Article Published in the Author Account of

Steven Ringquist

Regenerative Medicine for Diabetes Treatment

Abstract: The best and enduring treatment for the autoimmune, type 1 diabetes is to restore a patient's internal insulin production capacity. This may be possible by controlling the autoimmune destruction of islet β cells, the use of stem cells, and gene therapy, or a combination of the three.

Type 1 diabetes (T1D) is an autoimmune disease in which insulin-producing beta-cells contained within the pancreatic islet of Langerhans are destroyed by autoreactive T cells. T1D patients are treated via insulin hormone replacement therapy by subcutaneous injection of recombinant insulin (produced by molecular engineering). Blood glucose levels must be monitored many times a day to determine the appropriate quantity of insulin to be injected in order to control blood glucose levels (glycemia). Under the insulin-based treatment, the large and sustained effort that a patient must make to strive for near optimal control of glycemia over many decades, frequently beginning in childhood, often make this approach impractical. As a result, T1D contributes substantially to the high rate of nephropathy, neuropathy, retinopathy, and generalized microvascular disease experienced by this population.

Since insulin replacement therapy alone does not completely protect these individuals from severe complications, more appropriate treatments for curing T1D are needed. Transplantation of the whole pancreas or isolated pancreatic islets, have both been proposed in the aim of more effectively treating patients with complicated T1D. However, tempering the initial enthusiasm over transplantation has been the reported worsened survival rate for recipients of the pancreas alone, when compared with the survival of waiting-list patients receiving conventional insulin therapy, and the follow-up studies on islet recipients in which a gradual loss of islet function has been observed with time. Other problems arising from the use of powerful immune suppression drugs, required to avoid host rejection of islet graft and recurrence of autoimmunity, make transplantation an unsuitable alternative treatment for young T1D patients (Rother and Harlan, 2004).

Even in the event that transplantation-based approaches will improve and eventually prove to be superior in the long term reestablishment of euglycemia (normal blood glucose levels), reduced incidence of T1D complications, and overall patient health, the number of available donor organs will continue to limit the number of individuals who can be treated. Soberingly, the number of people in the U.S. with T1D is estimated at more than 1 million with roughly 30,000 new cases diagnosed annually. As current islet transplantation methodology uses multiple pancreases per recipient, less than 0.1% of the total U.S. T1D population can hope to be treated annually, only half of whom would be expected to achieve temporary insulin independence. Considered as a whole, these results illustrate the urgent need for the exploration of additional avenues in order to realize the goal of efficiently curing T1D.

Regeneration of endocrine pancreas function has been documented in partial pancreatectomy models and in sporadic reports involving spontaneous remission in T1D patients. This indicates at least the potential for the body to heal itself. Combining our knowledge in embryonic stem (ES) cells, adult stem cells, and gene therapy into a unified approach is the challenge awaiting us in our effort to cure T1D (Trucco, 2005).

Regeneration of Pancreas Endocrine Function

What we know about pancreas development continues to have a profound influence on experimental approaches to regenerative treatment of this organ since it is believed that regeneration recapitulates embryogenesis. In vertebrates, the process of gastrulation reorganizes the embryo into 3 layers: ectoderm, endoderm, and mesoderm. The endoderm gives rise to the respiratory and digestive systems, including the gut tube, and serves as a template for the gastrointestinal tract. It is from the gastrointestinal tract that the embryonic pancreas buds. The distinctive branching morphogenesis of the pancreatic bud gives rise to the gland’s ducts and acinar (granular masses) components. In turn, aggregates of differentiated cells (islets of Langerhans) arise from endocrine (hormonal or “internal” secretions) progenitors that proliferate from the budding ducts. Islets are composed of at least 4 cell types, organized within a highly vascularized compartment of the pancreas, constituting an estimated 1-4% of the gland’s endocrine component.

Within the pancreatic islet, the alpha cells produce glucagon (which has the opposite effect to insulin, in that it promotes glucose release); the beta cells, insulin; the gamma cells, pancreatic polypeptide (regulates physiological functions; a deficiency may be linked to obesity); and the delta cells, somatostatin (inhibits gastric secretion and motility) (Figure 1). The beta cells specifically are the target of autoimmune destruction in T1D, while the function of adjacent cells, endocrine as well as exocrine (”external” secretion via ducts), are left more or less intact.

Knowledge of pancreatic embryogenesis and, in particular, evidence of mechanisms that are able to maintain islet homeostasis has given rise to the hypothesis that endocrine progenitors may be present in the pancreatic ducts of the adult gland. Histological evidence has implicated islet neogenesis (i.e., generation of beta cells from non-insulin producing precursors) to occur preferentially near the pancreatic ducts in animal models. A number of groups have also reported physiological (pregnancy) or pathological (injury and obesity) circumstances promoting de novo expression of the insulin gene, as well as of other beta cell-specific mRNA markers. However, recent evidence based on methods used in cell lineage analysis has cast doubt on the relevance of the histology data: Strong evidence has been reported (Dor et al. 2004) that the existing beta cells are able to physiologically regulate their own homeostasis in the adult mouse as well as to promote a reparative effort after pancreatectomy. Adult stem cells seem to contribute little to this process.

At present, this topic is being hotly debated. For example, a pair of high profile reports published recently provide data supporting both of the opposing arguments: 1) Gershengorn et al. (2004) describe a likely pathway by which pre-existing adult human beta cells replicate involving at its core a reversible transition between epithelial (marking the differentiated beta cell and therefore its functional stage) and mesenchymal (identifying a replicating form) cell types; and 2) Seaberg et al. (2004) reported the isolation from adult mouse pancreas of a putative endocrine precursor that can be induced to express several beta cell-specific molecular as well as physiological markers. Nevertheless, the progenitor, which is responsible during embryogenesis and/or the hypothetical cell active in the adult capable of giving rise to new beta cells, is still awaiting definitive identification. A less differentiated precursor or a pre-existing beta cell must first be characterized for both improving our understanding of endocrine physiology and for developing possible therapeutic approaches based on exploiting the ability of these cells to regenerate islet mass.

Controlling Autoimmunity

In animal models for T1D, the autoimmune destruction of the pancreatic beta cell that generates an absolute dependence on externally supplied insulin to maintain glucose homeostasis, can be averted in a number of ways, including the elimination of the majority of autoreactive T cells with specific antibodies or the substitution of all or part of the immune cell repertoire with bone marrow cells obtained from diabetes-resistant donors (i.e., hematopoietic stem cell therapy). Once autoimmunity is controlled, regeneration of endocrine function may proceed, eventually replenishing the population of insulin-producing beta cells to a number sufficient to maintain euglycemia without the need for exogenous insulin therapy (Trucco, 2005) (Figure 2). While this process may take a period of months, it can be performed reproducibly. It remains to be established whether the regulation of autoimmunity followed by the spontaneous regeneration of endocrine function as demonstrated in the T1D mouse model, recapitulates the situation in humans.

Studies reported thus far show great promise in this line of inquiry. In fact, it is likely that the number of beta cells can increase physiologically to meet expanded metabolic requirements, as observed in pregnant T1D women during the first trimester. It also appears that the majority of patients experience intermittent remission after onset (i.e., the honeymoon phase), and it is known that as many as 40% of T1D patients maintain measurable islet beta cell function (as judged from serum levels of reactive C-peptide) for many years after becoming dependent on insulin replacement therapy. Furthermore, there are a handful of reliable reports indicating that some individuals with T1D are spontaneously cured. For example, Karges et al. (2004) reported on the long-term recovery of beta cell function in a 13-year old T1D male patient in which autoimmunity disappeared.

The strong evidence for regenerative capacity in the endocrine component of the pancreas suggests that it is reasonable to assume that the pancreas could be exposed to either beta cell precursors or beta cell specific signals that promote the formation of new insulin-producing cells. We need to improve our understanding of these variables if we are to exploit them for therapeutic purposes.

How Do We Proceed?

Much effort is currently focused on the use of emerging technologies for their potential to provide curative therapies for T1D. The proposed use of ES cells, adult stem cells, as well as gene therapy approaches to the treatment of patient populations are at present still in their infancy. Relative to each of these approaches there have been reports of striking success followed by equally striking confounding results when followed-up by independent groups. While it is accepted that regeneration of beta cells occurs, basic work still needs to resolve the time frame and physiological conditions necessary for promoting regeneration. For instance, what physiological circumstances facilitate, or limit, regeneration and what conditions will enable clinicians to achieve the desired therapeutic results? We need to know whether it is possible to promote self-healing in T1D patients, where the regenerative process may have been repressed by autoimmunity for too long a period of time. In this case, successful moderation of autoimmunity alone may not allow for a full recovery. In the event that regeneration at a certain point in time is irreversibly compromised, it may still be possible to transplant into these individuals functional precursor cells obtained from healthy, currently diabetes-resistant donors.

Embryonic Stem Cells

In the case of directed differentiation of ES cells into insulin-producing cells, reports of success from a strategy based on the exposure of the pancreas to biological signals specific for the pancreatic endocrine system, a modification of the protocol previously adopted for successful generation of neural cells. Using this method several groups reported the generation of cells able to produce insulin, leading to great expectations as to the usefulness of ES cells in beta cell regeneration. However, these results quickly came under intense criticism primarily from evidence that cells in culture may concentrate from the culture medium the insulin they seemed to have produced. Additional work, perhaps using transgenic fluorescent proteins, will need to be performed to clarify the interpretation of this approach.

Adult Stem Cells

Likewise, adult stem cells have undergone intensive investigation and equally intensive scrutiny as to their usefulness for T1D treatment. There have been several reports on the existence of adult progenitors of beta cells. However, even the ability to recognize beta cell precursors does not reduce the difficulty of physical isolation of these cells from a patient’s pancreas, especially when present at very low numbers and occurring in a difficult to access location such as the pancreatic duct. Increasing the number of possible precursors in vitro while avoiding the activation of differentiation pathways is also considered problematic. Culturing and expanding beta cells in vitro has been difficult and perhaps limited to a few proliferative cycles.

In order to bypass these difficulties, promising work has been performed using isolated adult pluripotent stem cells, such as those present in bone marrow. The hypothesis being that spatially and temporally restricted signals from the environment of the endocrine pancreas will allow differentiation of these precursor cells into that of the target tissue. As proof of concept, hematopoietic stem cells from bone marrow have been induced to form liver cells in vivo and similar work has been reported converting mouse hematopoietic stem cells into Purkinje neurons. At the time these reports were published, host signals sent by secreted factors or by cell-cell contact were proposed to be able to guide the transplanted precursors to differentiate into the same type of cells surrounding them, even across different lineages. However, these conclusions have been challenged by the possibility that rather than represent transdifferentiation, the results were tainted by false positive signals originating from sporadic cell fusion events. While it is increasingly likely that hematopoietic stem cells, as they were used in these reports, do not by themselves contribute to endocrine pancreas regeneration, there is still a formal possibility that fusion of cells can be useful even if that wasn’t what the authors of the original studies intended to show.

Perhaps, hematopoietic stem cells are capable of reaching areas of the body that are reasonably well isolated, such as the endocrine compartment of the adult pancreas thus allowing the generation of newly functional cells in situ.

Gene Therapy

Gene therapy is another method whose potential is being intensely explored for the treatment of T1D. If the regenerative potential of the endocrine pancreas is, in fact, lost then use of gene therapy may provide an important approach to overcome the damage inflicted by autoimmune disease. Successful transfection of cells, even belonging to a completely different lineage, with genes able to convert them into cells carrying the characteristics of those lost, has been reported. For example, the gut K cells of the mouse have been induced to produce human insulin by transfecting human insulin gene linked to the 5′-regulatory region of the glucose-dependent insulinotropic polypeptide (GIP) gene (which is secreted by K cells). In these transfected cells, the important property of glucose-stimulated expression of the controlled gene is automatically transferred to insulin production and eventually secretion. Moreover, separate experiments using gene therapy approaches have shown that after transfection with the Pdx1 gene, under control of the rat insulin 1 promoter, liver cells were able to produce insulin. The latter system produced enough insulin to satisfy the needs of a diabetic mouse, which once treated, became and remained steadily euglycemic.

Other Approaches

These experiments, however, still need independent confirmation and offer numerous hurdles that need to be overcome in order to translate their results into viable treatments for curing T1D. The requirement for histocompatibility in cell-based therapy is necessary to avoid the need for immune suppression. This problem can possibly be addressed by nuclear transfer procedures using the recipient’s own cells as donor of the nuclei. Homeostasis of beta cell mass must be controllable so that there is neither too many nor too few newly formed beta cells. Failure to treat autoimmunity in these individuals will only insure that the non-rejected graft will eventually be destroyed anyway. Anti-CD3 monoclonal antibody treatment to reinstall tolerance has been successful in the non-obese diabetic (NOD) mouse but, in humans, succeeded only in delaying the autoimmune disease.

Research has also provided evidence that antisense oligonucleotides targeting CD40, CD80, and CD86 primary transcripts have reduced the incidence of T1D in the mouse but this needs to be validated in people to accurately address their effectiveness (Trucco, 2005). Also, T1D can be prevented in the mouse model for T1D using gene therapy approaches where disease susceptibility MHC class II alleles are supplemented with alleles associated with diabetes resistance. If this latter result can be extended to already diabetic individuals, it may provide a suitable alternative to immune suppressor drugs in controlling autoimmunity (Trucco and Giannoukakis, 2005).


Beginning in childhood, T1D patients must have their blood glucose levels monitored and then controlled by multiple daily injections of insulin. Even in early adulthood, when patients are capable of caring for themselves, they are at risk for incidents of acute hypoglycemia (low blood glucose levels). While several studies have shown that rigorous supervision of HbA1C levels (amount of glucose bound to the A1C form of the oxygen-carrying hemoglobin [Hb] in red blood cells) over a lifetime can lessen the risk of developing several of the chronic complications associated with diabetes, many patients will still be at risk, while others will find it impractical to maintain the necessary regimen. Modern insulin replacement therapy, while excellent at saving lives, is unable to protect these individuals from the risks associated with sub-optimal control of blood glucose. Thus, there is an urgent need to exploit the promises implicit in the modern scientific breakthroughs of regenerative medicine.

In the near future, however, advances in T1D care need to be coupled with advances in regulating autoimmunity, the underlying cause of islet beta-cell loss in these patients. Such an approach is a prerequisite for both the use of ES cell-based therapy and the re-establishment of euglycemia, capitalizing on the pancreas’s regenerative potential. Intensive multidisciplinary effort between scientists involved in basic research, clinician-scientists working in translational medicine, and ultimately collaborations with endocrinologists will be required to establish the knowledge necessary to transform theory into practice.


This work was supported by grants U19-AI056374-01 Autoimmunity Centers of Excellence (SR and MT), RO1DK24021 (MT) from the National Institutes of Health, and ERHS #00035010 (SR and MT) from the Department of Defense, USA.

References and Further Readings

Dor Y, Brown J, Martinez OI, Melton D. Adult pancreatic beta-cells are formed by self-duplication rather than stem-cell differentiation. Nature 429:41-46, 2004.

Gershengorn MC, Hardikar AA, Wei C, Geras-Raaka E, Marcus-Samuels B, Raaka BM. Epithelial-to-mesenchymal transition generates proliferative human islet precursor cells. Science 306:2261-2264, 2004.

Karges B, Durinovic-Bello I, Heinze E, Boehm BO, Devatin KM, Karges W. Complete long-term recovery of beta-cell function in autoimmune type 1 diabetes after insulin treatment. Diabetes Care 27:1207-1208, 2004.

Rother KI and Harlan DM. Challenges facing islet transplantation for the treatment of type 1 diabetes mellitus. Journal of Clinical Investigation 114:877-883, 2004.

Seaberg RM, Smukler SR, Kieffer TJ, Enikolopov G, Asghar Z, Wheeler MB, Korbutt G, van der Kooy D. Clonal identification of multipotent precursors from adult mouse pancreas that generate neural and pancreatic lineages. Nature Biotechnology 22:1115-1124, 2004.

Trucco M. Regeneration of the pancreatic beta cell. Journal of Clinical Investigation 115:5-12, 2005.

Trucco M and Giannoukakis N. MHC tailored for diabetes cell therapy. Gene Therapy 12:553-554, 2005.

[Discovery Medicine, 5(25):142-147, 2005]

Access This PDF as a Subscriber |
E-mail It