Integration of Islet/Beta-Cell Transplants with Host Tissue Using Biomaterial Platforms

Abstract

Cell-based therapies are emerging for type I diabetes mellitus (T1D), an autoimmune disease characterized by the destruction of insulin-producing pancreatic β-cells, as a means to provide long-term restoration of glycemic control. Biomaterial scaffolds provide an opportunity to enhance the manufacturing and transplantation of islets or stem cell–derived β-cells. In contrast to encapsulation strategies that prevent host contact with the graft, recent approaches aim to integrate the transplant with the host to facilitate glucose sensing and insulin distribution, while also needing to modulate the immune response. Scaffolds can provide a supportive niche for cells either during the manufacturing process or following transplantation at extrahepatic sites. Scaffolds are being functionalized to deliver oxygen, angiogenic, anti-inflammatory, or trophic factors, and may facilitate cotransplantation of cells that can enhance engraftment or modulate immune responses. This local engineering of the transplant environment can complement systemic approaches for maximizing β-cell function or modulating immune responses leading to rejection. This review discusses the various scaffold platforms and design parameters that have been identified for the manufacture of human pluripotent stem cell–derived β-cells, and the transplantation of islets/β-cells to maintain normal blood glucose levels.

Type I diabetes (T1D), which affects an estimated 1.6 million Americans, is caused by the autoimmune destruction of the pancreatic β-cells within the islets of Langerhans. Individuals with T1D must currently rely on lifelong exogenous insulin therapy. Advanced insulin pumps and continuous glucose monitoring and feedback systems have given individuals with T1D tighter control of their blood glucose levels in recent years (1, 2). Despite the use of this technology and advanced algorithms, blood glucose levels are not maintained within the normal range as effectively as native islets, which puts individuals at risk for vascular complications and dangerous hypoglycemic events (1-3). The Edmonton Protocol, first published in 2000, has been utilized clinically to transplant islets via hepatic vein infusion in conjunction with immunosuppressive therapy (4). Individuals receiving islet transplantation have shown lower levels of glycated hemoglobin, more time spent in the healthy glycemic range, and a reduced risk for hypoglycemic events compared with recipients of continuous subcutaneous insulin infusion or multiple daily insulin injections (5). The Edmonton Protocol has shown the enormous promise for long-term, tightly controlled glucose levels without exogenous insulin; however, several drawbacks prevent wide adoption of this strategy (6). Hepatic vein infusion exposes cells directly to the blood without protection, resulting in an instant blood–mediated inflammatory reaction (IBMIR). The required systemic immunosuppression brings harsh side effects, putting patients at risk for infection, organ damage, and developing cancer. The harsh environment and need for lifelong immunosuppression provide a strong motivation to develop extrahepatic delivery strategies.

The limited supply of acceptable islets from cadaver donors requires alternative cell sources to be investigated and developed for T1D cell replacement therapy. Alternative cell sources that have shown promise include xenogeneic sources and human pluripotent stem cell (hPSC) sources, which include embryonic stem cells (hESCs) and adult induced pluripotent stem cells (iPSCs). While xenogeneic sources, such as porcine islets, have shown positive results in preclinical nonhuman primate studies, the xenogeneic immune response is considered an additional barrier to an already challenging immune response (7). Numerous studies have demonstrated the ability to generate pancreatic progenitors, cells committed to the pancreatic lineage, from pluripotent cells (8-11). These hPSC-derived pancreatic progenitors provide a means of overcoming the shortage of allogeneic donor tissue for this therapy (12).

Biomaterial scaffolds are being developed as a platform for delivering islets or hPSC-derived β-cells to extrahepatic sites. Extrahepatic sites are desired to avoid IBMIR, which introduces new challenges. Immediately following transplantation, transplanted cells face multiple challenges: they must rely on diffusion for oxygen, nutrients, and waste exchange during the period prior to revascularization, and they must overcome or avoid the inflammatory response and host autoimmune and allogeneic response (Fig. 1) (13-18). Biomaterials have been widely used for delivering islets, largely in an encapsulation approach that creates a physical barrier between therapeutic cells and the host immune system (19). Innovation with encapsulation technologies, such as novel materials to limit fibrosis (20), or thinner coatings (21), continues to reduce the transplant volume. Nevertheless, this approach restricts access to vascular ingrowth that is present within native islets and contributes to their function. Nonencapsulating scaffolds, in contrast to encapsulating hydrogels, allow for integration of transplanted cells with host tissue, which can support long-term survival and function (16, 22-24). These scaffolds have included microporous or fibrous scaffolds that allow for cell infiltration following cell transplantation, as well as degradable hydrogel systems that initially encapsulate cells, with subsequent degradation allowing for host tissue infiltration and integration. Scaffolds can provide support for cell adhesion and signals through functionalization of the material with peptides or proteins, or localized delivery of trophic factors. Collectively, scaffolds provide the opportunity to engineer the local microenvironment at the site of transplantation to support the integration, survival, and long-term function of transplanted cells.

Figure 1.

Survival and function of islets or β-cell clusters on microporous scaffolds posttransplantation. Following seeding, islets or hPSC-derived β-cells settle into the pores of the microporous scaffold. Three of the major challenges to survival and function include vascularization, inflammation, and adaptive immune responses. Scaffold modifications can be used to enhance vascularization or to attenuate the impact of inflammation or adaptive immune responses, with the arrows indicating that scaffold modifications can increase vascularization, transition a proinflammatory to an anti-inflammatory environment, or induce Tregs that can attenuate Teff responses at the graft. Vascularization: Release of growth factors such as VEGF or platelet-derived growth factor and/or co-transplant of supportive cells can enhance vascularization within the scaffold (13-15). Inflammation: Release of anti-inflammatory cytokines such as CXCL, IL-4, and IL-10 can be used to transition from a proinflammatory to an anti-inflammatory state (16, 17). Adaptive immune response: Modifying the scaffold with FasL108 or CCL22105 microparticles have been shown to induce regulatory T-cells upon release, as has systemic treatment with ECDI treated splenocytes (18). (Original diagram content developed by authors JL King and LD Shea with illustration by Jennifer Harley).

Herein, we provide an overview of scaffold platforms that are being employed for manufacturing hPSC-derived β-cell products and can deliver islets or β-cells to extrahepatic locations for integration with the host (Fig. 2A-D) (25-29). We review strategies towards enhancing the maturation and function of therapeutic β-cells, supporting survival of transplanted insulin-producing cells amid revascularization and inflammation, and the application of immune engineering methods to allow transplanted cells to avoid autoimmune or allogeneic host immune responses.

Figure 2.

Porous scaffolds used for Islet/β-cell transplantation. Biocompatible materials used to fabricate microporous scaffolds (A-D) (A) Poly (lactide-co-glycolide) (PLG) with 250-425 µm pores—scale bar 500 µm. Adapted from Gibly RF, Zhang X, Graham ML, et al. Extrahepatic islet transplantation with microporous polymer scaffolds in syngeneic mouse and allogeneic porcine models. Biomaterials. 2011;32(36):9677–9684. doi:10.1016/j.biomaterials.2011.08.084 with permission from Elsevier. (B) Polydimethylsiloxane (PDMS)—scale bar 500 µm. Adapted from Jiang K, Weaver JD, Li Y, Chen X, Liang J, Stabler CL. Local release of dexamethasone from macroporous scaffolds accelerates islet transplant engraftment by promotion of anti-inflammatory M2 macrophages. Biomaterials. 2017;114:71–81. doi:10.1016/j.biomaterials.2016.11.004 with permission from Elsevier. (C) Silk fibroin 200-400 µm pores—scale bar 50 µm. Adapted from Mao D, Zhu M, Zhang X, et al. A macroporous heparin-releasing silk fibroin scaffold improves islet transplantation outcome by promoting islet revascularization and survival. Acta Biomater. 2017;59:210–220. doi:10.1016/j.actbio.2017.06.039 with permission from Elsevier. (D) Polyethylene glycol (PEG) hydrogel—scale bar 1 mm. Adapted from Rios PD, Skoumal M, Liu J, et al. Evaluation of encapsulating and microporous nondegradable hydrogel scaffold designs on islet engraftment in rodent models of diabetes. Biotechnol Bioeng. 2018;115(9):2356–2364. doi:10.1002/bit.26741 with permission from John Wiley and Sons. (E) PLG scaffold used to culture pancreatic progenitors into insulin producing β-cells: top, extracellular matrix deposition within pores of the scaffold by hPSC derived β-cells; bottom, effect of pore size on insulin gene expression. Adapted from Youngblood RL, Sampson JP, Lebioda KR, Shea LD. Microporous scaffolds support assembly and differentiation of pancreatic progenitors into β-cell clusters. Acta Biomater. 2019;96:111–122. doi:10.1016/j.actbio.2019.06.032 with permission from Elsevier.

Manufacturing Platforms for Cell Transplantation

While hPSC-derived pancreatic progenitors have shown promise as allogeneic donor tissue, further development of manufacturing methods is required prior to successful clinical application. Large-scale production of hPSC-derived insulin-producing cells requires strict control of aggregate formation and differentiation factors for weeks of culture. The traditional culture system for cell-based therapies involves bioreactors for large-scale manufacturing, in which the cells are provided reagents at distinct stages and the cells self-assemble into clusters. Upon the completion of culture, these clusters are collected for direct infusion or embedded within materials for transplantation (10, 20). Biomaterial platforms provide a tool that can assist with assembly of cells into structures of tunable size, which is controlled by the material properties rather than the shear rate of the culture media (30). Additionally, as cells are cultured within the platform, the factors they secrete are deposited within a niche that may support their long-term function. Finally, upon completion of culture, the material platform can be directly transplanted with minimal cell manipulation and without disruption of the cellular niches.

Degradable hydrogels have been widely used for 3D differentiation of hPSCs (31-33). Matrigel-based materials have successfully produced islet organoids, as well as increased functional insulin secretion compared with 2D culture (34). Recent work has shown that functional β-cells can be produced from 2D culture under specific conditions. A recent protocol reproducibly generated cells with comparable or better functionality than suspension cultures (9). The condition optimized for this protocol was the addition of a 24-hour treatment with latrunculin A, which is a factor that depolymerizes the actin cytoskeleton and increases endocrine induction (9). While promising, Matrigel is sourced from murine tumors and generally not considered a translatable platform due to its batch to batch variability, undefined composition, and potential to transfer pathogens and facilitate tumor formation and growth (35-37). Collagen scaffolds and fibrin gels are two naturally occurring hydrogels that have been shown to support 3D hPSC differentiation into β-cells (32). A number of hydrogels have been employed for organoid culture with the ability to promote the formation of 3D structures that contain multiple cell types found in islets (33, 34, 38). These materials provide mechanical support for the formation of structures, and are modified with biological cues such as growth factors, cell adhesion ligands, immunomodulatory drugs, or enzyme degradation sites (39, 40). Proteomic and ribonucleic acid sequencing analysis was used to probe regulatory pathways during pancreatic differentiation of hPSCs, and the results suggested 3D confinement of the cells promoted integrin signaling and β-cell differentiation (31). These hydrogels can generate cells with a greater expression for β-cell genes and enhanced glucose-responsive function compared with 2D culture (31, 34).

Microporous scaffolds are also being developed as a platform for manufacturing hPSC-derived β-cells (29). Microporous scaffolds have commonly been used in regenerative medicine to allow seeded cells to form into tissues for subsequent transplantation. Specific to islets, the size of transplanted islets impacts insulin secretion and viability, and the pore size can regulate the size at which islet organoids form (41). After settling into pores, hPSCs form cell–cell interactions and mature into β-cells with transcription marker expression and glucose-stimulated insulin secretion comparable with or higher in scaffold cultured cells than in suspension culture (29). Additionally, hPSCs differentiated in microporous scaffolds were shown to form an extracellular matrix through deposition of collagen IV, laminin, and fibronectin (Fig. 2E). Extracellular matrix proteins could also be deposited onto the material surface in order to provide cues that promote the differentiation to islet organoids. Multiple biomaterials have been investigated for their potential to support islet organoid formation, such as polyethylene glycol or the poly (lactide-co-glycolide) (PLG), both of which have shown promise for islet transplantation in diabetic mouse models (22, 42-44). Importantly, the scaffolds that initially function as a support for organoid maturation can then be directly transplanted. Transplanting β-cells differentiated on a microporous scaffold avoids disruption of the existing cell–cell and cell–matrix interaction that may occur during the handling of suspension cultured cells for transplantation.

Transplantation

Pancreatic islets and β-cells are highly metabolically active with oxygen and glucose requirements that are greater than many other cell types, and the extrahepatic transplantation site must support these metabolic demands for a relatively large number of cells (≈500 000 islet equivalents (IEQ) per human transplant recipient) (45). In preclinical models, the kidney capsule is widely used, though this site is challenging to surgically access, cannot readily be altered as with the liver, and has the potential comorbidity of nephropathy, which would likely reduce healthy vasculature (46). The intraperitoneal and omental pouch have been widely used in preclinical and for clinical transplantation due to its large capacity for transplanting cells, which can also be delivered on scaffolds or with biologics that create a supportive environment. The subcutaneous space has an advantage of easy surgical access and could have sufficient space for large numbers of cells, though this site is less vascularized than other sites (47). The reduced vasculature in the subcutaneous tissue can lead to cell necrosis at the center of unmodified scaffolds as early as 24 hours after implantation (13). The intramuscular site offers a rich vascularization (48, 49), yet choosing an appropriate method of delivery at an intramuscular site is key to avoiding a detrimental fibrotic response (49). The native home of islets, the pancreas, has been considered as a transplant site, but it is widely considered less clinically promising due to the site requiring an invasive procedure (46). Collectively, a major challenge for clinical translation is providing an environment at the transplantation site that is able to support the metabolic needs of a large number of metabolically active cells, with hPSC-derived β-cells also requiring an environment that supports their maturation.

Enhancing survival

The initial challenge following transplantation is survival of the transplanted cells, and the properties of the transplantation site are critical. Oxygen is considered a limiting factor following transplantation, and a robust vasculature in close proximity to the graft can supply oxygen by diffusion. Scaffolds allow for oxygen and nutrient diffusion through the pores, yet passive diffusion is often insufficient to support the entire cell population for significant times. Alternatively, prevascularizing of the transplant site to enhance the local vascularity can enhance survival following the addition of islets. The temporary placement of a biomaterial to promote a proangiogenic inflammatory response successfully prevascularized a subcutaneous transplantation site and supported restoration of euglycemia following syngeneic islet transplantation (50). Fingolimod (FTY720), which targets sphingosine-1-phosphate (S1P) receptors on endothelial cells and smooth muscle cells to increase the density of surrounding microvascular networks, was loaded in scaffolds to prime an extrahepatic transplant site, and increased the formation of microvessels in diabetic mice, which supported islet function posttransplantation (51). Scaffolds loaded with vascular endothelial growth factor (VEGF) and fibrinogen have effectively prevascularized the site prior to islet transplantation (52-54). However, the efficacy of VEGF to induce vascularization is site dependent (55). A polyethylene glycol (PEG) hydrogel loaded with VEGF enhanced vascularization and reduced blood glucose when implanted in the epididymal fat pad, yet had no significant effect when implanted in the subcutaneous space or small bowel mesentery (55). While the epididymal fat pad, analogous to the omentum in humans, is considered a clinically translatable site, other methods to promote vascularization may be needed at alternative sites in order to support the needs of the cell graft. At a subcutaneous site, poly(D,L,-lactide-co-ε-caprolactone) scaffolds were allowed to prevascularize for 4 weeks prior to islet delivery. All mice in the diabetic cohort receiving 1200 islets into prevascularized poly(D,L,-lactide-co-ε-caprolactone) scaffolds achieved normoglycemia, while none of the mice in the cohort receiving the same islet dose into an unmodified subcutaneous pocket achieved normoglycemia (53). The Sernova Corporation has developed a device for islet transplantation called the Cell Pouch™, which is currently in a clinical trial, and is based on allowing 6 weeks for prevascularization in the subcutaneous space prior to islets being transplanted within cylindrical immune-isolating chambers (56). However, the islet loading levels and the diameter of cylinders within the Cell Pouch™ may contribute to hypoxia induced cell death, and future devices must limit the thickness of avascular tissue requiring oxygen diffusion (57). A complementary strategy to prevascularization involves drug delivery to inhibit apoptosis. Diabetic mice receiving islet transplants codelivered with V5, a BCL2-associated X protein–inhibiting peptide, had grafts with significantly enhanced function and a larger number of surviving islets (58). Collectively, prevascularizing the transplantation or blocking apoptosis are emerging strategies aimed at enhancing graft survival at early time points.

Oxygen has been directly delivered at the transplantation site to enhance survival prior to vascularization of the islets or islet organoids. Beta O₂ has developed an implantable device called βAir, which has a refillable oxygen chamber designed to provide oxygen to encapsulated islets. This system has successfully restored euglycemia in diabetic rats and minipigs (59, 60). However, significant functionality was not yet achieved in a phase I clinical trial (61). Alternatively, oxygen generating compounds can be embedded within materials that can immediately provide oxygen to the transplanted cells for days to weeks, which spans the critical time prior to integration with the host vasculature (Fig. 3A) (43, 62-64). Incorporation of hydrogen peroxide or calcium peroxide into scaffolds has been developed to locally provide oxygen release following transplantation and has been shown to improve graft function (62-64). Once the oxygen generating compound is depleted, hypoxia will return unless sufficient integration with the host vasculature has occurred. Increasing compound loading onto scaffolds may increase the length of oxygen delivery, yet the loading and release must be balanced to provide the necessary oxygen without the excessive production of harmful reactive oxygen species (64).

Figure 3.

Strategies for enhancing survival post transplantation. (A) Fibrin–heparin and VEGF-loaded collagen scaffolds containing oxygen generating microparticles (MPs) enhanced islet revascularization and improved restoration of euglycemia. Adapted from Montazeri L, Hojjati-Emami S, Bonakdar S, et al. Improvement of islet engrafts by enhanced angiogenesis and microparticle-mediated oxygenation. Biomaterials. 2016;89:157–165. doi:10.1016/j.biomaterials.2016.02.043 with permission from Elsevier. (B) Layered microporous PLG scaffolds loaded with the trophic factor exendin-4 enhance restoration of euglycemia compared to blank scaffolds. Adapted from Hlavaty KA, Gibly RF, Zhang X, et al. Enhancing Human Islet Transplantation by Localized Release of Trophic Factors From PLG Scaffolds. Am J Transpl. 2014;14(7):1523–1532. doi:10.1111/ajt.12742 with permission from John Wiley and Sons.

Enhancing integration

Long-term function of the transplanted cells is enabled by integration of the cells with the host vasculature, which can provide the necessary nutrients and allow for the sensing of blood glucose and the distribution of secreted insulin. Long-term islet function is maintained in part through a capillary network among the islets, indicating the need to integrate the cells with the vascular network posttransplantation (65). One of the major barriers to successful integration is the foreign body response. Viacyte’s cell replacement product VC-01 is an encapsulating device used to subcutaneously deliver hPSC-derived pancreatic progenitors. The recent clinical trial of VC-01 showed surviving cells out to 24 months, yet the function and integration of the device were limited by the foreign body response (66). The choice of biocompatible materials and the use of degradable platforms can influence the fibrotic response (67, 68). Microporous scaffolds, and degradable hydrogels support engraftment and revascularization (42, 69, 70). Microporous scaffolds typically have a large surface area to volume ratio, so that the distribution of cells near the surface minimizes the distance to host vasculature and the competition for nutrients. Additionally, the pores of the scaffold are generally sufficiently large to support vessel ingrowth; whereas the degradable hydrogels must undergo local degradation to create space for vessel ingrowth. Collectively, these examples illustrate that multiple strategies are being developed to enhance vascularization and engraftment, including the release of angiogenic factors and cell cotransplantation.

Angiogenic growth factors have been incorporated into scaffolds with the goal of enhancing the rate of vascularization and graft survival and function. The formation of robust vasculature networks depends on many aspects of the microenvironment, and the specific factors needed may differ depending on the properties of the site and the transplantation platform (71). Growth factors such as VEGF, platelet-derived growth factor, and acidic fibroblast growth factor have all been shown to increase vascularization of transplanted scaffolds and decrease time to euglycemia (13-15). The identity of the angiogenic factors, as well as its loading and release rate, can impact both vascularization and function. A collagen scaffold utilizing fibrin-conjugated heparin affinity based release of VEGF showed increased formation of new microvessels compared with control scaffolds containing only fibrin and VEGF (Fig. 3A) (62). Local release of these angiogenic factors increases the rate of vascularization, and importantly used factors that are endogenous to islets and pose minimal risk to the transplanted cells. As mentioned above, FTY720 was able to enhance prevascularization of scaffolds; however, the delivery of cells concurrently with FTY720 decreased islet function and failed to restore euglycemia (72, 73). Furthermore, not all growth factors result in long-term improvement of function. Acidic fibroblast growth factor, in particular, resulted in worse long-term function compared with control groups (15). The scaffold itself may also be modified to induce the expression of angiogenic factors by the transplanted or infiltrating cells. Heparin modified scaffolds upregulated VEGF expression and decreased time to restore euglycemia in diabetic mice when compared with control scaffolds (27). This release or induction of angiogenic factors can be combined with oxygen generating compound, with the coupling of these 2 approaches enhancing function following depletion of the oxygen generators (62). Combination strategies increase complexity of scaffold design, and require careful consideration of possible negative interaction. However, note that incorporation of multiple elements is likely to increase the regulatory burden.

Cotransplantation of support cells with β-cells can enhance integration with the host, and may provide cues that are beneficial to β-cells (74-76). Cotransplant of islets or hPSC-derived β-cells have been conducted with a variety of cell types, including endothelial cells (ECs), parathyroid glands, and fibroblasts (77-80). The mechanism of enhanced engraftment occurs through stimulation of growth factor release (parathyroid glands and ECs stimulate VEGF), and by providing the cells needed to generate new vasculature (ECs) (78, 79). Providing multiple cell types has the potential to recapitulate native pancreatic tissue, which may lead to enhanced function and engraftment. ECs have enhanced the maturation of hPSC-derived pancreatic progenitor cells in vitro (38, 80). An important consideration when designing an EC/β-cell cotransplantation strategy is that ECs experience increased cell death and dysregulation in elevated glucose environments (81). In order to support engraftment of hPSC-derived β-cells in vivo, ECs must be able to survive the posttransplant environment. Furthermore, the cotransplantation approach may differ for islets, which are already functional for restoring euglycemia, relative to hPSC-derived β-cells that may require further in vivo maturation.

Enhancing function

Local delivery of trophic factors to transplanted β-cells can stimulate receptors and signaling pathways known to regulate survival, proliferation, differentiation, and insulin secretion. The glucagon-like peptide-1 agonist exendin-4 stimulates glucose-dependent insulin secretion in β-cells, promotes β-cell proliferation, and protects against apoptosis (82, 83). In vitro culture of cells in a supporting hydrogel and receiving Exendin-4 had higher insulin secretion and less apoptosis than controls (84). The local and sustained release of exendin-4 allowed for up to a 25% decrease in the number of human islets in order to enhance glycemic control (Fig. 3B) (43). Scaffolds similarly releasing exendin-4 were used to support hPSC-derived pancreatic progenitors, and enhance maturation and c-peptide secretion (85). Alternatively, nerve growth factor (NGF) has promoted glucose responsive insulin secretion in β-cells through activation of calcium channels (86). Islets were transplanted into diabetic mice on a gelatin sponge, known as Gelitaspon, with or without NGF (87). Mice receiving islets transplanted on Gelitaspon soaked in 100 ng/mL NGF had significantly lower blood glucose and an improved response to intraperitoneal glucose tolerance test compared with the control group. A number of other trophic factors (Prolactin (88), Growth hormone (GH), also known as somatotropin (16, 43, 82, 83, 89), Hepatocyte growth factor (90), insulin-like growth factor-1 (91,92), Betacellulin, and Activin-A (93)) have been identified with potential to influence β-cell responses posttransplantation, yet these factors have not been investigated with scaffold based transplantation.

Immune Protection

The process of immune recognition and immune destruction of transplanted cells for an allogeneic transplantation occurs in a 4-step manner described as follows: (1) inflammation; (2) activation of allogeneic or self-dendritic cells (DCs) and migration to draining lymph nodes; (3) T cell activation by DCs, resulting in expansion of antidonor T cells; and (4) migration of T cells to the graft where they mediate cytotoxicity (16, 94). The development of the Edmonton Protocol dramatically improved clinical outcomes in islet transplantation by using sirolimus, low-dose tacrolimus, and induction anti-interleukin-2 receptor antibody. While improvements have been made in the immunosuppression protocol, these patients require lifelong immunosuppression, which can be problematic for a disease with a juvenile onset (95). The side effects brought by the permanent use of such systemically administered agents include increased vulnerability to infection, cytotoxicity or even diabetogenicity to the transplanted islets/cells. This section describes emerging research counteracting immune recognition and immune destruction, either locally at the site of transplant and/or systemically, that could enhance engraftment of the allogeneic cell transplant.

Local immune modulation

The local modulation of immune responses is motivated by at least 2 observations: (1) that the inflammatory response contributes to islet loss following hepatic infusion (96), and (2) immunoprivileged sites, such as those in the thymus, brain, and testis (46), can limit immune responses without the need for systemic immune suppression. The inflammatory response following cell transplantation reduces transplanted cell survival, often substantially, and is mediated initially by innate immune cells, whereas adaptive immune cells mediate cell killing in many autoimmune and allogeneic cell transplantations (16). The function of innate and adaptive immune cells can be modulated at immunoprivileged sites in the body, such as the anterior chamber of the eye and the testis. The immunoprivileged properties for both the ocular and testes microenvironments result from a combination of structural features and immunoregulatory and immunosuppressive molecules (97). In the testes, the Sertoli cells provide a physical blood barrier to isolate germ cells from the immune system along with a host of immunoregulatory compounds that suppress proinflammatory responses from the immune cells to maintain the immunoprotected environment (98). The eye and testes are generally not considered translationally relevant due to the large number of islets that would need to be delivered into these small organs. Nevertheless, the basic principles associated with these immunoprivileged microenvironments have served to inspire approaches to protect allogeneic cell transplants at extrahepatic sites.

The local release of cytokines from scaffolds has been employed to modulate the inflammatory reaction and enhance both cell survival and function. These drugs and cytokines stimulate the recruitment of suppressive immune cells or polarize immune cells toward an anti-inflammatory phenotype. Examples of immune regulatory molecules that have been investigated for protection of islets and β-cells include transforming growth factor-β (TGF-β), interleukin (IL)-10, IL-4, C-C motif chemokine ligand 2 (CCL2 [MCP-1]), C-X-C motif chemokine ligand 12 (CXCL12 [SDF1]), indoleamine 2,3-dioxygenase 1 (IDO1), leukotriene B4 (LTB4), IL-33, and prostaglandin E2 (PGE2) (16, 19). An important consideration is the impact these factors have at the specific site of transplantation. While IL-33 exerted graft-protective effects in both murine skin transplant and cardiac transplant models (99-101), IL-33’s ability to protect islets has shown mixed results. IL-33 delivery with islet transplantation into the peritoneal fat IL-33 upregulated graft-protective T cells and prolonged graft survival; however, at this site, IL-33 also led to an increased expression of proinflammatory cytokines responsible for potential host protective immunity (44). TGF-β1 delivered from porous scaffolds downregulated tumor necrosis factor-α, IL-12, and MCP-1 population by at least 40% compared with control scaffolds (102). The transplanted allogenic islets on the TGF-β1 releasing scaffolds had an enhanced engraftment and functioned significantly longer than those transplanted on empty scaffold controls.

Chemokine delivery influences the function of numerous cell types within the environment. CXCL12 delivery protected transplanted islets and stem cell–derived β-cells, through either protection or enhanced function for allogeneic and xenogeneic sources (103). CXCL12 directs monocyte differentiation towards a macrophage phenotype with immunosuppressive functions (17). Additionally, dendritic cells derived following CXCL12 modulation induce activity from antigen-specific regulatory T cells (17, 104). CXCL12 can directly signal to β-cells for survival and regeneration (105), and increased the homing of autologous MSCs and HSCs that can substantially reduce the infiltration of inflammatory cells, and improve blood vessel formation (106, 107).

Chemokines invoked by tumor cells that protect tumors from immune destruction have also been employed for local immunomodulation. CCL22 is released by tumor cells to recruit regulatory T cells (Tregs) to the site through a chemokine receptor (CCR4) expressed on Tregs. Tregs possess potential capabilities of mitigating autoimmunity in T1D and inducing local and systemic tolerogenic effects. This tolerogenic effect is achieved by suppressing T-cell and DC maturation and activation by secreting immunosuppressive factors, such as IL-10 and TGF-β1 (108, 109). Donor-specific tolerance to fully mismatched islet allografts has been achieved by locally recruiting Tregs in CCL22-expressing allografts (Fig. 4A) (110-112). Allogeneic islets transduced to express CCL22 conferred prolonged protection and preservation of transplanted murine β-cells from rejection and maintained normoglycemia in 75% of recipients over 80 days, with elevated number of Tregs in the islet grafts and an absence of antidonor antibodies or lymphocyte proliferation after exposure to donor splenocytes. In a different transplant model, CCL22-releasing microparticles delivered locally at a graft prolonged complete MHC mismatched allograft survival over 200 days (111). Effector T cells were significantly reduced in skin and draining lymph nodes of treated transplant recipients, with a resulting tissue architecture similar to muscle and skin biopsies from normal tissue.

Figure 4.

Immunomodulation strategies. CCL22 released microparticles in VCA model and Fas-L attached microgels in T1D model. (A) Characterization and 40-day release profile of recruitment-MP. Treatment with Recruitment-MP (50 mg) prolongs allograft survival indefinitely. Hindlimb allograft survival curve showing indefinite survival (>200 days) in 6 of 8 rats treated with Recruitment-MP (50 mg). Adapted from Fisher JD, Zhang W, Balmert SC, et al. In Situ Recruitment of Regulatory T Cells Promotes Donor-Specific Tolerance in Vascularized Composite Allotransplantation. Vol 6.; 2020. http://advances.sciencemag.org/. with permission from Creative Commons Attribution-Non Commercial license. (B) Flow-focusing microfluidics were used to generate biotinylated microgels from biotin-functionalized PEG-4MAL macromers. Islet graft survival: SA-FasL-presenting or control microgels and unmodified BALB/c islets were co-transplanted under the kidney capsule of chemically diabetic C57BL/6 recipients. Adapted from Headen DM, Woodward KB, Coronel MM, et al. Local immunomodulation with Fas ligand-engineered biomaterials achieves allogeneic islet graft acceptance. Nat Mater. 2018;17(8):732–739. doi:10.1038/s41563-018-0099-0 originally published in Nature Materials, with permission from author Dr. Lonnie D. Shea.

Natural immunoprivileged sites often employ constitutive expression of FasL to control the entry of lymphoid cells expressing Fas. A synthetic FasL was developed by Yolcu et al., which consisted of an engineered streptavidin-FasL (SA-FasL) that could be attached to biotinylated pancreatic islet grafts. This FasL modification of islets in combination with a transient course of rapamycin treatment (0.2 mg/kg daily for 15 days) resulted in indefinite survival and function of allogeneic cells in the absence of chronic immunosuppression (113). In another study, graft recipients were also systemically immunomodulated by intraperitoneal injection of donor SA-FasL- engineered splenocytes on days 1, 3, and 5 after the transplantation of the S-FasL-attached islets. This regimen resulted in the survival of all allogeneic islet grafts for the 250-day observation period since SA-FasL protein preferentially induced apoptosis in autoreactive T-effector cells while sparing CD4⁺CD25⁺FoxP3⁺ T-regulatory cells (114). This result has recently been translated to biomaterials used for islet transplantation, with the materials being biotinylated and functionalized with FasL rather than the islets. PEG hydrogel beads modified with FasL that were cotransplanted with islets, or FasL-modified porous PLG scaffolds supported allogeneic islet transplantation and function for over 200 days (Fig. 4B) (112, 115). Decreased numbers of both CD4+ and CD8+ T-effector cells and increased number of T-regulatory cells were observed in the tissues of mice receiving SA-FasL-engineered microgels. These islets were transplanted into either the kidney capsule or the epididymal fat pad with transient administration of rapamycin. These islets had sustained survival and normalized blood glucose levels in chemically diabetic graft recipients. Long-term graft recipients generated a normal immune response to donor alloantigens, implicating localized tolerance to the graft.

An alternative to delivery or presenting factors within the environment is to directly transplant cells to modulate the immune response. Biomaterial scaffolds can serve to support the colocalization of the cell types. Perhaps the most widely studied cell type for cotransplantation involves mesenchymal stem cells (MSCs) (116, 117). MSCs, derived from the infrapatellar fat pad, have been cotransplanted with islets in order to capitalize on the immunomodulatory (118) and proangiogenic (119) properties of MSCs. MSC treatment significantly enhanced islet engraftment and function, resulting in stabilized blood glucose levels, reduced exogenous insulin requirement, and increased numbers of regulatory T cells in peripheral blood (120). An alternative to recruiting Tregs is the direct transplantation of Tregs with islets. Tregs cotransplanted with islets on a microporous PLG scaffold provided long-term graft protection from an autoimmune response (121). The seeded Tregs were replaced by recipient-derived Tregs over time, suggesting that host-derived Tregs induce tolerance to islet grafts on the PLG scaffold. Normoglycemia was restored, and cotransplanted Tregs extended graft survival indefinitely in several instances. Interestingly, in this study, local codelivery of Tregs with an initial islet transplant also led to protection against autoimmune destruction of a second islet transplant (without Tregs), indicating systemic tolerance to islet antigens.

Antigen-specific immune modulation

Antigen-specific immune modulation strategies represent a significant opportunity for all cell and solid organ transplantation, which currently employ systemic immunosuppression to prevent rejection of the graft yet is associated with increased risk of infection and neoplasia for the patient. Modulation of systemic immune responses may be an alternative or complementary strategy to local immune modulation. These systemic therapies may be administered prior to cell transplantation, and/or as needed following transplantation. These strategies aim to either (1) delete autoreactive or alloreactive T cells, (2) induce anergy in the naïve cells, and/or (3) induce the formation of regulatory T cells that can modulate the graft reactive T cells. Immunosuppressive drugs, such as those used in the Edmonton Protocol, function by suppressing the function of one or more cell types in an antigen-independent manner. Localized delivery of these immunosuppressive factors can reduce the required dose and avoid some of the systemic side effects of the compounds (122). This localized delivery strategy can prolong graft survival; however, the graft contains a finite quantity of drug and seems unlikely to support indefinite graft survival and function. Numerous strategies are being investigated, such as pharmaceuticals, nanoparticles, or engineered cells, to limit allograft rejection and have been reviewed elsewhere (123-125), and we describe a few representative examples applied to islet transplantation.

Antigen-specific immune modulation offers the opportunity to prevent rejection of the graft while leaving the remainder of the immune system intact. One approach toward this goal has been the use of donor splenocytes, which have donor antigen and upon treatment with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (ECDI) and subsequent intravenous infusion has had success in inducing tolerance in a variety of autoimmune and allotransplant models (18, 126-128). Treated splenocytes, which induce apoptosis in the donor splenocytes, are infused with one dose prior to transplantation and 1 dose posttransplantation. Infusions of ECDI-treated donor splenic antigen-presenting cells result in indefinite survival of allogeneic islet grafts in the absence of immunosuppression (Fig. 5A) (18, 129, 130). The site of islet transplantation does influence the effectiveness of tolerance induction, with the kidney capsule and epididymal fat pad having high levels of success, and with 1 report indicating lower levels of tolerance induction in the liver (128). Recently, this approach of ECDI-treated splenocytes has been translated to studies in nonhuman primates using ECDI-treated apoptotic donor leukocytes. The regimen promoted stable islet allograft tolerance (>356 days) in monkeys by suppressing effector cell expansion and expanding regulatory networks (131).

Figure 5.

Immunomodulation strategies. ECDI-treated donor lymphoid cell infusion and ECDI-attached PLG nanoparticles in T1D model. (A) Glycemic control posttransplant: 4 ECDI-DLI experimental group and 2 control group diabetic nonhuman primates each promptly achieved normoglycemia post-allogeneic islet transplant. Adapted from Lei J, Kim JI, Shi S, et al. Pilot Study Evaluating Regulatory T Cell-Promoting Immunosuppression and Nonimmunogenic Donor Antigen Delivery in a Nonhuman Primate Islet Allotransplantation Model. Am J Transplant. 2015;15(10):2739–2749. doi:10.1111/ajt.13329 with permission from John Wiley and Sons. (B) Scanning transmission electron microscopy image and size of PLG particles prior to lysate coupling with ECDI. A short course of low dose rapamycin with PLG-dAg demonstrated significantly greater islet allograft survival (n = 11) compared with mice treated with rapamycin alone (n = 9) **P < .01. Adapted with permission from Bryant J, Hlavaty KA, Zhang X, et al. Nanoparticle delivery of donor antigens for transplant tolerance in allogeneic islet transplantation. Biomaterials. 2014;35(31):8887–8894. doi:10.1016/j.biomaterials.2014.06.044 with permission from Elsevier.

The capability of ECDI-coupled splenocytes to treat autoimmune disease and prevent transplant rejection has been extended to synthetic nanoparticles, which have the potential for greater consistency in GMP manufacturing that would facilitate translation. In autoimmune models of diabetes, antigen-loaded nanoparticles have been able to prevent islet destruction with nanoparticles administered prior to and after disease initiation (132). Additionally, this approach has been extended to allogeneic cell transplantation by coupling of cell lysates to the nanoparticles and subsequent infusion prior to and after allogeneic cell transplantation (114). The particles in combination with a 3-day course of low-dose rapamycin were able to induce long-term graft function in the absence of sustained immunosuppression (Fig. 5B) (130). The particles were effective at limiting T-cell activation induced by indirect antigen presentation; however, the particles were not able to influence T-cell activation by direct antigen presentation (133). This inability to affect T-cell responses resulting from direct antigen presentation likely reflects that antigen coupling to particles or cells does not appear to modulate activation induced by major antigen mismatch (133).

Final Thoughts

Substantial advances have been made in recent years for the differentiation of hPSC to β-cells. These advances have substantially raised the prospect of a cell-based therapy for T1D and highlight the critical need for cell manufacturing and delivery systems that can provide adequate numbers of cells that can efficiently engraft and function for long times without the need for sustained immunosuppression. Advances in CRISPR technology may also allow for the engineering of cells that have improved capacity for insulin secretion, or the ability to evade the immune system (134). These opportunities in cellular engineering will benefit from technologies that can provide control over the cellular microenvironment, both during cell manufacturing and following transplantation. These biomaterial platforms provide the tools to sequester key signaling factors, facilitate the organization of cells into structures, support cell–cell and cell–matrix interactions, recruit or deliver accessory cells to aid maturation or function, limit inflammatory responses, and evade the adaptive immune responses. Establishing a supportive environment at the time of manufacture or transplantation can create a stable niche that can support cell function for extended periods of time. Collectively, the capabilities of hPSC-derived β-cells have advanced, and engineering the microenvironment can help to translate these capabilities.

Glossary

Abbreviations

DC

dendritic cell

EC

endothelial cell

ECDI

1-ethyl-3-(3-dimethylaminopropyl)carbodiimide

ESC

embryonic stem cell

hPSC

human pluripotent stem cell

IBMIR

instant blood-mediated inflammatory reaction

IL

interleukin

iPSC

induced pluripotent stem cell

MSC

mesenchymal stem cell

NGF

nerve growth factor

PEG

polyethylene glycol

PLG

poly (lactide-co-glycolide)

PTH

parathyroid gland

T1D

type I diabetes mellitus

TGF

transforming growth factor

Treg

regulatory T cell

VEGF

vascular endothelial growth factor

Additional Information

Disclosure Summary: L.D.S. consults for and has financial interests in Cour Pharmaceuticals.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.