| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Review Article |
Department of Pharmaceutical Sciences, University of Tennessee Health Science Center, Memphis, Tennessee
Abstract
Abstract I. Introduction II. Obstacles to the Success of Islet Transplantation A. Loss of Islet Viability during Isolation and Culture B. Inadequate Revascularization of Transplanted Islets C. Autoimmune Recurrence and Immune Rejection D. Islet Mass and Site of Transplantation III. Biological Strategies for Improving the Success of Islet Transplantation A. Prevention of Immune Destruction of Transplanted Islets 1. Immunosuppression. 2. Immune Modulation and Tolerance. a. Antibody pretreatment for xenografts. b. Removal of passenger leukocytes for allografts. c. Cytokine modulation. d. Intrathymic islet or alloantigen injection. e. Peripheral tolerance by T-cell inactivation or depletion. f. Coactivation and costimulation blockade. g. Dendritic cell infusion. B. Counteracting Insufficient Tissue Supply 1. Xenotransplantation. 2. Regeneration Therapy. a. Replication of pre-existing beta-cells. b. Ectopic expression of beta-cell phenotype. c. Using embryonic stem cells. d. Using adult stem cells. 3. Insulin-Producing Cell Lines. IV. Biomaterial-Based Strategies for Improving the Success of Islet Transplantation A. Immunoisolation of Transplanted Islets 1. Types of Devices for Immunoisolation. a. Intravascular macrocapsules. b. Extravascular macrocapsules. c. Extravascular microcapsules. 2. Biocompatibility Considerations. 3. Methods for Microencapsulation of Islets. 4. Enhancing the Performance of Microencapsulated Islets. B. Surface Modification of Islets V. Nucleic Acid-Based Therapeutics for Improving the Success of Islet Transplantation A. Nonviral Gene Delivery B. Viral Gene Delivery 1. Vector Backbone Modification. 2. Surface Modification of Viral Vectors. C. Antisense Oligonucleotides and RNA Interference for Gene Silencing VI. Concluding Remarks
Islet transplantation may be used to treat type I diabetes. Despite tremendous progress in islet isolation, culture, and preservation, the clinical use of this modality of treatment is limited due to post-transplantation challenges to the islets such as the failure to revascularize and immune destruction of the islet graft. In addition, the need for lifelong strong immunosuppressing agents restricts the use of this option to a limited subset of patients, which is further restricted by the unmet need for large numbers of islets. Inadequate islet supply issues are being addressed by regeneration therapy and xenotransplantation. Various strategies are being tried to prevent 
-cell death, including immunoisolation using semipermeable biocompatible polymeric capsules and induction of immune tolerance. Genetic modification of islets promises to complement all these strategies toward the success of islet transplantation. Furthermore, synergistic application of more than one strategy is required for improving the success of islet transplantation. This review will critically address various insights developed in each individual strategy and for multipronged approaches, which will be helpful in achieving better outcomes.
Diabetes mellitus is a global disease with immense economic and social burden. type I diabetes is insulin-dependent diabetes mellitus (IDDM1) and is often called juvenile-onset diabetes. This autoimmune disorder leads to the destruction of insulin-producing pancreatic -cells (Mathis et al., 2001
). On the other hand, type II diabetes is noninsulin-dependent diabetes mellitus (NIDDM) and is often called adult-onset diabetes. Type II diabetes arises from peripheral resistance to insulin, leading to insulin overproduction by islets. As the disease progresses, the insulin-producing -cells in the islets of Langerhans in pancreas get desensitized to the persistently high glucose signal, thus leading to reduced insulin production by islets in response to normal glycemic stimulation (Costa et al., 2002). Islet dysfunction plays a key role in late-stage type II diabetes (Porte and Kahn, 1995; Pratley and Weyer, 2001). Late-stage type II diabetic patients often require insulin therapy in much higher doses because of peripheral insulin resistance (Holman and Turner, 1995). The number of people with diabetes is expected to exceed 350 million by 2010, and 10% of these are expected to have type I diabetics (Serup et al., 2001). Although some of the approaches outlined in this review will also be relevant for treating type II diabetes, we will focus mainly on islet transplantation as a treatment option for type I diabetes.
Current approaches for treating type I diabetes include 1) exogenous insulin therapy and 2) pancreas transplantation. Although daily glucose monitoring and exogenous insulin administration has been the standard therapy since the discovery of insulin, the poor control of blood glucose fluctuations with this therapy leads to many severe complications including neuropathy, nephropathy, retinopathy, heart disease, and atherosclerosis (Bailes, 2002; Bloomgarden, 2004; Hill, 2004). In 1993, the Diabetes Control and Complications Trial showed that strict control of blood glucose levels reduced the risk of developing diabetes-related complications. Many improvements in the formulations and delivery systems of insulin promise to improve therapeutic outcomes of insulin administration. However, poor patient compliance, the inherent complications of using certain devices for insulin delivery, and the risk of hypoglycemia have prompted the search for a "cure" of diabetes. Pancreas transplantation is currently the only option available that promises to cure the disease. This procedure, however, requires major surgery and lifelong immunosuppression (Robertson et al., 2000). Therefore, most pancreas transplantations are done in diabetic patients with severe late-stage complications and undergoing kidney transplantation and immunosuppression. These procedures are known as simultaneous pancreas kidney or pancreas after kidney transplantation (Sutherland et al., 2001). Pancreas transplantation is, therefore, not available to a vast majority of diabetic patients as a therapeutic option.
Islet transplantation, on the other hand, promises to be a cure at least as effective as pancreas transplantation, while being much less invasive. Islet transplantation involves the isolation of functional islets from cadaveric, multiorgan donors. These islets are then injected into the hepatic portal vein of the diabetic patient, from where they get deposited in well-perfused liver sinuses. Islet transplantation differs from other tissue and organ transplantation approaches in being a heterotropic graft, a graft that is located on a site other than the natural location of the tissue (Rossini et al., 1999). Islets are not transplanted homotropically in the pancreas of the recipient because the pancreas is a highly sensitive tissue. Any injury or manipulation of the pancreas leads to severe pancreatitis, with accompanying pain and tissue destruction (Morrow et al., 1984). Islet transplantation can provide certain advantages that are not available with pancreas transplantation including the potential for modifying tissue immunogenicity through in vitro culture or gene therapy approaches, tissue encapsulation for immunoisolation, potential for engraftment in immunoprivileged sites, and the possibility of using alternative tissue sources including xenogenic islets and stem cell derived -cell lines.
Scientists have made many advances in islet transplantation in recent years. In June 2000, Shapiro et al. at the University of Alberta in Edmonton, Canada, published results of an exceptionally successful case of islet transplantation wherein seven of seven patients were insulin-free at the end of 1 year (Shapiro et al., 2000). The latest results of their phase II clinical trials indicate a success rate of 82% with islet transplantation alone carried out in type I "brittle" diabetic patients (Oberholzer et al., 2003). This success rate matches that of pancreas transplantation (Sutherland et al., 2001) and has essentially revitalized this field and led many new centers to initiate the program of islet transplantation (Chang et al., 2004). Widespread clinical application of this procedure, however, is currently limited by the need for lifelong immunosuppression and the need for two to four donor pancreases per recipient.
The challenges to successful transplantation of islets include 1) isolation, culture, characterization, and preservation of islets, 2) inflammation and autoimmune-mediated destruction and alloimmune rejection of transplanted islets, 3) failure to revascularize, 4) low transplanted mass and high metabolic demand on the tissue, and 5) a limited supply of islets for widespread clinical use (Table 1) (Robertson, 2001; Lakey et al., 2003; Ricordi and Strom, 2004). The major barrier to islet transplantation is the need for lifelong immunosuppression of the recipient. Once this barrier is overcome, the limitation of islet supply will hamper the use of this procedure (Larkin, 2004). In the present review, we discuss these current challenges to the success of islet transplantation and the application of biological and biomaterial-based approaches toward improving the clinical success of islet transplantation.
|
Prevention of immune rejection of transplanted islets has conventionally been the use of generalized immunosuppression with its significant side effects (Inverardi et al., 2003; Chang et al., 2004). Several immune-modulation strategies have recently evolved to selectively block the immune responses against the graft. Some of these strategies are already used clinically for other applications, e.g., muromonab-CD3 in acute renal rejection (Smith, 1996), whereas others have been tested individually or in combination in several animal models. These will be discussed in section III.A.
To generate alternative sources of islets, both the use of islets from alternative species (xenotransplantation) and generation of islets and/or insulin-producing cells of human origin (regeneration therapy) have been explored (Scharfmann, 2003; Hussain and Theise, 2004). Whereas the former source suffers primarily from enhanced immune destruction of transplanted tissue due to xenoantigen rejection, the latter has met with limited success hitherto in generating large amounts of tissue. The use of replicating cell lines, on the other hand, has significant safety concerns for human applications. These issues and progress in these fields are discussed in section III.B.
Biomaterials have the potential to improve the outcome of islet engraftment by encapsulating islets before transplantation. This strategy of immunoisolating islets has met with limited success due to engineering, process, and biomaterial limitations. Furthermore, the diffusional barrier limits the free supply of oxygen and nutrients, resulting in hypoxia and lack of revascularization of islets. These problems may be circumvented by using surface coating of islets. These approaches are elaborated on in section IV.
An alternative to physical isolation of the transplanted tissues from the host involves modulating the gene expression profile of islets before transplantation by nucleic acid-based approaches. These strategies include both gene expression and gene knockdown approaches through the use of either nonviral or viral vectors. Furthermore, ex vivo transfection of islets before transplantation has the potential to be safely included in the clinical islet transplantation protocol. Several promising strategies will be discussed in section V. We finally conclude with an outlook for the future and the strategies that hold the most promise for solving some of the toughest problems currently hampering the clinical success of islet transplantation.
II. Obstacles to the Success of Islet Transplantation
Islet transplantation may be done in three different modalities (Federlin et al., 2001). Transplantation of islets isolated from the same animal is referred to as autologous or autotransplantation, transplantation of islets from the same species is referred to as allogeneic or allotransplantation, and xenogenic or xenotransplantation refers to the use of islets from a different species. Syngeneic transplantation is a special case of allotransplantation in which the graft donor and the recipient are monozygotic. Autotransplantation involves transplanting islets of the patient to himself or herself in patients who necessitate total pancreatectomy, e.g., chronic pancreatitis (White et al., 2000). Although allotransplantation is the preferred modality for immunological and safety reasons, severe constraints in tissue availability necessitate exploring getting islets from alternative sources (xenotransplantation) to make this option available to a substantial fraction of patients.
In the overall process of islet transplantation, islets are obtained from the pancreas of cadaveric multiorgan donors in the case of allotransplantation or from animals (e.g., pig) in the case of xenotransplantation. The pancreases are digested with collagenase that disintegrates the intercellular matrix of collagen, thus releasing islets. The islets are isolated, purified, tested, and cultured before being transplanted into the recipient, which is usually done by a simple injection in the hepatic portal vein that deposits the islets in the liver. The overall process of islet isolation, purification, preservation, and quality control poses serious challenges to the clinical outcome of islet transplants (Menger and Messmer, 1992; London et al., 1994; Lakey et al., 2002).
A. Loss of Islet Viability during Isolation and Culture
Efficient isolation of pure islets, without inflicting significant damage, is the key to successful islet transplantation. The pancreas contains exocrine, endocrine, and ductal cells. The endocrine cells, arranged in islets within the pancreas, consist of 
-, -, 
-, and PP cells that secrete glucagon, insulin, somatostatin, and pancreatic polypeptide, respectively (Menger et al., 1994). Figure 1 shows the location of the pancreas inside the abdominal cavity and the microanatomy of islets. Islet portal circulation, with blood flow from -to -to -cells, as well as afferent innervation from the central nervous system, have a role in hormone secretion from constituent cells (Helman et al., 1982; Steffens and Strubbe, 1983). Islet isolation from the pancreas essentially involves dissociation of islets from the exocrine pancreas by enzymatic digestion combined with mechanical agitation. Isolated islets are then purified by density gradient centrifugation. A critical balance of composition, process, and duration of collagenase digestion is required for isolating islets with high purity, integrity, and viability with a sufficient yield. The enzymatic digestion process disrupts islet-to-exocrine tissue adhesive contacts (Wolters et al., 1992). Thus, whereas lower duration or inappropriate composition of collagenase will lead to incomplete purification of islets from exocrine tissue along with reduced yield, increased duration of collagenase exposure adversely affects within-islet cell-to-cell adhesion, leading to loss of islet integrity and viability.
|
) in rat, dog, pig, and human pancreas. The authors found that intraislet associations are predominantly cell-to-cell in all four species, whereas islet-to-exocrine tissue interactions are predominantly cell-to-cell in pig pancreas and cell-to-matrix in canine pancreas, which has completely encapsulated islets. In the case of rat and human pancreas, the situation was intermediate with a tendency toward predominance of cell-to-matrix interactions. These observations explain the reason that pig islet isolation is difficult (White et al., 1999; Omer et al., 2003) and also point to the kind of digestive enzyme mixtures that may be used for a given species. For example, intraislet cell-to-cell adhesion is protease-sensitive, whereas the extraislet cell-to-matrix adhesion is collagenase-sensitive (McShane et al., 1989; Wolters et al., 1992). Thus, the use of highly pure collagenase preparations is desirable to isolate pure islets with the least possible damage to the islets themselves. The presence of protease in the collagenase preparations reduces the yield and quality of isolated islets (Vos-Scheperkeuter et al., 1997). However, removing exocrine tissue is more efficient with the use of collagenase preparations containing protease for the isolation of pig islets (van Deijnen et al., 1992). Different enzyme composition and process modifications have been evaluated for isolation of pancreatic islets (Lakey et al., 1998b; Bucher et al., 2005). These include the use of additives in the collagenase solution (Arita et al., 2001), composition of density gradient materials (Lakey et al., 1998a), and digestion procedures (Lakey et al., 1999). Such studies have led to development of various species-specific liberase enzyme blends (Linetsky et al., 1997; Brandhorst et al., 1999; Olack et al., 1999) and the automated Ricordi chamber (Poo and Ricordi, 2004) for optimized islet isolations. These aspects, however, are not discussed in detail here, and the interested reader is referred to these publications.
The use of slightly impure islet preparations and coculture with extracellular matrix components such as collagen (Nagata et al., 2002) have been shown to enhance the viability and function of isolated islets. In addition, supplementation of culture medium with small intestinal submucosa was shown to improve islet functioning and viability (Lakey et al., 2001a). Media composition, seeding density, and temperature play a significant role (Falqui et al., 1991; Murdoch et al., 2004). In addition, islet coculture with pancreatic ductal epithelial cells was shown to be useful for maintaining islet viability and function after isolation (Gatto et al., 2003). Pancreatic ductal epithelial cells have been considered as putative stem cells for islets and an essential component of the extracellular matrix, which plays an important role in secreting appropriate growth factors that support islet viability. Gatto et al. (2003) found that long-term culture, as well as cryopreservation, decreased the viability of human pancreatic islets, which was prevented by coculture with ductal epithelial cells at 33°C. In a different study, coculture with ductal epithelial cells helped maintain structural integrity and viability of the islets (Ilieva et al., 1999).
Fraga et al. (1998) have evaluated different media supplements for the extended culture of human pancreatic islets. Islet culture in Connaught Medical Research Laboratories medium (Life Technologies, Inc., Rockville, MD) was compared with the supplementation of either 10% fetal bovine serum (standard medium) or insulin-, transferrin-, and selenium-containing medium (also known as the Memphis medium). Long-term culture of islets in insulin-, transferrin-, and selenium-containing medium was shown to maintain the viability of islets with no adverse effect on in vivo function in the NOD-SCID mouse model (Gaber et al., 2001; Gaber and Fraga, 2004; Rush et al., 2004) and correlated with islet function after transplantation in human subjects (Gaber et al., 2004).
Islet viability during culture is also adversely affected by hypoxia to the cells in the inner core of islets (Dionne et al., 1993; Gorden et al., 1997; Vasir et al., 1998). Although it may be difficult to prevent a hypoxic condition of the inner islet cell mass during in vitro culture, genetic modulation of islets to express genes that promote rapid revascularization upon transplantation and reduced culture time could play an important role in preventing hypoxic damage to the islets (Mahato et al., 2003; Cheng et al., 2004; Narang et al., 2004). These approaches are discussed in section V.
B. Inadequate Revascularization of Transplanted Islets
Islets are like an organ in themselves with extensive intraislet vasculature, formed of fenestrated capillary endothelial cell lining, which is essential for the supply of oxygen and nutrients to the cells in their inner core (Carroll, 1992; Menger et al., 1994). To determine the presence and orientation of intraislet vasculature, Menger et al. (1992) transplanted 8 to 10 isolated hamster islets into the dorsal skinfold chamber of syngeneic animals. Fourteen days post-transplantation, the microvasculature of the transplanted islets was perfused by an injection of 200 µl of 5% fluorescein isothiocyanate-conjugated dextran (mol. wt. 150,000), and the islet vasculature was analyzed by intravital fluorescence microscopy. As seen in Fig. 2, the supporting arterioles penetrate into the periphery of the islet and break into capillaries within the graft. Glomerulus-like capillary perfusion is directed to microvessels located within the core of the islet (Menger et al., 1994).
|
1% of the pancreas by weight (Lifson et al., 1980; Jansson and Hellerstrom, 1983). This vasculature gets disrupted during the process of islet isolation and culture, which causes an accumulation of endothelial fragments and compromises perfusion of the core of islets. Therefore, rapid revascularization is crucial for islet engraftment, survival, and function post-transplantation (Brissova et al., 2004). Successful islet grafts have been observed to regenerate the microvasculature within 10 to 14 days of transplantation (Menger et al., 1994; Vajkoczy et al., 1995; Merchant et al., 1997; Beger et al., 1998; Furuya et al., 2003). However, the proportion of islets that restore their original vasculature upon transplantation is limited and variable. This issue is as a fundamental factor in determining long-term graft survival and function.
Because of the disruption of intraislet vasculature, islets in culture, as well as during the initial few days of transplantation, depend on the diffusion of oxygen and nutrients from the periphery. Vascular endothelial cells are lost during culture (Mattsson, 2005), making endothelial cell expansion essential to the islet revascularization process. Islet survival and long-term function after transplantation are often antagonized by the lack of reestablishment of capillary networks within the islets (Narang et al., 2004), which also exacerbates immune destruction of transplanted islets (Lukinius et al., 1995).
Although transplantation in highly perfused organs such as the liver promises to provide adequate tissue bathing to provide nutrition by diffusion, the cells in the inner core of the islets still do not receive an adequate supply of oxygen and nutrients. These cells depend on intraislet capillary-mediated flow of blood. This limitation leads to lower oxygen and nutrient supply in the inner core of islets, which constitutes predominantly the insulin-secreting -cells, and ultimately leads to hypoxia and cell death. This phenomenon was elegantly demonstrated by Vasir et al. (1998), who stained islets cultured for 24 and 48 h with propidium iodide (red color) and calcein-AM (green color) to demonstrate the progressive loss of islet viability in the center of the islets. This loss of viability was associated with hypoxia in the inner core cells of islets. Culturing the same islets for an additional 24 h under hypoxic conditions exacerbated cell death (Vasir et al., 1998).
Islets try to revascularize themselves by secreting proangiogenesis molecular mediators such as vascular endothelial growth factor (VEGF) and its receptors (Vasir et al., 2000. 2001). Revascularization of islets post-transplantation occurs from the surrounding host tissue vasculature (Konstantinova and Lammert, 2004; Narang et al., 2004; Zhang et al., 2004). Secretion of VEGF and similar pro-angiogenic factors by the islets tends to promote this process. Achievement of rapid revascularization is expected to improve the viability and functioning of transplanted islets. This may be achieved by ex vivo VEGF gene delivery to islets (Narang et al., 2004) or by coencapsulating VEGF protein with islets during microencapsulation (Sigrist et al., 2003a). These approaches are discussed in section V.
C. Autoimmune Recurrence and Immune Rejection
Immunological challenges to islet survival, engraftment, and function post-transplantation are 2-fold: alloimmune destruction and autoimmune rejection. Although the former is common to all organ and tissue transplantation situations, type I diabetes offers additional challenges because it is autoimmune in origin. Diabetes is characterized by the presence of -cell-reactive autoantibodies and T-cells in the patient. Although the -cell lesion is mediated by -cell-specific autoreactive T-cells, the specific nature of effector T-cells remains elusive. The host has pre-existing antibodies and primed immune cells against -cell surface epitopes and insulin, which participate in graft destruction, in addition to the immune cells that infiltrate in response to nonself antigens (Jaeger et al., 2000). The host also reacts to nonself proteins originating from the transplanted tissue in the case of allo- and xenotransplantation. In terms of nonself antigens, immunological closeness of the graft to the host significantly influences the magnitude of an immune attack and determines overall graft survival outcome. Hence, success rate of transplantation is in the following order: autotransplantation > allotransplantation > xenotransplantation.
An understanding of the mechanisms underlying host immune responses in the context of islet transplantation is useful to the application of various immunosuppressive, immune-modulating, and immune-tolerizing approaches to improving transplantation outcome. The immunological bases of islet rejection overlap with transplant rejection situations involving heart, lung, liver, pancreas, and kidney transplantation such that the knowledge generated by research in these areas can be applied to islet transplantation. Although the overall basics of immunology and its application to the allograft situation are reviewed elsewhere (Rossini et al., 1999; Janeway et al., 2001), we will focus on the strategies that have been applied to islet transplantation and the underlying immune processes that are modulated.
When foreign tissue is transplanted, the host recognizes foreign antigens. This recognition follows the process of "antigen presentation" to the host immune cells, whereby peptide fragments of various surface, secreted, and shed proteins are brought in direct contact with the host immune cells attached to glycoproteins known as the major histocompatibility complex (MHC). The MHC can be class I or class II, depending on the cells that express these complexes. Whereas MHC class I molecules are expressed on all cells of the body, MHC class II molecules are expressed only on the surface of certain immune cells including macrophages and dendritic cells, which are known as professional antigen-presenting cells (APCs). Both class I and class II MHCs are able to present foreign peptides to host immune cells, but only class II MHC-bound peptides elicit an immune response because of the "costimulation" requirement of host T-cells for their full activation (Rossini et al., 1999).
Foreign antigen presentation to the host can be done by the host's own APCs or those of donor origin. The host APCs may be the mononuclear cells that infiltrate the graft and migrate away from the graft, or they may be the circulating APCs that encounter soluble donor antigens that have diffused away from the graft. Such soluble donor antigens are present predominantly in the case of xenografts. Donor APCs, on the other hand, are usually the dendritic cells, macrophages, and the circulating T-lymphocytes (called passenger leukocytes) that are transferred along with the graft. The antigens carried by the MHC II molecules on these APCs act as foreign antigens to the transplant recipient when these lymphocytes migrate to the lymph nodes of the host.
Donor antigen presentation to the host T-cells could be mediated via a direct or indirect pathway. The direct pathway involves antigen presentation by the donor APCs, whereas the indirect pathway involves host APCs, which pick up and process donor antigens for presentation as illustrated in Fig. 3. In the direct pathway of immune destruction, predominant in the case of allotransplantation, donor APCs migrate from the implantation site and present antigens to the host T-cells resulting in the development of CD4+ helper Th1 cells (Nicolls et al., 2001; Jiang et al., 2004). Th1 cells, in turn, produce a set of cytokines that favor expansion and activation of cytotoxic CD8+ T-cells. These are the primary effector cells mediating allogeneic cell damage. Xenotransplantation of islets, however, leads to the activation of the indirect pathway whereby antigens shed by the donor are taken up by host APCs and displayed on MHC II molecules (Watschinger, 1995; Game and Lechler, 2002; Jiang et al., 2004). In the case of xenotransplantation, humoral responses targeted mainly toward the surface -(1,3)-galactose moiety already exist. This moiety is present in animal tissue but has been evolutionarily lost in humans. The presence of these preformed antibodies is the main cause of hyperacute rejection observed with xenotransplantation (Cozzi et al., 2000; Ramsland et al., 2003). Although donor APCs are present in the transplanted xenogenic islets also, they are not able to activate the direct pathway because the MHCs displayed on donor APCs are not able to efficiently engage host T-cells (Chitilian et al., 1998; Vallee et al., 1998).
|
An understanding of these pathways is critical to the design of approaches for preventing immune destruction of transplanted islets. Although allogeneic islet destruction involves predominantly cytokine action and cytotoxic CD8+ T-cells, xenogenic tissue destruction proceeds via the antigen-antibody reaction pathway (Makhlouf et al., 2003). Thus, preventing leaching of cellular antigens is of critical significance to the biomaterial-based approaches to xenotransplantation. In contrast, preventing the migration of donor APCs from the transplant site could be adequate for allotransplantation. These criteria have significant implications in terms of the porosity and diffusibility requirements for encapsulated systems containing allogeneic or xenogenic islets, the latter being more stringent. Hence, whereas ultrafiltration (2–50 nm pore size) membranes are required for xenograft applications, microfiltration (0.1–1 µm pore size) membranes are adequate for allotransplantation of islets (Chaikof, 1999).
In both pathways, the molecular level mechanism of T-cell activation involves the following five steps (Fig. 3): 1) recognition of MHC and the bound peptide by the T-cell receptor (TCR) on the host T-lymphocyte, which, together with intimately associated CD45 transduces Signal 1, 2) transient CD40L (also called CD154) expression on the responding T-cell, 3) interaction of CD154 (CD40L) with the constitutively expressed CD40 on the APC (Coactivation), 4) up-regulation of costimulatory molecules B7-1 and B7-2 on the APC membranes, and 5) interaction of B7-1/2 with constitutively expressed CD28 or transiently expressed cytotoxic T-lymphocyte antigen-4 (CTLA-4) on the T-cell (Costimulation, Signal 2). In the absence of costimulation, the host T-lymphocytes are not activated by contact with foreign antigen and may, in fact, undergo apoptosis. Various immune tolerizing strategies attempt to block immune responses at one or more of these steps. These are discussed in section III.A.
D. Islet Mass and Site of Transplantation
The total number of islets present in an adult human pancreas is approximately 1 million, however, only about one half or fewer of these are successfully isolated. Thus, whereas transplantation of one intact pancreas is adequate to achieve glucose homeostasis in a diabetic recipient, islet transplantation requires the use of islets from two to four donor pancreases. Therefore, >10,000 IE/kg were transplanted in diabetic patients using the Edmonton protocol (Street et al., 2004a), whereas Gaber et al. (2004) used 11,000 to 15,000 IE/kg over three different infusions. The islet mass requirement for transplantation is reflected not only in the achievement and maintenance of normoglycemia in transplant recipients, but also in terms of long-term graft survival and function (Rickels et al., 2005). Often transplanted islets do not engraft well, leading to primary nonfunction. Primary nonfunction occurs because of nonspecific events that are not related to the classic immune rejection phenomena. It is caused by the poor quality of islet preparation, cytokine-mediated local inflammation and apoptosis, blood clotting, and hypoxia before revascularization of the islets (Bretzel, 2003). Islets further experience high metabolic demand in the recipient because of insulin resistance, diabetogenic and toxic immunosuppressive agents (glucocorticoids, cyclosporine A, and tacrolimus), and low transplanted islet mass. If and when inadequate numbers of islets are transplanted, the increased metabolic demand and persistent hyperglycemia may lead to graft destruction from islet apoptosis (Rossetti et al., 1990; Leahy et al., 1992).
The number of transplanted islets plays a critical role in short- and long-term islet function and metabolic normalization in the transplant recipient (Beattie and Hayek, 1993; Tobin et al., 1993). Various researchers have investigated the effect of number of islets transplanted on various aspects of islet function post-transplantation. For example, Finegood et al. (1992) evaluated the time required for normalization of fed-state plasma glucose levels during the 5 weeks after syngeneic transplantation of 500 to 3000 IE in streptozotocin-induced diabetic Wistar Furth rats by portal vein infusion. Islet mass had an inverse correlation to the time to glycemic normalization (Fig. 4A). Animals receiving 500 IE required approximately 5 weeks to achieve normoglycemia, whereas animals receiving 2000 to 3000 IE achieved normoglycemia within 2 weeks of transplantation. Bell et al. (1994) observed that the blood glucose levels were inversely proportional to the islet mass. Increasing islet masses improves both short-term and long-term glycemic normalization and leads to prolonged graft survival, due to reduced hyperglycemic stress to the islets (Fig. 4B). In the xenotransplantation of human islets in the subcapsular space of NOD-SCID mice, Rush et al. (2004) demonstrated improvement in insulin production upon transplantation of a higher number of islets. They concluded that 2000 IE/mouse are adequate for in vivo assessment of islet function.
|
-cells. Islets release insulin in a typical biphasic manner upon an increase in glucose concentration (Bratanova-Tochkova et al., 2002; Kennedy et al., 2002; Straub and Sharp, 2002). Persistently increased glucose concentrations, however, are toxic to the islets (Kaneto et al., 1999; Francini et al., 2001; Leibowitz et al., 2001; Biarnes et al., 2002; Maedler et al., 2002b). Thus, rapid and effective normalization of blood glucose after islet transplantation—from either sufficient mass of islets, secretion of the required amount of insulin, or exogenous insulin administration—is critical to the survival and function of islet graft. Thus, Biarnes et al. (2002) found that increased islet apoptosis and increased islet mass upon suboptimal syngeneic islet transplantation (100 IE/mouse) in streptozotocin-induced diabetic mice under a chronic (30 days) hyperglycemic state compared with mice in which normoglycemia was maintained by exogenous insulin administration. Exogenous insulin administration is a standard part of the protocol of islet transplantation, and the insulin amount required is slowly tapered off over a period of time after the islets are transplanted.
The site of transplantation of islets also influences graft performance (al-Abdullah et al., 1995; Mahmoud et al., 1998). The various sites that have been evaluated for islet transplantation include the liver (Contreras et al., 2004a), spleen (Weitgasser et al., 1996), abdominal cavity (in the omentum) (Kin et al., 2003), testes (Gores et al., 2003), and renal subcapsular space (Molano et al., 2003). The rationale for selecting different sites in experimental systems often depends on a host of factors. For example, abdominal implantation is preferred for micro- and macroencapsulated systems because of higher volume of the graft. Transplantation in immunoprivileged sites such as the testes and under the kidney capsule have been preferred to reduce immunological challenges to the engrafted tissue (Ksander and Streilein, 1994). In fact, Sertoli cells, which contribute to the immunoprivileged status of the testes, have been cotransplanted with islets under the kidney capsule in an attempt to prolong islet allograft survival without systemic immunosuppression (Selawry and Cameron, 1993; Kin et al., 2002).
Immune privilege at these sites has been correlated to several factors, including high levels of Fas ligand (FasL) expression on cells (Bellgrau et al., 1995; Griffith and Ferguson, 1997). FasL-expressing cells interact with Fas-expressing T-lymphocytes in the graft region and lead to apoptosis of the infiltrating cells by the natural Fas-FasL-mediated process. This knowledge has led to attempts to ectopically express FasL in the islets or at the graft site to prevent acute graft rejection. Thus, Lau et al. (1996) genetically engineered myoblasts from the donor to express FasL. Islets from mice were allotransplanted with syngeneic myoblasts under the kidney capsule of streptozotocin-induced diabetic mice, leading to long-term normoglycemia. Transplantation of myofibroblasts expressing FasL on the other kidney led to graft failure, indicating the need for local FasL expression (Lau et al., 1996). The expression of FasL on islets themselves, however, met with failure, probably because the expressing islets brought infiltrating T-lymphocytes in closer proximity, which led to islet destruction (Kang et al., 1997). This result was also observed when FasL transgenic mice were prepared with -cell promoter specific expression of FasL, such that it was expressed only on -cells (Chervonsky et al., 1997). These studies indicate that FasL expression in proximity to the graft, but not from the graft itself, can prevent immune-mediated destruction of transplanted islets.
In a study directly comparing islet transplantation under the kidney capsule versus that under the spleen capsule, Weitgasser et al. observed more prolonged normoglycemia when syngeneic islets were transplanted under the kidney capsule in streptozotocin-induced diabetic rats (Weitgasser et al., 1996). The most widely used site for islet transplantation, however, is the liver, which follows portal vein administration of islets as a suspension (Shapiro et al., 2000). Transplantation in the liver is the least invasive. It ensures that each islet receives an ample amount of blood supply, and insulin production and utilization follow the physiological route (Arbit, 2004).
Islets transplanted in the hepatic portal vein lodge themselves in the sinusoids of the liver. Although intraportal transplantation has been used in most clinical studies, certain factors lead to significant islet damage and trauma. The benefit of `bathing in blood' toward rapid diffusive transport of nutrients is offset by the adverse inflammatory reactions initiated by islets when they suddenly come in contact with blood in the portal vein. This instant blood-mediated inflammatory reaction of islets is characterized by the activation of coagulation and complement systems, islet infiltration of host leukocytes, and binding of host platelets (Bennet et al., 1999; Badet et al., 2002; Moberg et al., 2003). Islet damage also occurs because of nonspecific activation and dysfunction of intrahepatic host endothelial cells. These endothelial cells, in response to islets lodging in hepatic microcapillaries, up-regulate intercellular adhesion molecule (ICAM)-1 and P-selectin and produce nitric oxide (NO) and inflammatory cytokines such as TNF-, IL-1, and IFN-
(Xenos et al., 1994; Bottino et al., 1998; Contreras et al., 2004a). Resident islet macrophages, Kupffer cells of the liver, and liver sinusoidal endothelial cells have been implicated as primary mediators of inflammation-mediated loss of islets when transplanted in the liver (Barshes et al., 2005). Furthermore, portal islet transplantation leads to bleeding, portal venous thrombosis, and portal hypertension (Robertson, 2004). These complications are partly offset by portal blood pressure monitoring and the use of anticoagulants during the procedure (Robertson, 2004).
Islet mass required for glycemic normalization is also influenced by the site of transplantation. For example, in the case of canine islet autografts, Kaufman et al. (1990) found that whereas the threshold number of islets required to achieve normoglycemia in the liver and spleen were similar (4500 IE/kg), this number failed to ameliorate hyperglycemia when transplanted in the renal subcapsular space. Whereas both the islet mass and the site of islet transplantation in the host play an important role in graft survival and function, their definitive optimization is difficult because of a host of factors that influence ultimate graft survival. Although portal vein transplantation of islets at a count of >10,000 IE/kg recipient is contemporarily practiced, these variables may be optimized for further improvement with various interventions being explored to improve islet graft survival and function.
III. Biological Strategies for Improving the Success of Islet Transplantation
The two major impediments to the clinical success of islet transplantation are the immune destruction of transplanted islets and the limited supply of islet tissue. Several approaches have been proposed and tested to address these problems. Among the biological strategies used to overcome immune rejection are the use of novel immunosuppressive agents and regimens, and donor-specific induction of immune tolerance in the host. To address the foreseeable dilemma of unmet tissue demand, xenotransplantation, in vitro stem cell differentiation, and regeneration therapy of -cells have been explored.
A. Prevention of Immune Destruction of Transplanted Islets
Islet graft rejection process can be divided into three categories depending on the etiology, severity, and the timing involved— hyperacute rejection, acute rejection, and chronic rejection (Rossini et al., 1999). Hyperacute rejection is the immediate rejection process that proceeds within hours and depends on preformed and primed antibodies within the host against the graft. This process is observed predominantly with xenotransplantation wherein preformed antibodies exist against the -(1,3)-galactosyl residues present on endothelial cells of lower animals. The humoral response leads to complement fixation and thrombosis within minutes to hours of engraftment and finally to transplant failure (Groth et al., 1994). The acute rejection process depends on self- and nonself-recognition and is predominant with allotransplantation. This rejection process is characterized by rapid infiltration of immune cells, followed by T-cell responses. Immunosuppressive drugs inhibit the acute rejection process. The chronic rejection process, however, proceeds even in the presence of immunosuppression and is characterized by fibrosis and distortion of the architecture of transplanted tissue, leading to graft failure (Orloff et al., 1995). Various mechanisms have been proposed to account for the chronic rejection process. These include the wound healing process, delayed type hypersensitivity reaction, antibody-mediated humoral immunity, and endothelial cell damage (Rossini et al., 1999).
Various approaches have been attempted to target these different rejection processes. Hyperacute rejection of xenografts can be obviated with use of animals knocked out for specific genes, e.g., -(1,3)-galactose, or a component for complement fixation (White and Yannoutsos, 1996) or use of complement inhibitors, complement depletion, and plasmapheresis of the recipient to remove natural antibodies (Rossini et al., 1999). The acute rejection process has been addressed predominantly through the use of immunosuppressant drugs, whereas chronic rejection is being addressed through the use of tolerogenic strategies (Womer et al., 2001a). The combination of immunosuppression and tolerance approaches is now being proposed for improved clinical outcomes in both islet and solid organ transplantation (Adams et al., 2001).
1. Immunosuppression.
Generalized immunosuppression of the transplant recipient is the standard protocol today to prevent graft rejection by the host immune system. The first generation drugs that were applied to this end, include azathioprine, glucocorticoids, and antilymphocyte serum (ALS). Azathioprine is a calcineurin inhibitor. Calcineurin is a cytosolic calcium-dependent serine/threonine phosphatase protein that acts to remove phosphates from cytoplasmic regulatory proteins, which then penetrate the nucleus and act as transcription factors. Inhibition of calcineurin activity leads to inhibition of production of various cytokines including IL-2 and other gene products essential for T-cell activation. Although they are highly effective, these drugs have significant toxicity. Nephrotoxicity is prevalent in three-fourths of all patients (Burke et al., 1994). Additional side effects include hypertension, hepatotoxicity, neurotoxicity, hirsutism, gingival hyperplasia, and gastrointestinal toxicity. Other agents widely used for immunosuppression include glucocorticoids. These act through inhibiting T-cell proliferation and expression of genes encoding specific cytokines. They further block IL-2 production and also act by nonspecific inflammatory and antiadhesion effects. Their long-term administration, however, is associated with severe toxic effects including ulcers, hyperglycemia, osteoporosis, and increased risk of infection and neoplasms (Corbett et al., 1993; Diasio and LoBuglio, 1996; Dantal et al., 1998; Newstead, 1998). Antilymphocyte serum, e.g., polyclonal antithymocyte globulins to deplete T-cells in the host, is also widely used for immunosuppression during organ transplantation (Beiras-Fernandez et al., 2003).
Although some second-generation drugs are still used, those with higher potency and larger therapeutic window have been added to the drug cocktail. These include cyclosporine and tacrolimus (FK506) (Rossini et al., 1999). Cyclosporine acts on T-lymphocytes by forming a heterodimeric complex with cytoplasmic receptor protein, cyclophilin. Tacrolimus, on the other hand, binds to a cytosolic protein called FK 506-binding protein (Diasio and LoBuglio, 1996). However, many of these agents, including tacrolimus, cyclosporine, and steroids, are diabetogenic and toxic to the islets (Drachenberg et al., 1999). The use of immunosuppressive agents that do not challenge the islet graft is thus warranted. Although newer agents are constantly being developed, e.g., FTY720 (Fu et al., 2002) and lisofylline (Yang et al., 2004), improvements have also been reported with novel combinations of existing agents.
The islet transplant center at the University of Alberta in Edmonton, AL, Canada, has reported a high success rate of islet allotransplantation by sequential islet transplantation 2 to 10 weeks apart using two or more pancreases to achieve adequate mass of engrafted islets and by using a glucocorticoid-free immunosuppressive regimen that includes IL-2 receptor antibody (daclizumab), sirolimus (rapamycin), and low-dose tacrolimus (Shapiro et al., 2000). Daclizumab is given intravenously right after transplantation and then is discontinued. Sirolimus and tacrolimus must be taken for life. Daclizumab does not adversely affect islet function or glucose metabolism (Bretzel, 2003). The Edmonton group reported insulin independence in all seven patients with transplantation of an islet mass of 11,500 IE (basal diameter of 150 µm) per kg b.wt. In a follow-up study, this group reported an 80% success rate in terms of insulin independence for 1 year, which was maintained by 12 of 15 patients transplanted with 9000 IE/kg b.wt. (Ryan et al., 2001, 2002). Based on these encouraging results, a multicenter clinical trial to coordinate the implementation of the Edmonton protocol, called the Immune Tolerance Network, was initiated by the National Institutes of Health together with Juvenile Diabetes Foundation International with seven centers in the United States and Canada, and three in Europe (Bluestone and Matthews, 2000; Bluestone et al., 2000). The Immune Tolerance Network sponsors investigator-initiated research in targeted prevention of immunemediated transplant rejection by blocking immune signals at three different levels: T-cell recognition of antigen/MHC complex on APCs, costimulation to augment T-cell proliferative response to antigenic stimuli, and targeting clonal activation/deletion. One of these trials reported an insulin independence rate of 90% at the end of 1 year and long-term graft function in all 31 of 31 patients receiving transplants (Ricordi et al., 2005).
2. Immune Modulation and Tolerance.
An inspiration toward the possibility of avoiding immunosuppression of the recipient without accompanying graft loss came from early observations of selective graft acceptance of twin animals that share common placental circulation during gestation. Graft tolerance in this case was ascribed to the exposure of neonatal animals to foreign antigens (Billingham et al., 1953). Such a selective immunological acceptance of a "foreign" graft by the immune system is known as immune tolerance. Tolerance is defined as the specific immune nonresponsiveness to an immunogenic stimulus (Samstein and Platt, 2001). There are some underlying assumptions to this definition of tolerance. It presupposes that the recipient is immunocompetent and that the immune response to the desired transplant is only qualitatively different from an immune response to other foreign tissue or pathogens. These assumptions, however, may not hold true in many cases because of a more profound suppression of the immune system with the strategies used and the inability to actually assess third-party graft rejection. The clinical and experimental criterion mostly used for the success of an immune tolerance intervention is the prolonged survival of the graft without immunosuppression, whereas stress on histological evidence of the absence of chronic rejection is now increasing.
Immune tolerance to an allogeneic or xenogenic islet transplant can be achieved at various stages in immune system development. These approaches target the graft, the graft donor, or the host. Based on the mechanistic point of interference, they may be classified as modulation of transplant immunogenicity, removal of passenger leukocytes, or induction of transplant tolerance. These are summarized in Table 2 and discussed below.
|
a. Antibody pretreatment for xenografts.
Islet graft rejection is a predominantly T-lymphocyte-mediated process that occurs by several postulated mechanisms, e.g., provision of an adherence signal to T-cells by binding to the graft, antibody-dependent cell-mediated cytotoxicity, formation of immune complexes that physically block the vasculature and impair graft function, and complement fixation (Rossini et al., 1999). Therefore, the use of anti-T-cell antibodies should minimize islet graft rejection. Antibodies against the graft, however, can also have a protective role by masking the donor MHC class I antigens. Coating with an antibody that does not elicit a host immune response has been used to protect islet grafts against immune destruction. This strategy was used by Faustman and Coe (1991) for human islet xenotransplantation by precoating the donor antigens in islet tissue with the variable region of an antibody against donor MHC I molecules. They observed graft survival and histological improvement beyond 200 days without the need for immunosuppression.
b. Removal of passenger leukocytes for allografts.
In the direct antigen presentation pathway, donor APCs migrate to the host lymph nodes to present donor antigen. Therefore, this pathway can be blocked by removing donor APCs from the graft before transplantation. Faustman et al. (1981) applied this strategy to prevent islet allotransplant rejection by pretreating the graft with antisera and complement to remove donor passenger leukocytes. In another strategy, islets were cultured in vitro at a reduced temperature (24°C) for 7 days before transplantation, a process that apparently led to the removal of passenger leukocytes. This approach, in addition to a single injection of ALS, achieved islet allograft survival for >3 months without immunosuppression (Lacy et al., 1979; Chervonsky et al., 1997).
c. Cytokine modulation.
Cytokines may play either a destructive or an immunomodulatory role in islet graft rejection. Cytokines that contribute to graft destruction directly or by activating effector cells include IL-1, TNF-, and IFN-; whereas the cytokines that may impair graft rejection include IL-4, IL-10, and TGF-. Some cytokines are classified as Th1 or Th2-based on the T-helper (CD4+) lymphocytes that produce them. Th1 cytokines include IL-2, IFN-, TNF-, and IL-12; Th2 cytokines include IL-4, IL-5, and IL-10. Th1 cytokines activate both T-cells and macrophages and promote cellular immune responses that serve as terminal effector mechanisms. Th2 cytokines produce reactions that favor humoral, IgE-mediated allergic, and mucosal immune reactions (Antin and Ferrara, 1992; Mosmann and Sad, 1996; Nickerson et al., 1997).
Expression of immunoregulatory molecules from the islet grafts themselves by various gene transfer approaches during ex vivo islet culture is an attractive option for preventing islet graft rejection. For example, Gallichan et al. (1998) expressed IL-4 in islets using lentiviral vector for stable transfection. They observed absence of inflammatory infiltrates in grafts and, upon transplantation in diabetes prone mice, protection of animals from autoimmune insulitis and islet graft destruction. This was consistent with the observation of switching of islet-antigen-specific T-cell responses toward a Th2 phenotype. However, autoimmune disease recurrence was not prevented by IL-4 gene transfer to islets before transplantation into diabetic NOD mice using transiently expressing adenoviral (Adv) vectors despite a significant level of transgene expression (Smith et al., 1997).
Transfection of islets with Adv encoding interleukin-1 receptor antagonist prevented IL-1-mediated islet destruction and loss of islet function (Giannoukakis et al., 1999b). IL-1 antagonism is beneficial not only in the prevention of graft destruction, but also in the prevention of autoimmune insulitis and in the pathogenesis of diabetes. Other cytokines that have been evaluated toward this end include IL-10 (Benhamou et al., 1996) and TGF- (Ise et al., 2004). TGF- was reported to mediate the effects of anti-CD3 antibodies in NOD mice in abrogating autoimmunity (Belghith et al., 2003). TGF- was also implicated in the beneficial effect of mitomycin C on islet xenograft survival in a rat-to-mouse model (Ise et al., 2004) and in the protective effect mediated by Sertoli cells in a mouse allotransplantation model (Suarez-Pinzon et al., 2000). In the latter study, islets were transplanted into the left renal capsule of diabetic NOD mice whereas Sertoli cells were transplanted under the right renal capsule. Improvement in the survival and function of islets in Sertoli cell-transplanted mice were correlated to elevated plasma levels of TGF- and its production by Sertoli cells. Following this lead, Suarez-Pinzon et al. investigated whether Adv-mediated ex vivo transfection of islets with TGF- improved the outcome of islet transplantation. NOD mouse islets were transfected with porcine latent TGF-1 using Adv-TGF-1 and Adv vector alone. TGF-1 overexpression from the islets resulted in longer normoglycemia (median period of 22 days versus 7 days for control), reduced CD45+ T-cell infiltration of the graft, and reduced apoptosis of transplanted -cells (Suarez-Pinzon et al., 2002).
TGF- is postulated to act by generating CD4+CD25+ regulatory T-cells (Tregs) from CD4+CD25-). Tregs are potent suppressors of innate inflammatory responses and have been shown to enhance syngeneic islet transplant survival. TGF-2 was shown to induce Foxp3 expression in CD4+CD25- T-cells resulting in Foxp3+ cells that behave like conventional Tregs (Fu et al., 2004). A recent study has further shown that systemic TGF1 gene therapy by intravenous injection of Adv-TGF1 induces the production of Foxp3+ cells that restores self-tolerance by inhibiting autoimmune-mediated destruction of islets in the pancreas of NOD mice (Luo et al., 2005). These authors also transplanted 500 syngeneic islets under the kidney capsule of these mice 7 to 14 days after Adv-TGF1 injection. Islet graft survival time was prolonged (50 days in Adv-TGF1 injected mice versus 17 days in Adv-control injected mice) and was associated with peri-islet mononuclear cell infiltrate staining positive for CD4, CD25, and Foxp3. These studies demonstrate a cytoprotective role of TGF- that could be used for both reducing autoimmunity and inducing transplant tolerance.
Free radical (NO· and 
)-induced -cell death is initiated by macrophage secretion of cytokines IL-1 and TNF-. The prominent role of TNF- in stimulating the immune system indicates that antagonism of TNF- receptor binding may protect islet grafts from cytokine-mediated destruction. TNF- expression is up-regulated in inflamed islets during the development of type I diabetes (Held et al., 1990), soluble TNF receptor administration blocks TNF- mediated dysfunction (Farney et al., 1993), and transgenic mice expressing soluble type 1 TNF receptors secreted from -cells escape insulitis and diabetes (Hunger et al., 1997). Furthermore, TNF- injection to NOD mice led to an earlier onset of disease, whereas administration of anti-TNF monoclonal antibody resulted in complete prevention of diabetes development (Yang et al., 1994a). These observations indicate that islet treatment to antagonize TNF- might improve the outcome of islet transplantation. Thus, Dobson et al. (2000) investigated the utility Adv-mediated transfection of an inhibitor of TNF (TNFi), whereas Machen et al. (2004) explored the use of soluble type 1 TNF receptor-Ig fusion protein. Dobson et al. (2000) transfected human pancreatic islets with Adv-producing TNFi and transplanted 2000 IE under the kidney capsules of NOD-SCID mice. Fifteen days after transplantation, the mice were injected with human peripheral blood leukocytes (huPBL) or buffer control. TNFi-transfected islets exhibited improved graft survival and function. The authors observed that TNFi effectively limited damage to -cells by huPBL, although leukocyte infiltration was not affected. In addition, no difference between TNFi-treated and -untreated groups was observed in mice not injected with huPBL (Dobson et al., 2000). TNFR-Ig transfection, on the other hand, was shown to reduce cytokine-induced apoptotic human islet death in vitro and prolongation of normoglycemia a
