FOR CELL TRANSPLANTATION

BACKGROUND

[0001] Cell replacement therapy is a promising potential treatment option for a wide variety of diseases. Many clinical conditions and disease states result from the lack of factor(s) produced by living ceils or tissues, including, for example, diabetes, in which insulin production is inadequate; Parkinson's disease, in which dopamine production is decreased; and anemia, in which erythropoietin is deficient. Such conditions or diseases may be treated by cell/tissue implants that produce the missing or deficient factor(s).

[0002] However, many challenges remain in the field of ceil replacement therapy. The viability and functionality of transplanted cells is compromised by, for example, lack of mechanical protection, lack of necessary

factors/nutrients (e.g., due to inadequate vascularization or inability of the vascular system to reach parts of the transplant), and inflammatory responses. Thus, there is a need for methods and devices that optimize the viability and functionality of implanted cells.

[0003] Type 1 diabetes meilitus (T1 DM) is an autoimmune disorder characterized by the destruction of insulin producing beta ceils found in the islets of Langerhans of the pancreas. Clinical transplantation of insulin-

producing cells offers a solution to restoring beta cell function through the intrahepatic transplantation of, e.g., allogeneic pancreatic islets into a diabetic recipient (see Fig. 1A). Such transplantation can lead to improved control in blood glucose levels, higher C~peptide levels, and insulin independence for several years, delaying the onset and reducing the intensity of diabetes- related complications. However, the success of clinical insulin-producing cell transplantation is hindered by the high rate at which transplanted cells are destroyed and/or rendered nonfunctional. This is determined in part by the standard intrahepatic location of the implant site, which is prone to mechanical stresses and inflammatory responses, is exposed to high drug and toxin loads, and renders the transplanted cells irretrievable. Transplantation of insulin- producing cells at alternative sites, such as the subcutaneous space, may alleviate many of these issues. However, relocating insulin-producing cells to alternative sites requires adequate mechanical protection and spatial distribution of the transplanted cells, as well as access to a fully developed vasculature.

SUMMARY OF THE INVENTION

[0004] To address these issues, we have designed and developed a highly porous organosilicone (po!ydimethylsiloxane, PDMS) scaffold capable of providing structural support and adjustable spatial distribution to ceils without hindering nutrient delivery. The scaffold provides a superior, more

physiological environment for the cells, leading to improved viability and function. Furthermore, the scaffold material itself can be modified to provide sustained delivery of biologically active agents that improve the engraftment, survival, function and long-term viability of the cells. Such agents include, but are not limited to, oxygen generating, releasing or transport-enhancing agents, growth factors or growth-stimulating factors, anti-inflammatory compounds and immunosuppressive agents. Table 1 (below) indicates certain advantages of the scaffold and select reasons for their importance.

[0005] The implanted cells may be, e.g., insulin-producing ceils. Structural support to insulin-producing ceils in the form of a scaffold is critical to reducing pelleting and agglutination of the ceils, which results in decreased availability of nutrients to the cells and thus leads to cell death. The highly porous organosilicone scaffold maximizes nutrient delivery by creating a structure that supports and spatially distributes the insulin-producing cells, and also promotes vascular infiltration.

[0006] Highly porous, biocompatible and biostable scaffolds within alternative transplantation sites offer a rational strategy for improving overall cell engraftment (Fig. 1 B). These scaffolds provide mechanical protection to the cells, afford retrievabiiity, and for ceils with high metabolic demand (e.g., insulin-producing cells), allow for both intra-device vascularization and a means to spatially distribute the ceils within the device to avoid ceil death resulting from inadequate nutrient supply where the high density of metabolic demand cannot be satisfied due in large part to diffusion limitations. The scaffold surface and void spaces, or the scaffold material itself, may be modified with one or more different adhesion proteins and optionally other biological factors (e.g., anti-inflammatory factors) for enhanced cell adherence and viability.

BRIEF DESCRIPTION OF THE DRAWINGS

[0007] FIGURE 1 : Schematic of scaffold platform and clinical islet transplantation.

a) Diagram of procedure for islet cell transplantation.

b) Schematic of the multi-functional platform of the macroporous scaffold.

The scaffold material provides three-dimensionai distribution and protection from mechanical stress for cells of multiple sizes. The porosity of the scaffold material is sufficient to permit vascular infiltration. The scaffold material can be surface modified with bioactive factors to enhance adhesion or modulate the surrounding environment, and can serve as a means to locally deliver agents. Further, the scaffold material may serve as a platform for the delivery of adherent cells, which can grow and proliferate on the scaffold.

[0008] FIGURE 2: Analysis of scaffold pore structure,

a) Photographic image of macroporous scaffold structure.

b) Image of macroporous scaffold structure by scanning electron microscopy (JEOL, JSM-5600LV, 29Pa, 20kV).

c) image of macroporous scaffold structure by scanning electron microscopy at higher magnification (JEOL, JSM-5600LV, 29Pa, 20kV).

d) Distribution graph of scaffold porosity (n=20). The green square indicates average scaffold porosity at 85%.

e) Confocal images (left) and merged confocal and fluorescent images (right) of noncoated scaffold (top) and fibronectin coated scaffold (bottom) stained with anti-fibronectin-biotin primary antibody and streptavidin-FITC secondary antibody (Zeiss LSM510). [0009] FIGURE 3: Scaffolds are biostable for up to 30 days after implantation into rats.

a) Histological cross-sections of scaffolds at 10X magnification stained with Masson's trichrome stain at days 3, 14 and 30 after implantation into 1 cm subcutaneous pockets in male Lewis rats. Top row: control (absence of materials). Second row: noncoated PDMS macroporous scaffoldv s. Third row: fibronectin-coated PDMS macroporous scaffolds. Bottom row:

negative control (Dacron material). [0010] FIGURE 4: identification of blood vessel infiltration into scaffolds.

Histological cross-sections of noncoated and fibronectin coated scaffolds at 40X magnification stained with Masson's trichrome stain at day 14 and day 30 after implantation into 1 cm subcutaneous pockets in male Lewis rats. BV: blood vessels. SS: scaffolds.

[0011] FIGURE 5: Analysis of collagen deposition and total infiltration into scaffolds.

A) Quantification of collagen deposition into noncoated and fibronectin coated scaffolds via Metamorph Analysis of blue areas in samples stained with

Masson's trichrome stain.

B) Quantification of total cell and matrix infiltration into noncoated and

fibronectin coated scaffolds via Metamorph Analysis of total area not occupied by scaffold material in samples stained with Masson's trichrome stain.

[0012] FIGURE 6: Cellular adhesion and growth on scaffolds.

Visualization of mesenchymal stem cells loaded in noncoated (left) and fibronectin-coated (right) scaffolds and cultured in standard culture conditions (24 well plate, full culture media) for two days (top) and seven days (bottom). Samp!es were stained with fluorescent live/dead dye (calcein AM (green) and EthD-1 (red)).

[0013] FIGURE 7: Adherence of islets seeded in scaffolds.

A) Graph showing retention of rat islets within scaffold alone (black bar) and scaffold with fibrin (gray bar) as measured by a colorimetric MTT assay at day 0 after loading.

B) Graph showing retention of non-human primate islets within scaffold alone (black bar) and scaffold with fibrin (gray bar) as measured by MTT assay at day 0 after loading.

[0014] FIGURE 8: Viability of islets seeded in scaffolds.

Viability of 1500 IEQ within: a standard two-dimensional culture dish control (white bar); a PDMS macroporous scaffold without fibrin (black bar); or a fibronectin-coated PDMS macroporous scaffold with fibrin (gray bar); as measured by MTT assay. Fibrin was loaded onto the scaffold following islet loading, to fill in the scaffold void space. Islets were cultured for 0 or 24 hours for rat islets and 24 hours for non-human primate and human islets. A: rat islets; B: non-human primate islets; C: human islets.

[0015] FIGURE 9: Spatial distribution and viability of islets seeded in scaffolds.

Confocai microscope images of islets stained with fluorescent live/dead dye (calcein AM (green) and EthD-1 (red)) within PDMS macroporous scaffolds following 24 hours of culture. A: rat islets; B: non-human primate islets; C: human islets.

[0016] FIGURE 10; Insulin secretion of islets seeded in scaffolds, insulin ELISA assay showing functional insulin secretion rates after !ow (40 mg/dL) and high (300 mg/dL) glucose stimuli for 150 IEQ in: a standard two- dimensional culture dish control (white bar); a PDMS macroporous scaffold without fibrin (black bar); or a PDMS macroporous scaffold with fibrin (gray bar). Results are expressed as stimulation index, which is the total insulin output at high glucose divided by the total insulin output at low glucose.

Human and rat islets were cultured for 24 hours, while non-human primate islets were evaluated six hours after loading. A: rat islets; B: non-human primate islets; C: human islets.

[0017] FIGURE 11 : Average diameter of islets retained in scaffolds. A) Table showing average diameter of non-human primate islets seeded in scaffolds.

B) Table showing average diameter of human islets seeded in scaffolds.

[0018] FIGURE 12; Comparison of islet viability in macroporous and microporous structures.

Graph showing islet viability of 1000 IEQ cultured for 48 hours at 5% oxygen in a PDMS macroporous scaffold (black bar, left) or a microporous 2% agarose gel (gray bar, right), as measured by MTT assay. [0019] FIGURE 13: in vivo performance of scaffolds for euglycemia restoration in a diabetic mouse model,

A) Staining of scaffold for islets (red) and nuclei (DAPI) 74 days after

implantation into the epididymal fat pad of a diabetic mouse model.

B) Postoperative blood glucose levels of diabetic mice implanted with islets alone (light blue diamonds), islets seeded in a fibronectin-coated scaffold

(dark blue line, black diamonds) or islets seeded in a fibronectin-coated scaffold then coated with fibrin/PDGF (white diamonds).

C) Diabetic state graphs for mice with free islets (light blue solid line), islets seeded in a fibronectin-coated scaffold (dark blue beaded line) and islets seeded in a fibronectin-coaied scaffold then coated with fibrln/PDGF (dotted line),

[0020] FIGURE 14: In vivo performance of scaffolds in a diabetic rat modeL,

A) Photographs of a PDMS macroporous scaffold loaded with islets in a

spread omentum (left) and wrapped up in the omentum (right) of a Lewis rat,

B) Graph showing nonfasting blood glucose levels of chemically induced diabetic Lewis rats following the transplantation of 1500 IEQ syngeneic islets in a PDMS macroporous scaffold into the omental pouch site

(squares) or as free islets into the standard kidney capsule transplant site (triangles). The graft was removed at 188 days.

C) Graph showing results of intravenous glucose tolerance test performed on functional graft recipients over 100 days post transplant. Black diamond, solid line: islets in silicone scaffold in omentum. Open circle, dashed line: islets freely loaded in omentum. Gray triangle, solid line: islets freely loaded into the kidney capsule.

D) Photographs showing 10X (A) and 20X (B) magnifications of

immunofluorescence-stained islets within PDMS macroporous scaffolds following expiantation of the grafts (blue: nuclei stained with DAPI; green: islets).

E) Graph showing nonfasting blood glucose levels of chemically induced diabetic Lewis rats following the transplantation of 3000 or 5000 IEQ allogeneic islets in a PDMS macroporous scaffold with fibrin/PDGF into the omental pouch site (3000: triangles; 5000:squares) or as free islets into the standard kidney capsule transplant site (diamonds).

[0021] FIGURE 15: in vivo performance of scaffolds in a diabetic baboon model. Graph showing fasting blood glucose (FBG; yellow), post-prandial blood glucose (PBG; green), and exogenous insulin levels (insulin/kg; blue) for a baboon receiving 25,000 allogeneic lEQ/kg loaded within 6 PDMS

macroporous scaffolds and implanted into the omental pouch. The baboon received intravenous anti~CD154 mono-therapy (20 mg/kg) on post-operative days 0, 1 , 4, 10, 18 and 28 and every week thereafter as maintenance.

[0022] FIGURE 16: Eiution profiles for dexamethasone incorporated in scaffolds.

Eiution profiles for 10% or 20% w/v dexamethasone incorporated within a

PDMS macroporous scaffold when incubated in a buffer solution. Scaffolds of 8 mm and 10 mm in diameter were tested.

[0023] FIGURE 17; Design of agent-releasing elements,

A) Examples of potential drug-releasing geometries. Drugs may be

incorporated into a biocompatible material such as PDMS and shaped into a disk, rod, or outer cage for implantation with the macroporous scaffold. B) Photograph of a drug-releasing "cage" design, where the drug-releasing material is fabricated into an outer cage that can house the macroporous scaffold.

[0024] FIGURE 18: Eiution profiles for the release of dexamethasone incorporated Into agent-releasing elements.

A) Eiution profiles for the release of dexamethasone incorporated at 0%, 5%, 10% or 20% w/v into PDMS disks.

B) Eiution profile for the release of dexamethasone incorporated at 10% w/v into a PDMS cage.

[0025] FIGURE 19: In vivo performance of agent-releasing elements in a diabetic mouse model. a) Urine measurements of dexamethasone level in mice with a subcutaneous implant of a dexamethasone releasing PDMS disk (with dexamethasone incorporated at 0%, 5%, 10%, or 20% w/v) as a function of days after implant.

b) Graph illustrating the time to reversal to normog!ycemia for chemically

induced diabetic mice with syngeneic islet transplants within the scaffold at the epididymal fat pad site, with: no dexamethasone releasing rod surrounding the scaffold (squares), one dexamethasone releasing rod surrounding the scaffold (triangles) or two dexamethasone rods

surrounding the scaffold (upside-down triangles).

DETAILED DESCRIPTION OF THE INVENTION

[0026] The invention relates to biocompatible scaffolds with a controllable pore size range, in some embodiments, the pore size range is 25-850 μηι. In certain embodiments, the pore size range is 250-425 μηι, which is conducive to housing, e.g., typical 150 ,um diameter pancreatic islets. In other

embodiments, some or all of the pore sizes are smaller (e.g., 225 μνη, 200 μηι, 175 μίϊΐ, 150 μηι, 100 μηι, or less), for housing smaller cells (e.g., smaller islets, or individual ceils or aggregates of individual cells such as beta cells and other therapeutic agent-releasing cells). The pore sizes may in certain embodiments be larger, e.g., for housing larger cells or tissue samples. The pores may also be fabricated with a mixture of pore sizes. In certain embodiments, the scaffold has pore sizes ranging from 50-500 μηη, 80-450 μηι or 75-400 um. Pore sizes may be randomly distributed or may be fabricated to form patterns. For example, the pores may be positionaiiy asymmetric, in some embodiments, the pores may be fabricated with larger pores in one part of the scaffold and smaller pores in another part of the scaffold. In certain embodiments, the pores may be fabricated with larger pores on one side of the scaffold (e.g., the top) and smaller pores on another side (e.g., the bottom), creating a gradient of pore sizes. In some embodiments, the pores are fabricated to provide substantially uniform vascularization throughout the scaffold. The scaffolds may be fully or partially macroporous. The scaffolds have high biocompatibi!ity after extended implantation and are suitable for in vivo applications. In some embodiments, the viability and functionality of ceils (e.g., insulin-producing cells) are not adversely affected by loading within the scaffolds. In some embodiments, ceils (e.g., insulin-producing cells) and/or cell aggregates distribute evenly throughout the scaffold.

[0027] Macroporous scaffolds may be fabricated from any suitable biocompatible material, using any suitable technique that delivers high porosity, in certain embodiments, the scaffolds are fabricated using high internal phase emulsion (HIPE) polymerization, fiber bonding, gas foaming, critical freeze drying and/or electrospinning.

[0028] in certain embodiments, the scaffolds are fabricated using the solvent casting and particulate leaching technique (SCPL). The scaffolds may be fabricated from, e.g., silicone. In some embodiments, the scaffolds are fabricated from organosilicone. Silicone molds can be created by combining varying ratios of particulates and silicone polymer. In some embodiments, the ratio is 50-90% v/v. in some embodiments, the ratio is 50%, 55%, 80%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91 %, 92%, 93%, 94%, 95%, 98%, 97%, 98%, or 99% v/v particulates to silicone polymer. In one embodiment, the ratio of particulates to silicone polymer is 90% v/v. In some embodiments, the particulates are salt crystals, gelatin spheres, paraffin spheres, sodium tartrate, sodium citrate and/or glucose grains. In certain embodiments, the particulates are sodium chloride. In certain embodiments, the silicone polymer is the organosilicone polymer polydimethylsiloxane

(PDMS). Pore size and degree of porosity can be individually controlled by varying the particle size and polymer to particle ratio, respectively, using methods known in the art. The molds in which the particulate/silicone mixture is loaded may vary in dimension depending on the type of cells to be transplanted and the site of implantation, in some embodiments, the scaffold has a porosity of 20-99%, 30-99%, 40-98%, 50-97%, 60-98%, 70-95%, 75- 90%, or 80-90%. In some embodiments, the porosity of the scaffold is greater than 20%, 30%, 40%, 50%, 60%, 75%, 80%, 81 %, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99%.

[0029] The scaffolds may be of any desired shape. The shape of the scaffolds may vary based on the types of ceils to be implanted, the intended therapeutic effect, and/or the location of the implant, for example. The skilled practitioner can assess the shape(s) preferred for the intended appiication(s).

[0030] For enhanced cell adhesion, the scaffold surface or void space may be modified or filled with one or more different adhesion proteins or matrices, in certain embodiments, extracellular matrix (ECM) materials or adhesion proteins, e,g, collagen type I or IV, !aminin, fibronectin, albumin, fibrin, or arginine-glycine-aspartate peptides, or non-natural materials that promote cellular adhesion, e.g. poly-i-lysine, albumin, alginate, or agarose, are incorporated in or on the scaffold (see e.g., Beck et aL, Tissue Eng 13(3): 1 -1 1 (2007)). The void space may be filled (partially or completely,

prior/during/after ceil loading) with a matrix, e.g., collagen, fibrin,

poiy(ethylene) glycol, etc. The matrix may comprise agents that aid in the adherence, long term survival and/or function of cells implanted in the scaffold, such as growth factors, anti-inflammatory factors, and/or vascular- promoting/adhesion-promoting agents, e.g., PDGF, fibronectin, etc. The agents may be directly tethered to the matrix or simply mixed within the materials. In some embodiments, the agents are incorporated within the scaffold materials, e.g., by mixing the agents with the material before polymerization is initiated, in certain embodiments, fibrin glue is loaded into the void spaces of the scaffold to aid cell retention. In certain embodiments, PDGF is loaded into the void spaces of the scaffold, in certain embodiments, the scaffold surface is coated with fibronectin. in some embodiments, the scaffold surface is coated with fibronectin, and fibrin glue and PDGF are loaded into the void spaces, either individually or in various combinations. In other embodiments, fibrin glue is loaded into the scaffold without PDGF or fibronectin.

[0031] In some embodiments, the scaffold has the capacity to be stable at greater than 50%, 60%, 70%, 75%, 80%, 85%, 90%, or 95% void space, in certain embodiments, the scaffold has the capacity to be stable at greater than 80% void space.

[0032] in certain embodiments, the invention relates to biocompatible, highly porous organosilicone scaffolds with controllable pore size and porosity that can be fabricated and used for transplantation of insulin-producing ceils. The scaffolds may be used for transplanting the cells into alternative, non-hepatic sites by addressing the issues of spatial distribution of insulin-producing cells, mechanical protection, and intra-device vascularization.

[0033] Besides pancreatic islets, which are considered one preferred cell/tissue type for regulating sugar and energy metabolism, and for treating diabetes, the scaffolds of the invention and methods involving those scaffolds may also be used for other cell therapy model systems. Ceils for implantation may deliver a therapeutic benefit, e.g. by expressing a therapeutic factor in vivo. Examples of such celis include, but are not limited to, celis that produce: dopamine to treat Parkinson's disease (Minquez-Caste!lanos et a!., J Neurol Neurosurg Psychiatry in press (2007)); growth hormone to treat dwarfism (Chang et al., Trends Biotechnoi 17:78-83 (1999)); factor VIII and factor IX (Chang et ai., Trends Biotechnoi 17, 78-83 (1999)) to treat hemophilia; and erythropoietin to treat anemia (Rinsch et ai., Kidney intern 62:1395-1401 (2002)). Many more beneficial cell produced factors or cellular/tissue activities may be imagined. The implanted cells may express and/or deliver more than one therapeutic factor, or may comprise two or more cell types delivering one or more therapeutic factors. The implanted cells may also or alternatively express and/or deliver an agonist, analog, derivative, chimera, fusion, or fragment of a therapeutic factor to deliver a therapeutic effect. [0034] The implanted cells may also or alternatively deliver a therapeutic effect without secreting a diffusible factor, e.g., by providing an enzymatic activity that, for example, converts a substrate into a product having a beneficial effect, and/or metabolizing, sequestering, or absorbing a detrimental substance. The implanted ceils may deliver a therapeutic effect through a biological material-linked factor, such as a cell surface-linked factor.

[0035] The cells may naturally deliver a therapeutic effect, without genetic modifications, or may be genetically engineered to do so. For example, the cells may be transfected with expression vectors that express one or more therapeutic and/or helper ceil factors, in another embodiment, the cells may comprise, consist of, or consist essentially of cells transfected with expression vectors that express one or more therapeutic and/or helper ceil factors. Such expression may be in a constitutive or in a regulated manner, e.g., in response to biological modulators in the bloodstream or tissues to which the scaffold is exposed. These and other expression systems and methods for making them are well known to the skilled practitioner.

[0036] The cells may be, for example, autologous, heterologous, syngeneic, allogeneic, or xenogeneic cells. The cells may be derived from cadaver tissue or from living tissue. The cells may be of non-mammalian or mammalian origin, non-human origin or human origin, self or non-self. The ceils may be pluripotent, muitipotent, totipotent, or differentiated embryonic or adult stem cells; primary differentiated cells; or immortalized cells, among other cell types. Stem ceils may comprise, e.g., cells derived from cord blood, amniotic fluid, menstrual blood, placenta, Wharton's jelly, cytotropobiasts, and the like. The cells may also comprise any combination of the above-listed cell types.

[0037] The cells that provide a therapeutic effect may be implanted alone or in combination with other ceil types (e.g., Sertoli ceils, mesenchymal and bone marrow derived cells, endothelial progenitor cells, stem cells, regulatory T cells T _reg , etc., each referred to generically as implant "helper cells") that provide growth factors and/or other beneficial agents for establishment, maintenance or expansion of the implanted cells, or otherwise to help the implanted cells deliver a therapeutic effect. In one embodiment, the ceils that provide a therapeutic effect are implanted with mesenchymal stem ceils (MSCs).

[0038] The scaffolds and methods of this invention may be used to treat disorders including, but not limited to: diabetes, Parkinson's disease, anemia, dwarfism, hemophilia, amyloidosis, immune system disorders, inflammations, chronic pain, arthritis, hypertension, disorders of the nervous system, metabolic disorders, endocrine disorders, iymphoproiiferative disorders, myeloproliferative disorders, myeiodyspiastic syndromes, stem cell disorders, phagocyte disorders, histiocytic disorders, abnormalities of erythrocytes or platelets, plasma ceil disorders, acute ieukemias, chronic ieukemias, malignancies (breast carcinoma, Ewing Sarcoma, neuroblastoma, renal ceil carcinoma, etc.), hypothyroidism, hypopituitarism, hypogonadism, graft failure, graft versus host disease (GVD), veno-occlusive disease, side effects from pre-transpiant chemotherapy (such as excessive bleeding, infertility, and renal as well as lung and heart complications), and other disorders and diseases that will be recognized by the skilled practitioner.

[0039] Exemplary therapeutic factors which may be delivered by the transplanted cells include, but are not limited to, one or more of: insulin, glucagon, erythropoietin; Factor VIII; Factor IX; hemoglobin; albumin;

neurotransmitters such as dopamine, gamma-aminobutyric acid (GABA), glutamic acid, serotonin, norepinephrine, epinephrine, and acetylcholine;

growth factors such as nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3), neurotrophin 4/5 (NT-4/5), ciliary neurotrophic factor (CNTF), glial cell line-derived neurotrophic factor (GDNF), cholinergic differentiation factor/leukemia inhibitory factor (CDF/LI F), epidermal growth factor (EGF), insulin-like growth factor (IGF), fibroblast growth factor (FGF), and platelet-derived growth factor (PDGF); pain inhibitors such as Substance P, catecholamines, dynorphins, endorphins, or

enkephalins; hormones such as parathyroid hormone or growth hormone; immunomodulators such as granulocyte-macrophage colony stimulating factor (GM-CSF); neuromodulators; !ymphokines; cytokines; cofactors; antibodies; aptamers; and enzymes. Choice of one or more therapeutic factors and the concentrations at which they are produced and released from the cells and thereby from the scaffolds to the subject are dictated by the needs of the subject (e.g., patient) being treated, the selected location of implantation and other factors that may be readily determined empirically by the skilled practitioner.

[0040] in some embodiments, the therapeutic factor has insulin-like or insulin-regulatory activity, in certain embodiments, the therapeutic factor is insulin or an insulin analog. In certain embodiments, the therapeutic factor is a precursor form of insulin, such as preproinsu!in or proinsuiin. In certain embodiments, the therapeutic factor is an insulin chimeric or fusion protein.

[0041] in some embodiments, the therapeutic factor(s) are released from the implanted cells due to the receipt of a stimulus or signal from the host (e.g., changes in blood levels of glucose, hormones, metabolic signaling agents, chemical signaling molecules, etc.).

[0042] in some embodiments, the therapeutic effect comprises regulation of insulin levels in the blood. In certain embodiments, the therapeutic effect comprises regulation of glucose levels in the blood. In other embodiments, the therapeutic effect comprises regulation of levels of one or more other biological response regulators in the blood of the subject.

[0043] in some embodiments, the scaffolds also contain agents that aid long term survival and function of the transplanted cells. Such agents include, e.g., agents for vascularization (e.g., VEGF), anti-inflammatory agents (e.g., anti- TNF alpha, lisofylline, pentoxifylline, alpha 1 -antitrypsin, inter!eukin-1 (!L-1 ). interleukin-10 (!L-10), interleukin-1 receptor antagonist peptide (IRAP), TGF- beta, antibodies to !L-1 , interferon gamma, TNF-alpha, anti-tissue factor, complement inhibitors, COX-2 inhibitors, calcineurin inhibitors (e.g., cyc!osporine, tacrolimus, etc.), glucocorticoids (e.g., dexamethasone, Cortisol, prednisolone, loteprednol eiabonate, flucinolone acetonide, etc.), and inhibitors of lymphocyte trafficking (e.g. fingo!imod. etc.)); oxygen generating, releasing or transport-enhancing agents (e.g., encapsulated peroxides or perfluorocarbons (PFCs)); cytoprotective/antiapoptotic agents/molecules, tolerance-inducing molecules (e.g., the Power-Mix described in Zheng et aL Immunity 19(4):503-514 (2003), or wherein said Power-Mix comprises (1 ) an agonist to IL-2, immunoglobulin, and/or a fusion protein; (2) antagonist-type !L- 15-reiated cytolytic immunoglobulin and/or a fusion protein; and (3) plus or minus rapamycin); IL-10 and IL-10 fusions; costimulatory blocking agents including antibodies, fusion proteins, small molecules, galectin-1 , aptamers, antibodies and aptamers to lymphocyte activation markers (e.g., 4BB1 );

adhesion molecules (e.g., CD103, etc.) and other molecules involved in the delivery of signals to lymphocytes (e.g., LFA1 , LFA3, 4BB1 , and CD45, etc.); EBNA-like molecules; IL-35-, IL12-, and IL12-receptor-targeting antibodies and aptamers; anti-IL-17 antibodies; anti-iL-17 receptor antibodies and aptamers; and anti-!L-6 antibodies and lL-6 receptor antibodies and aptamers; etc.); immunosuppressive agents (e.g., oATP, calcineurin inhibitors (e.g.,

cyclosporine, tacrolimus, etc.), protein kinase C inhibitors (e.g., AEB071 , etc.), inhibitors of proliferation signals (e.g., siroiimus, everolimus, JAK3 inhibitors, etc.), inhibitors of nucleotide synthesis (e.g., azathioprine, mycophenolic acid MPA / mycophenoiate mofetii MMF, lef!unomide, FK778, etc.), glucocorticoids (e.g., dexamethasone, Cortisol, prednisolone, loteprednol etabonate, flucinolone acetonide, etc.), inhibitors of lymphocyte trafficking (such as fingoiimod and other sphingosine-1 -phosphate receptor 1 modulators, etc); inhibitors of cell surface receptor activation (such as depleting or nondepleting antibodies and fusion proteins including but not limited to Thymoglobulin ATG, muromonab-CD3, a!emtuzumab. rituximab, daciizumab, basiiiximab, beiatacept, campath-1 H, Prograf, anti IL~2r, MMF, FTY, LEA, and others, etc.); oxygen generating, releasing or transport-enhancing products; and growth factors (e.g., IGF-I, IGF-l i, INGAP, exendin-4, GLP-1 , HGF, etc.). in certain embodiments, some or ail of the agents are released with slow/sustained release characteristics, e.g., in slow/sustained release cartridges, coatings, encapsulations, micro- or nanospheres, etc. In certain embodiments, some or all of the agents exhibit slow/sustained release for over at least 3, 10, 12, 20, 30, 40, 50, 80, 70, 80, 90, 100, 200, 300, 385, 500 or 730 days, or any other period adequate for the desired purpose. In one embodiment, the scaffold comprises slow-release dexamethasone. in one embodiment, the scaffold comprises slow-release fingolimod (FTY720).

[0044] in some embodiments, an agent is released from the scaffold. The agent may be incorporated into, e.g., a 3-D formulation, in some

embodiments, some or all of the agent is released with slow/sustained release characteristics. In certain embodiments, the agent is an anti-inflammatory or immunosuppressive agent, e.g., anti-TNF alpha, lisofyiline, pentoxifylline, alpha 1 -antitrypsin, interleukin-1 (IL-1 ), interleukin-10 (IL-10), snterleuksn-1 receptor antagonist peptide (!RAP), TGF-beta, antibodies to IL-1 , interferon gamma, TNF-a!pha, anti-tissue factor, complement inhibitors, COX-2 inhibitors, calcineurin inhibitors (e.g., cyciosporine, tacrolimus, etc.), glucocorticoids (e.g., dexamethasone, Cortisol, prednisolone, ioteprednoi etabonate, flucinoione acetonide, etc.), inhibitor of lymphocyte trafficking (e.g. fingolimod, etc.), protein kinase C inhibitors, inhibitors of proliferation signals (e.g., sirolimus, everoiimus, JAK3 inhibitors, etc), inhibitors of nucleotide synthesis (e.g., azathioprine, mycophenolic acid MPA / mycophenolate mofetii MMF, lef!unomide, etc.), and inhibitors of ceil surface receptor activation (e.g., depleting or nondepieting antibodies and fusion proteins including but not limited to Thymoglobulin ATG, muromonab-CD3, a!emtuzumab. rituximab, daciizumab, basiliximab, belatacept, campath-1 H, etc.). In certain

embodiments, the agent is a hydrophobic agent, in some embodiments, the agent comprises dexamethasone, prednisolone, cyciosporine, tacrolimus, sirolimus, everoiimus, fingolimod (FTY720), mycophenolic acid, etc.

Organosiiicone materials such as PDMS are particularly suitable for the preparation of sustained release formulations of hydrophobic drugs (Malcolm, K. et al. J, Contr. Rel. 2003, 90, 217). However, any material that may be used to prepare sustained release formulations of hydrophobic drugs is suitable for and encompassed by the present invention,

[0045] in some embodiments, the scaffold acts as a membrane enclosing an interior reservoir system with the agent(s) to be released. In some

embodiments, the agent is incorporated into the scaffold structure itself, in certain embodiments, the scaffold forms a non-erodible matrix system with the agent, wherein the agent is dispersed in the polymer. A number of siloxane- based implants for sustained release progestin delivery (8 months to 7 years) are commercially available (Croxatto, H. B. Contraception 2002, 85, 15) proving the feasibility and safety of this approach, in some embodiments, the material of the scaffold also serves as the polymeric matrix material for the sustained release of the therapeutic agent.

[0046] Alternatively, or additionally, the agent may be incorporated into an element that is not connected to the part of the scaffold containing the implanted cells, which is then placed within, surrounding or adjacent to the scaffold structure upon implant. In some embodiments, the separate element is nonporous. The separate element may have any shape capable of effectively releasing an agent for the purposes of the invention. In some embodiments, the separate element is shaped into a disk, rod or outer cage. In certain embodiments, the separate element is an outer cage thai surrounds ihe scaffold. In certain embodiments, the separate element is designed for periodic replacement, and provides sustained release of the incorporated agent for arbitrary prolonged periods as desirable. The incorporated agent may be released for periods of, e.g., hours, days, weeks, or years before replacement. The separate element may be made of, e.g., PDMS and/or any other biocompatible materials capable of being fabricated with and releasing agents. In some embodiments, the separate element is fabricated from the same material as the scaffold. In some embodiments, the separate element releases the same agent as the scaffold.

[0047] in one embodiment, anti-inflammatory and/or immunosuppressive molecules are tethered to or incorporated into the scaffold to reduce the host inflammatory response to the implant. Exemplary anti-inflammatory/ immunosuppressive agents include, e.g., anti-TNF alpha, lisofyiline, pentoxifylline, alpha 1 -antitrypsin, interieukin-1 (IL-1 ), interieukin-10 (IL-10), interieukin-1 receptor antagonist peptide (IRAP), TGF-beta, antibodies to IL-1 , interferon gamma, TNF-alpha, anti-tissue factor, complement inhibitors, COX- 2 inhibitors, caicineurin inhibitors (e.g., cyciosporine, tacrolimus, etc.), glucocorticoids (e.g., dexamethasone, Cortisol, prednisolone, loteprednol etabonate, f!ucino!one acetonide, etc.), inhibitor of lymphocyte trafficking (e.g. fingoiimod, etc.), protein kinase C inhibitors, inhibitors of proliferation signals (e.g., sirolimus, everolimus, JAK3 inhibitors, etc.), inhibitors of nucleotide synthesis (e.g., azathioprine, mycophenoiic acid MPA / mycophenolate mofetii MMF, lefiunomide, etc.), and inhibitors of ceil surface receptor activation (e.g., depleting or nondepieting antibodies and fusion proteins including but not limited to Thymogiobuiin ATG, muromonab-CD3, alerntuzumab, rituximab, daclizumab, basiiiximab, beiatacept, campath-I H, etc.) In one embodiment, extracellular matrix (ECM) molecules such as collagen type I or IV, iaminin, fibronectin, fibrin, or arginine-glycine-aspartate peptides are incorporated on the surface of the scaffold (Beck et a!., Tissue Eng 13(3): 1 -1 1 (2007)). in addition to surface modification, the void space may also be filled (partially or completely, prior/during/after ceil loading) with a matrix, e.g., collagen, fibrin, poly(ethylene) glycol, alginate, etc. The matrix may comprise growth factors, surface adhesion proteins, or other vascu!ar-promoting/adhesion-promoting agents, e.g., PDGF, fibronectin, etc. The agents may be directly tethered to the matrix or simply mixed within the materials. In certain embodiments, the agents may be coated or encapsulated for slow release properties. In one embodiment, the scaffold incorporates fibronectin on the scaffold surface, and fibrin and/or PDGF within the void space, individually or in various

combinations. In another embodiment, a !aminin-coated scaffold is filled with a collagen I matrix, individually or in various combinations, in one

embodiment, the scaffold material incorporates dexamethasone and/or fingo!imod (FTY720).

[0048] The scaffolds of the invention may be implanted in any appropriate place within the subject, in certain embodiments, the implant location may be, for example, intraomentai (in an omental pouch), subcutaneous,

intraperitoneal, intramuscular, or renal subcapsular. In one embodiment, the implant location is subcutaneous.

[0049] The scaffolds of the invention may be implanted into any animal host or subject, in some embodiments, the host or subject is a mammal. In a certain embodiment, the host or subject is human.

[0050] in any of the above embodiments, the scaffold may further comprise a tether to facilitate manipulation and/or retrieval of the scaffold from a subject.

EXAMPLES

Example 1 : Fabrication of Macroporous Siiicone Scaffolds

[0051] Macroporous silicone scaffolds were fabricated using the solvent casting and particulate leaching technique (SCPL). The silicone polymer was prepared by mixing PDMS monomer with platinum catalyst, 4:1 v/v. The silicone molds were created by combining varying ratios (50-90% v/v) of sodium chloride crystals (Mallinckrodt Baker, NJ) (250 to 425 μηι diameter) and silicone polymer solution. The sa!t/silicone mixture was loaded into prefabricated, stainless steel molds (10 mm diameter, 2 mm height), pressurized to 1500 psi and incubated at 37°C for 48 hrs to complete siiicone cross-linking. The NaCi was then leached out from the scaffolds for at least 72 hrs. Pore size and degree of porosity were individually controlled by varying the particle size and polymer to particle ratio, respectively. For enhanced cell adhesion, the scaffold surface was modified by incubating overnight with fibronectin at 250 μglmL·.

Example 2: Characterization of rnacroporous silicone scaffolds

[0052] The rnacroporous structure of the scaffold was visualized

photographically (Fig. 2A) and by scanning electron microscopy (SEM) (Fig. 2B, C). As seen in the SEM images, the scaffold is highly porous and the pore size is representative of the salt crystal diameter. Moreover, the pores are interconnected and tortuous. Final porosity was determined using gross measurements and weights (dry and wet), and calculated with the following formula:

Porosity of the scaffolds manufactured with 90% w/v sodium choioride crystals was determined to be 85% + 5% w/v (Fig. 2D).

[0053] The protein-modified scaffold surface was stained with anti- fibronectin-biotin primary and streptavidin-FITC secondary antibodies, and visualized through confoca! imaging. The fluorescence imaging demonstrated a homogenous!y modified scaffold surface with protein coating (Fig. 2E).

[0054] We performed in vivo studies to assess biocompatibiiity and stability of the silicone scaffolds, as well as vascular infiltration. In vivo biocompatibiiity was determined for noncoated and protein coated scaffolds by implanting the scaffolds into 1 cm subcutaneous pockets in male Lewis rats and performing histological analyses (HIE) at 3, 14, and 30 days (Fig. 3). Dacron material and absence of a material were also examined as controls.

[0055] Histological cross-sections showed the biocompatibiiity of the silicone scaffolds (with or without coating), with blood vessel formation around the scaffold (Fig. 4), significant deposition of collagen and no fibrotic tissue observed (Figs. 4 and 5A). The scaffolds were biocompatible and biostable for up to 180 days. Scaffolds coated with fibronectin encouraged a higher degree of collagen deposition (Fig. 5A) and a higher degree of cellular infiltration and ECM deposition compared to noncoated scaffolds, thereby demonstrating an enhanced integration of the biomateriai within the host (Fig. 5B).

[0056] The fibronectin-coated macroporous silicone scaffolds were shown to support greater cellular adhesion and growth of mesenchymal stem ceils (MSCs) than noncoated macroporous silicone scaffolds. MSCs cultured on the coated scaffolds for seven days exhibited their standard expanded phenotype (Fig. 8).

Example 3: Islet viability and function in macroporous silicone scaffolds

[0057] Pancreatic islets from male Lewis rats, non-human primate (NHP) baboons, and human sources were used in experiments to measure islet viability and function in macroporous silicone scaffolds, islets were loaded into the scaffolds at the desired islet equivalent (IEQ) density by suspending them in a small volume, pipetting them onto the scaffolds and applying a light pressure gradient to distribute the islets into micro-sized pores. Fibrin glue was added to select groups to evaluate islet retention. The islets were cultured in tissue culture dishes at 20% oxygen for up to 24 hours. Two- dimensional cultures were used as controls.

[0058] Scaffolds seeded with rat, non-human primate and human islets were inspected for islet spatial distribution and viability by fluorescent live/dead dye staining (calcein AM and EthD-1 ) and confocai microscopy (Fig. 9). The adherence of the rat and non-human primate islets, and the viability of the rat, non-human primate and human islets, were quantified by MTT assay

(Promega, WS), calibrated to cell number via standard curve, for 1500 IEQ per scaffold (Figs. 7 and 8). Functional insulin secretion rates were determined by collecting insulin samples from low (40 mg/dL) and high (300 mg/dL) glucose stimuli for 150 iEQ per scaffold and quantifying by insulin ELISA assay (Fig. 10). The average diameters of the non-human primate islets and the human islets are indicated (Fig. 1 1 ), demonstrating that islet retention is dependent on islet size (>35% non-human primate (26% islets > 100 iim); 80% human (71 % islets > 100 μιη)).

[0059] Rat islets within the scaffold (with or without fibrin) performed similarly to controls (p>0.05) for all groups and all assays for the same given day. Non-human primate islets within the scaffold (with or without fibrin) demonstrated similar viability and insulin secretion rate to controls (p>0.05) for all groups. Human islets within the scaffold (without fibrin) performed similarly to controls (p>0.05) for ail groups and ail assays. Islets within scaffolds with fibrin showed a statistically significant increase in adherence in comparison to islets within scaffolds without fibrin.

[0060] To compare islet viability in macroporous and microporous scaffolds, 1000 IEQ islets were loaded into macroporous PDMS scaffolds with fibrin, using a gentle pressure gradient; and into microporous scaffolds by uniformly mixing the islets with 2% liquid agarose, pouring into molds of the same dimensions as the PDMS scaffolds, and allowing to cool to room temperature. The islets were then cultured for 48 hours at 5% oxygen. As determined by MTT assay, viability was enhanced in macroporous scaffolds in comparison with microporous scaffolds (Fig. 12).

[0061] These studies indicate that macroporous silicone scaffolds exhibit improved retention of small islets when coated with fibronectin. Further, the macroporous silicone scaffolds retain similar islet viability compared to controls (p>0.05), as well as high functional indices, with no inhibition of islet function as compared to controls.

Example 4: In vivo performance of macroporous silicone scaffolds

[0062] To assess the in vivo efficacy of the scaffolds in restoring euglycemia in a syngeneic streptozotocin (STZ)-diabetic mouse model, fibronectin-coated scaffolds with or without fibrin and PDGF were loaded with 500 !EG mouse C57BL/6J islets and implanted into the epididymal fat pad (EFP) of STZ~ diabetic mice. Scaffolds were folded into the center of the EFP and secured. The control group received free islets into an EFP pocket. On day 74, the scaffold implant with fibrin and PDGF was stained for islets (red) and nuclei (DAPi) (Fig. 13A). The scaffolds with fibrin and PDGF restored

normogiycemia (defined as consistent nonfasting glycemia levels < 200 mg/dl) in the mice (Fig. 13B). Further, diabetic state graphs illustrate the accelerated restoration to normogiycemia for mice with grafts containing the macroporous scaffold with fibrin/PDGF (Fig. 13C).

[0063] To assess the in vivo efficacy of the scaffolds at restoring euglycemia in a chemically induced diabetic Lewis rat model, the rats were implanted with a scaffold loaded with 1500 IEQ in the omental pouch site (Fig. 14A) or 1500 !EG free islets in the standard rodent kidney capsule transplant site. An intravenous glucose tolerance test was performed on functional graft recipients over 100 days post-transplant by measuring blood glucose at timepoints following the injection of a bolus of glucose. Islets loaded in a silicone scaffold and implanted in the omentum, free islets implanted in the omentum, and free islets implanted in the kidney capsule all illustrated similar glucose clearance profiles (Fig. 14C). After removal of the implanted ceils at 188 days, hyperglycemia was restored, illustrating that the implants alone were responsible for the control of blood glucose levels (Fig. 14B). The scaffold was stained for islets (red) and nuclei (DAPI) after graft explanation, demonstrating that the islets were stil! viable and functional at the time of explantation (Fig. 14D).

[0064] The in vivo efficacy of the scaffolds in restoring euglycemia in an allogeneic sireptozoiocin (STZ)-diabetic mouse model was also assessed. Fibronectin-coated PDMS macroporous scaffolds with or without fibrin and PDGF, loaded with 3000 or 5000 IEG allogeneic islets, were implanted into the omental pouch site of chemically induced diabetic Lewis rats. A control group received 3000 1EG free islets into the standard kidney capsule transplant site. The scaffolds with fibrin and PDGF, seeded with allogeneic islets, were able to restore normoglycemia in the rats over a time course of at least 28 days (Fig. 14E).

[0065] To assess the in vivo efficacy of the scaffolds at restoring euglycemia in a diabetic baboon model, fasting blood glucose, post-prandial blood glucose and exogenous insulin levels were charted in a baboon (made diabetic by partial pancreatectomy and subsequent STZ administration) with 25,000 allogeneic !EQ/kg loaded within 8 PDMS macroporous scaffolds and implanted into the omental pouch. The baboon also received intravenous anti- rejection anti-CD154 mono-therapy at 20 mg/kg on post-operative days 0, 1 , 4, 10, 18, and 28 and every week thereafter as maintenance. At 384 days post- implant, the baboon continued to exhibit blood glucose control (Fig. 15). Example 5: Release of dexamethasorte from macroporous silicone scaffolds or separate elements.

[0066] To test the release of agents from the scaffold, 10% or 20% w/v of dexamethasone was incorporated within two different sizes of PDMS macroporous scaffolds (8 mm and 10 mm diameter) by mixing the

dexamethasone powder with the silicone polymer prior to loading into the molds. E!ution of the dexamethasone was measured upon incubation of the scaffold in a buffer solution. The incorporated dexamethasone showed sustained release from the scaffold material over a time course of at least 12 days (Fig. 16).

[0067] Agents may also be incorporated into and released from elements separate from the macroporous scaffolds. Fig. 17 illustrates examples of potential drug-releasing geometries. Dexamethasone incorporated into PDMS disks at 0%, 5%, 10% or 20% w/v or a PDMS cage at 10% w/v, where the dexamethasone was mixed with the polymer prior to pouring into the respeciive molds, demonslrated sustained release properties over the course of the study (Fig. 18).

[0068] To examine the performance of agent-releasing elements in vivo, chemically induced diabetic mice were implanted subcutaneously with PDMS disks in which dexamethasone was incorporated at 0%, 5%, 10% or 20% w/v. Dexamethasone in the urine of the mice was monitored over a time course and exhibited sustained release from the PDMS disks over the course of the study (Fig. 19A).

[0069] The effect of dexamethasone release on the performance of islets implanted in chemically induced diabetic mice was also examined. The mice were implanted with syngeneic islets within the macroporous scaffold at the epididymai fat pad site with zero, one or two 5% w/v dexamethasone releasing rods adjacent to the scaffold. No detrimental effect was observed in the time to reversal to normoglycemia in the presence of dexamethasone (Fig. 19B).

[0070] AH publications and patent applications cited in this specification are incorporated herein by reference as if each individual publication or patent application were specifically and individually indicated to be incorporated by reference.