Abstract
In the past years, in the field of β-cell replacement for diabetes therapy, the easy availability of bone marrow (BM) and the widely consolidated clinical experience in the field of hematology have contributed to the development of strategy to achieve donor-specific transplantation tolerance. Recently, the potential role of BM in diabetes therapy has been reassessed from a different point of view. Diverse groups investigated the contribution of BM cells to β-cell replacement as direct differentiation into insulin-producing cells. More importantly, while direct differentiation is highly unlikely, a wide array of experimental evidences indicates that cells of BM origin are capable of facilitating the survival or the endogenous regeneration of β-cells through an as yet well-defined regeneration process. These new experimental in vitro and in vivo data will expand in the near future the clinical trials involving BM or BM-derived cells to cure both type 1 and type 2 diabetes in humans. In this review we recapitulate the history of use of BM in diabetes therapy and we provide clinically relevant actual information about the participation of BM and BM-derived stem cells in islet cell regeneration processes. Furthermore, new aspects such as employing BM as “feeder tissue” for pancreatic islets and new clinical use of BM in diabetes therapy are discussed.
Introduction
The use of bone marrow (BM)derived cells to achieve donor-specific transplantation tolerance has been extensively addressed in animal models. It was first reported by Billingham and associates in 1953 (10) in neonatal mice. Subsequently, chimerism with donor-specific transplantation tolerance was achieved in adult animal models by combining BM transplant (BMT) and preconditioning the host with different regimens, which have included, among others, total body irradiation (50), total lymphoid irradiation (114), and antilymphocyte globulin (17,85).
The first studies on the possibility of combining BMT with insulin-producing cell transplant date back to the mid-1970s (62,97). The goal of BMT in these studies was to allow the survival of allogeneic insulin-secreting tissue by creating allogeneic chimeras through lethal irradiation and injection of allogeneic BM in mice or rats made diabetic with streptozotocin (STZ) (13).
The availability of murine models of spontaneous diabetes [“BB” rat (89) and non-obese diabetic mouse (NOD) (79)] permitted studying BM in relation with pancreatic islets in the autoimmune context in the early 1980s. The first studies devoted to define etiologic factors of β-cell destruction in “BB” rats clearly demonstrated that BMT has a complex key role in modulating autoimmune response against pancreatic islet (). An unexpected finding, which may be of paramount significance, was that the incidence of spontaneous diabetes is strikingly less in newborn “BB” rats (less than 24 h old) that received MHC-compatible BMT (50 × 106 cells IV) from normal donors (Wistar Furth rat) (4,87,88). Similarly, Ikehara et al. (49) reported that MHC-incompatible (H-2d in H2g) allogeneic BMT (2 × 107 cells IV) from young BALB/c nu/nu mice can prevent insulitis and diabetes in NOD. The same group showed that MHC-incompatible allogeneic BMT combined with pancreatic tissue from newborn BALB/c nu/nu grafted under the renal capsule can be used to cure overt diabetes in NOD (134). The relevance of BM in determining the autoimmune response was subsequently strongly supported by the evidence that F1 hybrids produced by outcross of NOD mice with diabetes-resistant strains are also diabetes resistant (i.e., [NODxNON]F1, [NODxB10]F1, and [NODxCBA]F1 mice), but this resistance is abrogated if F1 hybrids are lethally irradiated and then hematopoietically reconstituted with NOD BM (48,63,110,130). A report of clinical observation describing that type 1 diabetes occurred in a woman 4 years after BMT from her HLA-identical brother with type 1 diabetes definitively consecrated BM as a central element in the etiopathogenesis of type 1 diabetes (64).
All together these results provided evidence that BMT plus insulin-producing cell transplantation could be used to treat type 1 diabetes mellitus. The rational of the concomitant MHC-incompatible allogeneic BMT was explained as follows: (i) allogeneic BM induces immunological tolerance to the donor-type major histocompatibility complex-encoded determinants and permits the host to accept subsequent allogeneic insulin-producing grafts from the BM donor and (ii) BM replaces, through the preparative lethal whole-body irradiation, abnormal stem and immune autoreactive cells with stem and immune cells from a normal donor, thus eliminating the danger of a recurrence of diabetes. In the 1990s, on the basis of these concepts, different strategies combining BMT with different immunosuppressive agents were proposed to prolong pancreatic allogeneic islet graft survival or to obtain donor-specific allograft acceptance in different preclinical murine models of diabetes: peritransplant antilymphocyte serum plus posttransplant infusion of donor BM (80); single dose of antilymphocyte serum combined with inoculation of allogeneic BM into the thymus (101); transplantation of ultraviolet B-irradiated BM cells into lethally gamma-irradiated allogeneic recipients to obtain stable chimerism (55).
The clinical use of donor BMT to prolong the survival of organ allografts was first attempted in kidney transplant recipients (7,82,85) and subsequently in liver (106) and heart (58) allograft recipients. All these clinical studies were based on the premise that “space” needs to be created by preconditioning of the host with cytoreductive or cytoablative therapy, thus allowing for engraftment of donor BM with subsequent establishment and perpetuation of chimerism. Subsequently, the demonstration that donor chimerism is a naturally occurring phenomenon after transplantation of solid organs (109,118,119) and that a stable mixed hematopoietic chimerism is obtainable in humans with BMT after minimal nonmyeloablative conditioning regimens (113) provided the basis for the initiation of studies, in which unmodified MHC-mismatched donor BM cells were infused in recipients at the time of organ placement without any preconditioning of that host or deviation from the routine drugs and their therapeutic doses that were required for maintenance of adequate immunosuppression (36,102,103). On the basis of the hypothesis that donor cell chimerism, in a preclinical model of islet transplantation, is needed to attain a tolerant state and prevent autoimmune response, different strategies to combined BMT after immune-suppressive nonablative regimens with islet transplant were proposed (3,8,12,18,29,39,40,45,47, 68–71,76,104,105,111,131,136,139). In the 1990s, clinical trials of donor BM infusion were initiated in simultaneous pancreas and kidney and pancreas transplantation at centers in Pittsburgh and Miami (27,28). Although donor BM infusion in these cohorts of patients has been associated with persistent donor microchimerism, its clinical impact remained uncertain. Despite this, infusion of donor BM-derived cells to create a chimeric state continues to be tested in clinical protocols intended to induce specific immunologic tolerance.
New Perspective: BM as an Extrapancreatic Hideout for the Elusive Pancreatic Stem Cell
Diverse groups investigated the contribution of BM cells to β-cell replacement as direct differentiation into insulin-producing cells. There are several different types of adult stem cells found in BM. The most extensively characterized stem cell of mesodermal origin is the hematopoietic stem cell (HSC). In addition to HSC, mesenchymal stem cell (MSC) is also able to be isolated from BM and contributes to the formation of mesenchymal tissues, such as bone, cartilage, muscle, ligament, tendon, adipose tissue, and stroma. Moreover, a variety of progenitors with specific differentiation potential have been described: multipotent adult progenitor cells (MAPCs) (54), marrow-isolated adult multilineage inducible (MIAMI) cells (30), and tissue committed stem cells (TCSCs). The relationship between these stem cells remains uncertain.
In vitro multiple strategies involving exposure to various growth factors combinations under specific culture conditions, often augmented by genetic manipulation, have been explored to differentiate adult stem cells found in BM in insulin-producing cells. It has been seen in different models (mice, rats, and humans) that cells derived from BM are capable of differentiating into insulin-producing cells. At this time there is a consensus that MSC (19,20,22,24,44,61,73,75,86,120,121,125,133) but not HSC can be induced to exhibit pancreatic properties, including low-level insulin production and responsiveness to supraphysiological glucose levels. However, when transplanted these cells do not normalize blood glucose or respond as efficiently as β-cells to a glucose challenge (1,94,124,132,137). Attempts to transdifferentiate BM stem cells into insulin-producing cells in vivo have produced conflicting results. In 2003 Ianus and colleagues (46) reported BM as an extrapancreatic source of pancreatic β-cells that may play a role in β-cell turnover and possibly the adaptation of islet mass in response to physiological and environmental stimuli. In this study, BM cells that selectively express the enhanced green fluorescent protein (EGFP) if the insulin gene is actively transcribed were transplanted into lethally irradiated recipient mice and gave rise to EGFP-positive insulin-producing cells in pancreatic islets. After 4–6 weeks, recipient mice revealed both Y chromosome and GFP positivity in 1.7–3% of cells in pancreatic islets. The BM derived β-cells expressed insulin, Glut2, Pdx1, and Pax6 and secreted insulin in response to a glucose challenge.
Other teams did not confirm these data and following studies have reported much lower frequencies (0.004%) or even a total absence of regeneration phenomena in β-cells of BM origin. Using GFP transgenic mice as donors, Choi et al. evaluated the distribution of HSC in the pancreas after BMT and found that none of the GFP+ cells localized in islets expressed insulin. Even after pancreatic injury, obtained with low doses of STZ, no GFP+ cells expressing insulin were found in the islets or around the ducts of the pancreas (23). Lechner et al. in other pancreatic injury models confirmed the same results: mice were transplanted with GFP+ sex-mismatched BM after partial pancreatectomy and STZ administration. BM engrafted successfully but 3 months after transplantation only a few β-cells (2 out of 100,000 screened) were of BM origin (65). Further studies of BMT in normal and diabetic mice were performed using as donor transgenic mouse expressing GFP under the control of insulin promoter or β-Gal. Even in this model all the BM-derived cells engrafted in pancreas expressed hematopoietic markers and not insulin or Pdx1 (60).
New Perspective: BM as “Feeder Tissue” for Pancreatic Islets
Recently, the potential role of BM in β-cell regeneration has been reassessed from a different point of view. While direct differentiation is highly unlikely, a wide array of experimental evidences indicates that cells of BM origin are capable of facilitating the survival or the endogenous regeneration of β-cells through an as yet well-defined regeneration process. In fact, transplants of BM cells were reported to improve or reverse the diabetic state in a number of experimental models, including low-dose STZ (6,41,43,66), high-dose STZ (51), E2f1/E2f2 mutant mice (67), and the KK-Ay model of type 2 diabetes (126). Moreover, in vitro coculture with BM demonstrated the ability to repair damaged human islets, resulting in functional insulin release from islets in long-term culture (77).
This suggest that the role of BM in the cure of diabetes could be an indirect role in protecting remaining β-cells from further destruction at onset of disease or in the stimulation of endogenous β-cell replacement from tissue-resident stem cells or preexisting β-cells. Hess et al. reported that transplantation of BM-derived cells in STZ-treated mice resulted in endogenous pancreatic regeneration (43). They observed that the majority of BM transplanted cells found in pancreas were localized to ductal and islet structures and that none expressed insulin. In the same direction, Hasegawa et al. sustained that BMT induces regeneration of recipient pancreatic β-cells: they transplanted GFP+ BM in diabetic irradiated mice and observed an improvement in glycemia and partial recovery of pancreatic islet number and size without evidence of differentiation (41). More recently, the confirmation of the role of BM-derived cells in supporting pancreatic islet repair comes from Gao et al.: transplantation of GFP+ BM in diabetic mice resulted in reduction of hyperglycemia, increase in number of islets, increase in insulin+/GFP- proliferating cells, increase in Ngn3+ cells, increase in insulin and glucagon double positive cells and insulin-/Pdx1+ cells (38). It should be underlined that not in all experimental models tested was BMT able to promote pancreatic islet repair. After BMT into adult Akita diabetic mice (a pure model of β-cell apoptosis in the absence of any inflammatory activation), despite full hematological chimerism being achieved, no improvement was found in glucose levels, nor was there evidence for in vivo β-cell regeneration (2). This suggests that the type of diabetes and mechanism of injury may be important determinants of the potential for regeneration of the endocrine pancreas by BM.
Which Is the Cellular Component of BM That Plays the Main Role as “Feeder” Cells?
At this time, three different candidates have been proposed: HSC, endothelial progenitor cells (EPC), and MSC. C-kit+ and/or CD45+ HSC were suggested by Hess and Hasegawa (41,43). In fact, they reported respectively that (i) BM cells expressing c-kit marker could lead, when transplanted, to reduction in glycemia, whereas c-kit- cells had no effect and that (ii) BM-derived cells, detected around the islets, were CD45+. Alternatively, a possible explanation of the indirect role of BM on endogenous islet cells is the involvement of endothelial cells in tissue remodeling and stimulus to proliferation. This was supported by the finding that BM-derived cells in the host pancreas differentiated into endothelial cells and their presence was accompanied by a proliferation of recipient pancreatic cells that resulted in insulin production (83). For example, Mathews and colleagues reported that after BMT the frequency of endothelial cells (of both donor and recipient origin) increased in the animals in which β-cell injury had been induced (81). In the same direction, augmentation of EPC in the peripheral circulation after intrahepatic islet syngeneic infusion in mouse resulted in higher islet vascular density and engraftment (26). Finally, the role of MSC was strongly supported in more recent years by many experimental models (11,21,32,56,78,115,128). Human MSC delivered via intracardiac infusion on days 10 and 17 in NOD/scid mice treated with daily low doses of STZ on days 1–4 lowered blood glucose levels and preserved pancreatic islets and β-cells producing insulin without evidence of direct differentiation (66). Similarly, murine MSC administration resulted in β-pancreatic islets regeneration and prevented renal damage in diabetic mice (35).
Does BM Play a Role as “Feeder Tissue” by Modulating/Enhancing Vascularization?
One of the common features that emerged from in vivo studies using BM or BM-derived cells as “feeder tissue” is the suggestion of neovasculogenesis as potential mechanism for supporting endogenous (81) or exogenous islet cells (56,83,107). In fact HSC, MSC, and EPC derived from BM have all emerged as useful promoters of neovascularization, and two predominant mechanisms by which these cells contribute to neovascularization have been identified: (i) vessel formation by differentiation into mature endothelial cells (i.e., EPC) (5); (ii) release of proangiogenic factors such as hepatocyte growth factor (HGF) and vascular endothelial growth factors A (VEGF A) (31,98). In an interdependent physical and functional relationship with β-cells, endothelial cells are involved not only in the delivery of oxygen and nutrients to endocrine cells but also, via the vascular basement membrane, provide a niche for β-cells, induce insulin gene expression during islet development, affect adult β-cell function, promote β-cell proliferation, and produce a number of vasoactive, angiogenic substances and growth factors (93,135). Consequently, even if the main mechanism by which BM acts as feeder tissue for islet remains to be elucidated, it is reasonable to assume that the induced neovascularization plays a major role. This does not exclude that other mechanisms may be involved. The MSC-derived laminins binding to α6β1-integrin expressed on pancreatic β-cells could support insulin expression and β-cell proliferation (93). Similarly, the MSC-derived collagen IV binding to α1β1-integrin could stimulate insulin secretion (59). Moreover, the release from BM-derived cells of trophic molecules like HGF, IL-6, insulin-like growth factor binding proteins 4, VEGF A, and TGF-β can directly sustain β-cell survival and function (21,25,51,52,98). Finally, MSC displaying immunomodulatory properties, including anti-inflammatory activity (90,92,138), could improve β-cell survival and function antagonizing pathogenic inflammation and promoting tissue healing.
Does a Cross-Talk Between BM and Pancreatic Islets Exist in Physiologic and/or Pathologic Condition?
In mouse recipients of a GFP+ BMT, we recently conclusively demonstrated that pancreatic MSC originate from BM (117). So, even if we cannot exclude the existence of other sources of MSC, these evidences strongly suggest the existence of a cross-talk between BM and pancreas, evidences also supported by the fact that the pancreas is a preferential site of GFP+ cells localization after GFP+ BMT (117).
In humans, we recently found that part (about 6%) of the CD73+ MSC population in allogeneic pancreas removed after transplantation expressed recipient HLA, suggesting that an extrapancreatic source “refilled” the pancreas (117). A study by Butler and colleagues (15) analyzed 31 human pancreata obtained at autopsy from BMT recipients who had received their transplant from a donor of the opposite sex. Survival of BM recipients ranged from 13 to 1,235 days after transplant. The authors described scattered donor BM-derived cells present throughout the pancreas, with a density that increased with duration of survival from BMT, reaching a plateau at ~500 days. These cells were present in vascular endothelium, exocrine pancreatic ducts, and interlobular connective tissue.
Regarding the mechanisms involved in MSC recruitment, we previously reported that human pancreatic islets can attract BM MSC in vitro, and this attraction is principally mediated by two chemokines: CX3CL1 and CXCL12 (116). Similarly, Lin and colleagues reported, using a coculture microfluidic chip, that BM MSC have the ability to migrate to pancreatic islets (74). Currently, we are unable to determine whether MSC migration from the BM occurs in the absence of pancreatic injury, or if MSC constitute a renewable pool of cells that migrate early in development to create a pancreatic stem cell niche that replaces dying cells. It is possible that the resident MSC are derived from the vessels within the pancreas as self-renewing “vascular” stem cells resident pool. On the other hand, the BM could reinforce the pancreatic MSC population only in stress situations, as in the radiation-based BMT setting (33) or in the organ damage in the inflammatory events, and do not participate in the MSC turnover during healthy status.
Even more complex is trying to understand whether BM–pancreas cross-talk plays a role in the pathogenesis of diabetes. Diabetes, both type 1 and type 2, is associated to morphologic and functional alterations of the BM and BM-derived stem cells (42,72,96,99,112,123,127). Whether these alterations contribute to β-cell dysfunction, apoptosis, or lack of regeneration remains to be elucidate. More generally, whether these alterations are either preexisting or a consequence of diabetes remains to be elucidated and few papers addressed these issues (53,95).
New Perspective: BM for Diabetes Therapy in the Clinic
Experimental in vitro and in vivo data associated to the easy availability of BM and to widely consolidated clinical experience in the field of hematology has made possible designing clinical trials on humans involving BM to cure both type 1 and type 2 diabetes.
HSC and Islets Concomitant Allotransplantation in Type 1 Diabetic Patients
In experimental islet transplant models, infusion of high doses of donor HSC using minimal or nonablative conditioning has been shown to induce mixed donor/recipient chimerism and graft tolerance with reduction or discontinuation of immunosuppression (122). On these bases a clinical pilot trial of donor HSC transplant in islet-receiving patients is ongoing at Diabetes Research Institute in Miami (ClinicalTrials.gov Identifier: NCT00315614). The study's objective is to induce a stable donor/host mixed hematopoietic microchimerism, allowing an immunological tolerance condition and hence the suspension of immunosuppression therapy. Patients receive islet transplantation together with high doses of donor-purified HSC under a modified “Edmonton-like” immunosuppression protocol without ablative conditioning. Preliminary results (84) indicate that the combined HSC–islets allotransplantation did not lead to stable chimerism and graft tolerance; in fact, after immunosuppression weaning, islet dysfunction and graft failure were observed. Adjustment of transplantation conditions (higher dose of HSC, more intense immune-suppressive conditioning, engraftment improvement, and alloreactivity reduction) could maybe lead to a more stable mixed chimerism and therefore a more tolerogenic status.
HSC Autologous Transplantation in Type 1 Diabetic Patients
The potential use of BMT to alter the course of type 1 diabetes disease process was first proposed in animal studies in 1985 using allogeneic BM (49). Human studies suggested that the use of allogeneic BMT for treatment of malignancy resulted in reversal of several autoimmune diseases including type 1 diabetes (91). The first attempt to determine the safety and efficacy of a nonmyeloablative intense immune suppression with autologous HSC transplantation in type 1 diabetes patients at onset comes from a recent work by Voltarelli and colleagues (129) (ClinicalTrials.gov Identifier: NCT003 15133). In this study 23 new onset subjects (aged 13–31 years) within 6 weeks from type 1 diabetes diagnosis received autologous HSC cell transplantation. Patients underwent autologous stem cell mobilization with cyclophosphamide and daily granulocyte colony stimulating factor followed by leukapheresis and cryopreservation of HSC. Then they received an intensive immune-suppressive conditioning therapy (rabbit antithymocyte globlulin and cyclophosphamide) with reinfusion of the previously mobilized HSCs. Twenty out of the 23 patients reverted their insulin dependence. Twelve patients maintained this status for a mean 31 months (range 14–52 months) and 8 patients relapsed and resumed insulin use at low dose (0.1–0.3 IU/kg). Two patients developed bilateral nosocomial pneumonia, 3 patients developed late endocrine dysfunction, and 9 patients developed oligospermia. There was no treatment-related mortality. The study design did not include a randomized control group that either received no intervention or received only immunosuppression or immunomodulation. Only the result of a long-term monitoring of β-cell function over the coming months can tell whether the cost/benefit ratio of this approach can support the procedure. The question whether the beneficial effect of autologous HSC transplantation is due to β-cell regeneration, by the intensive immunomodulation that stops autoimmune destruction of remaining β-cells, or both, remains open.
Autologous BM Arterious Infusion Into the Pancreas After Femoral Catheterization in Type 1 and Type 2 Diabetic Patients
The use of an arterial infusion of mononucleate cells derived from BM directly in the pancreas through the splenic artery solely for the purpose of autologous β-cell regeneration without immunosuppression was reported in 2005 by the Argentine team directed by Fernandez-Vina. The data (reported at the 45th Annual Meeting of the American Society of Cell Biology in San Francisco in 2005) showed the procedure's tremendous benefit in type 2 diabetic patients (suspension of pharmacological treatment in 84% of the patients treated) but they have not been published yet. On the other hand, a prospective phase 1 study enrolling 25 patients with type 2 diabetes who received a combination therapy of intrapancreatic autologous stem cell infusion and peri-infusion hyperbaric oxygen treatment between March 2004 and October 2006 at Stem Cells Argentina Medical Center (Buenos Aires) was recently published (34). All metabolic variables tested (fasting glucose, HbAlc, fasting C-peptide, C-peptide/glucose ratio, and insulin requirements) showed significant improvement when comparing baseline to 12 months follow-up. Intrapancreatic autologous stem cell infusion was also reported as a safe and effective modality of treatment to improve β-cell function in 10 patients with type 2 diabetes treated at Postgraduate Institute of Medical Education and Research in India (9). Further randomized controlled clinical trials will be required to confirm these exciting findings. Phase I/II clinical trials of intra-arterial pancreatic infusion of total autologous BM and/or BM-derived stem cell are currently under way: (i) for the treatment of type 2 diabetes at Fuzhou General Hospital in China (in combination with hyperbaric oxygen therapy; ClinicalTrials.gov Identifier: NCT00767260), at Postgraduate Institute of Medical Education and Research in India (ClinicalTrials.gov Identifier: NCT00644241), at Shandong University in China (ClinicalTrials.gov Identifier: NCT00465478); (ii) for the treatment of type 1 diabetes at University of Moròn in Argentina (ClinicalTrials.gov Identifier: NCT00971503), at Hospital Clinic of Barcelona in Spain (ClinicalTrials.gov Identifier: NCT008 21899), and at Shandong University in China (ClinicalTrials.gov Identifier: NCT00465478).
Intra-BM Islet Allotransplantation in Type 1 Diabetic Patients
Because adult stem cells of BM origin may have a role in the initiation of endogenous β-cell regeneration and in favoring β-cell survival, this potential can also be exploited from a novel and opposite procedure, considering BM as the ideal microenvironment for islet survival (108). In theory, the BM offers an extravascular, protected, and well-vascularized (even if hypoxic) microenvironment (14,57), capable of sustaining islet grafts. Recently, a direct intrabone new administration route for cord blood cell transplantation has been successfully established in patients with acute leukemia (37) with the aim to provide an improved hematological recovery as a result of better stem cell homing. The procedure is easy and reproducible: a standard needle for BM aspiration is inserted in iliac crest and cells are gently infused in bone marrow cavities. Because of its broad distribution and easy access, BM has the potential to overcome not only the physiologic loss of islets, but also the technical limitations and complications encountered with the intraportal infusion. In accordance with this hypothesis we are exploring the possibility to use BM as the site for islet transplantation. Preliminary data in mice indicate that pancreatic islets engraft efficiently in BM and are able to maintain glucose metabolism up to 1 year. In the syngeneic model of marginal islet mass transplantation, BM guarantees a higher probability to reach euglycemia than liver (2.4-fold increase) and, after gain of normoglycemia, islets infused in BM guarantees a control of glucose metabolism similar to that of nondiabetic mice (16).
Because BM as a site for pancreatic islet grafts is clinically applicable, we started a pilot clinical trial at San Raffaele Diabetes Research Institute (HSR-DRI, Milan). At present, 5 patients having hepatic contraindications for liver islet transplant have received a single intra-BM islet infusion at the level of the iliac crest (100). Thus far we observed absence of adverse events related to infusion. Just as important, we observed presence of insulin-producing cells on a BM biopsy 1 month after transplantation, which was associated with significant levels of circulating C-peptide. Starting from this experience and promising results in animal models, we are planning to perform a phase II study in type 1 diabetes patients to compare the efficacy of islet transplantation in subjects infused either in BM or liver.
Conclusion
There is mounting evidence that BM and BM-derived stem cells can participate in regeneration of pancreatic islets following chemical and autoimmune injury in vivo. Experimental in vitro and in vivo data associated to the easy availability of BM and to the widely consolidated clinical experience in the field of hematology will expand in the near future the clinical trials involving BM to cure both type 1 and type 2 diabetes in humans. Future studies should evaluate the effect of BM and BM-derived stem cells on the prevention of and cure for diabetes by examining mechanistic bases first in mice and then in pilot human studies. The comprehension of the major mechanisms involved will permit the identification of new molecular target and development of new pharmacological strategies to cure both type 1 and type 2 diabetes. Besides this, a major effort in the near future should be made to verify the existence of a cross-talk between BM and pancreatic islets able to support β-cell mass in physiologic and/or pathologic condition in the absence of drug- or transplant-induced BM mobilization. The fascinating hypothesis that BM could reinforce during life the β-cell population in stress situations could open a new point of view on the etiopathogenesis and the therapy of diabetes.
Footnotes
Acknowledgments
This work was supported by EU FP7 (DIAPREPP: Project No. 202013, and BetaCellTher apy—Beta Cell Therapy in Diabetes: Project No. 241883).
