Stem Cell Transplantation in the Treatment of Type 1 Diabetes Mellitus: From Insulin Replacement to Beta-Cell Replacement

Type 1 diabetes mellitus (T1DM) is an autoimmune disease that attacks pancreatic β-cells, leading to the destruction of insulitis-related islet β-cells. Islet β-cell transplantation has been proven as a curative measure in T1DM. However, a logarithmic increase in the global population with diabetes, limited donor supply, and the need for lifelong immunosuppression restrict the widespread use of β-cell transplantation. Numerous therapeutic approaches have been taken to search for substitutes of β-cells, among which stem cell transplantation is one of the most promising alternatives. Stem cells have demonstrated the potential efficacy to treat T1DM by reconstitution of immunotolerance and preservation of islet β-cell function in recent research. cGMP-grade stem cell products have been used in human clinical trials, showing that stem cell transplantation has beneficial effects on T1DM, with no obvious adverse reactions. To better achieve remission of T1DM by stem cell transplantation, in this work, we explain the progression of stem cell transplantation such as mesenchymal stem cells (MSCs), human embryonic stem cells (hESCs), and bone marrow hematopoietic stem cells (BM-HSCs) to restore the immunotolerance and preserve the islet β-cell function of T1DM in recent years. This review article provides evidence of the clinical applications of stem cell therapy in the treatment of T1DM.


INTRODUCTION
Diabetes mellitus (DM) characterized by hyperglycemia, caused by insufficient insulin secretion or insulin resistance, is a group of chronic metabolic diseases. According to the International Diabetes Federation (IDF), the global adult diabetes population will exceed 537 million by 2021, and more than three-fourths of people with diabetes live in low-and middle-income countries, indicating that diabetes disproportionately affects the poor (http://www. diabetesatlas.org/). Diabetes is classified into four types: type 1 diabetes mellitus (T1DM), type 2 diabetes mellitus (T2DM), gestational diabetes mellitus (GDM), and monogenic diabetes mellitus (1)(2)(3)(4)(5). T1DM is an autoimmune disease, where autoreactive T cells attack pancreatic b-cells, leading to insulitisrelated islet b-cell destruction, which results in an absolute lack of insulin secretion causing hyperglycemia, abnormal glucose metabolism, and lifelong dependence on exogenous insulin. The majority of T1DM patients have poor blood glucose control and large blood sugar fluctuation. Chronic hyperglycemia results in the development of serious complications associated with diabetes, such as microvascular and macrovascular complications, reducing the quality of life and causing a considerable economic burden on T1DM patients and the society (6). The incidence rate of T1DM is increasing every year around the world (7,8). Although there is evidence that a combination of genetic susceptibility and environmental factors can increase the risk of immune disorder in T1DM patients, the exact etiology of the impaired immune system in T1DM is still unclear. More scientific efforts are needed to prevent b-cell loss and improve the quality of life in T1DM.

THE DIFFICULTIES OF INSULIN REPLACEMENT AND b-CELL REPLACEMENT IN T1DM
At present, the treatment and preventive options for T1DM are limited, mainly through insulin replacement therapy. T1DM cannot be cured; patients must rely on exogenous insulin injections for the rest of their lives to maintain glycemic control. Lente and NPH insulin were the only effective methods for the treatment of T1DM in the past (9,10). In recent years, novel approaches to insulin treatment, such as the introduction of glycosylated hemoglobin assays (HbA1c) and continuous glucose monitoring (CGM), have been used, and the effectiveness of basal/bolus therapy using portable continuous subcutaneous insulin infusion (CSII) pumps and closed-loop artificial pancreas system has been demonstrated. Artificial pancreas combining CGM with CSII pumps could automatically administer an appropriate insulin dose via a dosing algorithm. Some randomized controlled trials proved that the artificial pancreas system could efficiently adjust the glycemic index by automatically delivering exogenous insulin with dosing algorithms based on sensor glucose levels (11). However, the lag time of glycemia detected by CGM and the risk of hypoglycemia and infections limit the application of artificial pancreas, and some of the patients with unawareness of hypoglycemic events such as brittle type T1DM are not qualified to use the artificial pancreas (12,13). Also, insulin replacement therapy can only supplement the missing insulin and cannot fundamentally restore the function of the pancreas. Although these achievements can better manage blood glucose and large blood sugar fluctuation in T1DM, they can hardly prevent the occurrence of a series of complications, including microvascular, macrovascular, and neuropathy complications (14,15). As a result, many adjunctive therapies, such as dietary and weight management, nutrition therapy, physical activity and exercise, and some drugs used to treat T2DM, have been proposed to treat T1DM, which alleviate blood glucose fluctuation and reduce the lifetime risk of complications to some extent, but their effectiveness is limited. Therefore, it is very important to develop better technology and equipment for diagnosis and treatment options to prevent T1DM (16)(17)(18).
b-Cell replacement has also been proven as a curative measure in T1DM, which may be achieved through pancreas or islet transplantation in selective candidates (19). Pancreatic transplantation has the potential of re-establishing physiologicregulated insulin production, obviously decreasing the risk of hypoglycemic unawareness and finally decreasing the longtime risk of mortality from severe hypoglycemic complications (20). Since 2000, b-cell replacement through intrahepatic isolated islet transplantation has proven efficacious, indicating that islet transplantation is also an important option in the treatment of T1DM (21). Compared with the artificial pancreas system, islet transplantation and pancreatic transplantation were the better options to relieve the symptom of T1DM patients with unawareness of hypoglycemic events such as brittle type T1DM for a long time (22). T1DM patients can be clinically alleviated through improved control of the levels of blood glucose and restored awareness of hypoglycemia, resulting in the prevention of several life-threatening complications associated with diabetes, such as diabetic foot, microvascular and macrovascular diseases, kidney failure, nerve damage, and blindness (23). During the process of pancreatic or islet transplantation, both the autoimmune and alloimmune systems are still major threats to increase the transplantation risk. Patients treated with cell replacement therapies require immunosuppressive drugs as life-long treatment, and in many cases, these drugs lead to toxicities and side effects that made the adoption of this treatment strategy limited to only the most severe disease cases, inhibiting the widespread adoption of pancreatic or islet transplantation therapies in T1DM (24).
Besides the immune problem, the logarithmic increase in the global population of people with diabetes, the limited donor supply, and the need for lifelong immunosuppression restrict its widespread use (25). Numerous therapeutic approaches have been reported to solve this problem, including the search for bcell substitutes, porcine islet xenotransplantation, and stem cell transplantations, which present solutions to the donor shortage and may be the most likely alternatives (26,27).
Although the artificial pancreas system and pancreatic transplantation in T1DM can normalize and improve glycemic control in T1DM, the application of artificial pancreas systems and pancreatic transplantation is still limited due to their shortage. To solve the problem, stem cell transplant is a promising new strategy for patients with T1DM. There are many advantages of stem cells in the treatment of T1DM: first of all, stem cells such as bone marrow-derived stem cells (MSCs) can easily be obtained from bone marrow, umbilical cord blood, adipose tissue, etc. compared with islet and pancreas; secondly, the pluripotent stem cells could differentiate into b-cells and increase the secretion of insulin; thirdly, stem cells can moderate the immunome effect by inhibiting T-cell proliferation and reduce the inflammatory response, which can protect b-cells from autoimmune attack; and finally, stem cells can secrete cytokines by paracrine effects to enhance the antioxidant and proliferation ability of cells, which can help improve the survival of b-cells. To better understand the constitution of immunotolerance and preservation of islet b-cell function, we reviewed the progression of stem cells in recent years and tried to provide support for the clinical applications of stem cell therapy in the treatment of T1DM, especially in the brittle type T1DM.

STEM CELL TRANSPLANTATION THERAPY FOR T1DM
Stem cells are undifferentiated cells capable of self-renewal, giving rise to virtually any tissue or organ (28)(29)(30)(31)(32)(33). Stem cells can be grouped into four broad categories based on their origin: adult stem cells (ASCs), fetal stem cells (FSCs), embryonic stem cells (ESCs), and induced pluripotent stem cells (iPSCs). iPSCs and ESCs are pluripotent stem cells (PSCs), whereas ASCs are unipotent or oligopotent (34)(35)(36). PSCs, such as human-induced PSCs (iPSCs) and human embryonic stem cells (ESCs), offer a reproducible source of human cells at a very early developmental stage with the potential to form any cell type in the adult body (37)(38)(39). iPSCs, human cord blood-derived multipotent stem cells (CB-SCs), hematopoietic stem cells (HSCs), and MSCs were used for the preservation of b-cells by islet protection and regeneration, and another potent function of stem cells is the ability to re-establish peripheral tolerance toward b-cells through remodeling of the immune response as well as through inhibition of autoreactive T-cell function (40,41). In general, stem cells can increase the mass of islets by the ability of differentiation to b-cells-like organoids, and reconstitute immunotolerance by inhibiting the immune response of T cell and Th1 cells through TGF-b and inflammatory pathways ( Figure 1). As T1DM is featured as an autoimmune disease by activating immune cells to attack and destroy pancreatic b-cells, the immunomodulatory properties of stems cells and its potential ability of differentiation into insulin-producing cells should be considered when using stem cell therapy for T1DM treatment.

MSC Transplantation in T1DM
MSCs are one of the best candidate cells used as cell therapy for T1DM. MSCs are fibroblast-like, multipotent stromal, nonhematopoietic cells that could easily be sourced from various tissues, including adipose tissue, bone marrow, and umbilical cord blood (42). MSCs rapidly undergo mesodermal lineage differentiation, such as adipocytes, myoblasts, cardiomyocytes, chondrocytes, and b-cell-like cells (43)(44)(45). The bone marrow and umbilical cord blood could be separated over a gradient of Percoll by density gradient centrifugation to collect the MNCs, and the MNCs were washed with PBS and transferred to a 100mm culture dish to induce MSCs. The redundant tissues such as arteries and veins were removed from the adipose tissue, human umbilical cord, etc. The region of the remaining adipose or human umbilical cord tissues was diced into small fragments and seeded into a 100-mm culture dish to collect the MSCs. The induced MSCs were stored in liquid nitrogen and cultured for up to five passages for transplantation via intravenous injection. The characteristics of MSCs were defined by the International Society for Cell Therapy (ISCT) as follows: adherence to plastic; expression of the surface molecules CD73, CD90, and CD105 in the absence of CD34, CD45, HLA-DR, and CD14 or CD11b and CD79a or CD19; and the capacity for differentiation to adipocytes, osteoblasts, and chondroblasts in vitro (46). The potential of MSCs as a cell-based therapy in the treatment of immunologic disorders has been well established (47). MSCs can alter the microenvironment in tissues and promote existing bcell survival and regeneration, resulting in increased b-cell mass and normal blood glucose recovery (48)(49)(50)(51). Injection of bone marrow MSCs into diabetic mice can increase insulin levels and downregulate hyperglycemia; the exosomes derived from human umbilical cord stem cells (hUCMSCs) can enhance insulin sensitivity (52). Similarly, monotherapy with human umbilical cord MSCs reverses autoimmunity, promotes islet cell regeneration, and improves blood glucose control (53)(54)(55).
The allogeneic MSCs have been attempted in clinical trials, which can improve the level of insulin and C-peptide and reduce blood glucose. Although MSC xenotransplantation was not used in the clinics, several pieces of evidence showed that humanderived MSCs could alleviate the diabetic symptom through bcell-like organoid differentiation and immunomodulation in NOD mice, rats, and monkeys, while far-red light, gene editing, and other modifications could enhance the function of MSCs in a T1DM animal model, indicating that the intervening and xenograft MSCs are the potential option for T1DM treatment.

Immunomodulatory Ability of MSCs
MSCs can protect b-cell, increase the secretion of insulin, and reduce glycemia in patients with T1DM by regulating the immune system. The application of MSCs in eliminating autoimmune diseases has been fully proven in an animal model, and MSCs have a wide range of regulatory effects on immune cells. Domouky et al. showed that MSCs could reduce hyperglycemia in diabetic rats on day 15 (56). The inhibition of T-cell proliferation in islets and the presence of increased Treg in T1DM were features of MSCs' autoimmune properties (57). Shigemoto-Kuroda et al. developed T1DM mouse models for autoimmune diseases and discovered that MSCs could suppress type 1 helper T cell (Th1) development and delay the onset of T1DM in mice. CD4 + cells were found in significant numbers in the islets of mice treated with PBS, while fewer CD4 + cells were found in the islets of MSC-treated mice. The level of insulin in plasma was increased by MSC treatment, and there was a significant reduction in the production of IL-12, IFN-g, p70, and tumor necrosis factor (TNF) (58). Bassi et al. isolated MSCs from epididymal fat tissue from 8-week-old male Balb/c mice and characterized through immunophenotyping its capacity to prevent the proliferation of CD4 + T cells (59,60). Treatment of NOD mice with MSCs attenuated hyperglycemia of early-onset autoimmune diabetes and increased amylin levels, reflective of autoimmune diabetes improvement; reduced the amount of inflammatory cell infiltration, maintaining insulin expression in pancreatic islets by suppressing the Th1 immune response in the pancreas; and promoted the high expression of active TGF-b1 (60). Meanwhile, syngeneic MSCs were detected for a significantly longer period, albeit with diminishing persistence in immune-deficient mice model (61). In another study, van Megen et al. found that activation of MSCs can take up and process antigen and increase HLA-DR expression and immune inhibitory markers, while their metabolic profile was maintained without enhancing T-cell proliferation. MSCs can also enhance immunosuppressive capacity without stimulating alloreactive T cells (62). In an in-vitro study, Montanucci et al. provided preliminary evidence that immunoisolatory microcapsule-hHUCMS (CpS-hUCMS) may represent a functional biohybrid artificial system, where molecular products can induce effective immunomodulatory effects in vitro and in T1DM patients, making it possible to further clarify their therapeutic potential in humans (63). Montanucci et al. isolated and microencapsulated human umbilical cord Wharton jelly-derived mesenchymal stem cells (hUCMS) for  xenograft (TX) in a spontaneous T1DM mouse model (NOD mice). At 10 days of TX, Treg cells did not increase, while at 216 days of TX, CD4 + CD25 high cells increased in terms of both percentage and number. Further research found that at 216 days of TX, only the mild T1DM NOD mice presented sustained and full alleviation of hyperglycemia, while no alleviation of hyperglycemia was observed in severe T1DM NODs. These findings suggested that the successful hUCMS therapy approach for the treatment of T1DM in NOD mice depended on the stage of the T1DM disease process, with severe T1DM NODs exhibiting a continuous decrease in residual b-cell mass (64). All these results provide encouraging first steps in the clinical translation of the use of preactivated MSCs as a cellular immune intervention therapy, which helps to treat inflammatory and autoimmune disorders, including T1DM.  (66). Similar results were reported by several other groups that isletlike clusters can be formatted in vitro by cultured MSCs given the appropriate procedure (67,68). Human umbilical cord Wharton jelly cells (hUCWJCs) are a subtype of MSCs, which were transplanted into a T1DM mouse model with renal damage, and the therapeutic effect of transplantation was evaluated. It was found that hUCWJCs can promote the level of C-peptide and insulin in mice, which certified the potential of intraperitoneal injection of hUCWJCs and the ability of hUCWJCs to migrate to damaged tissues to enhance the secretion of insulin from nonpancreatic local cells (69).  (50). Another clinical trial revealed that MSC injection through liver puncture could successfully decrease the levels of insulin, islet cells, and glutamic acid decarboxylase (GAD) antibody in two patients within 1 year, with a decreased concentration of blood glucose and HbA1c and increased concentration of C-peptide, indicating immune regulatory cell tolerance (71).

HSC Transplantation in T1DM
The conception of HSCs was generated in the 1950s with the discovery that intravenously injected bone marrow cells could rescue irradiated mice from lethality through re-establishing blood cell production (72). The ability to manage expansion and the characteristics of self-renewal of the hematopoietic compartment while maintaining the capacity for differentiation into HSCs were demonstrated (72,73). Peripheral hematopoietic stem cells are mobilized with cyclophosphamide and granulocyte colony-stimulating factors. Leukapheresis using a continuousflow blood cell separator was initiated when the rebounding CD34 + cells reached 10 cells/ml. Apheresis was continued daily until the number of harvested progenitor cells reached a minimum of 3.0 × 10 6 CD34 + cells/kg body weight. Unmanipulated peripheral blood stem cells were frozen in 10% dimethyl sulfoxide in a rate-controlled freezer and stored in the vapor phase of liquid nitrogen (74). Then, the collected cells were injected intravenously. HSCs have proven to be safe in human subjects and have been widely utilized as an effective treatment for hematological malignancies (75). Recently, HSCs have been used in T1DM for the suppressed function of the immune system response in both in-vitro and in-vivo studies.

Immunomodulatory Ability of HSCs
Immunomodulatory activity is the most important ability of HSCs in patients with T1DM. HSCs can inhibit the occurrence of T1DM (76)(77)(78). Patients with recent-onset T1DM have been triumphantly reverted to euglycemia by autologous hematopoietic stem and progenitor cell transplantation (AHSCT), and modulation of autologous hematopoietic stem and progenitor cells (HSPCs) with prostaglandins (PGs) in vitro enhances their immunoregulatory properties through increasing the expression of the immune checkpoint-signaling molecule PD-L1 (79). Wang et al. demonstrated a lower proportion of proliferating T conventional cells (Tcon) and a higher absolute number and percentage of Treg cells in pancreatic lymph nodes from resistant mice among the younger recipients compared to the rapid progressors among the older recipients, and older NOD mice progressed more rapidly to the end stage of diabetes (80). Although mixed chimerism with MHC-matched nonautoimmune donor bone marrow (BM) transplants did not prevent T1DM in NOD mice models, induction of either mixed or complete chimerism with MHC-mismatched BM transplants inhibited T1DM in the same mice (81). This limited the translational applications of HSCs to reshape the autoimmune response by myeloablative agents/approaches. The genetically modified HSCs were used to overcome the disadvantage. Ex-vivo genetic manipulation of NOD HSCs to encode proinsulin and transgenically target MHC class II could successfully prevent T1DM onset (78,82). The increased CXCL12 (SDF-1) level in bone marrow-derived HSCs of NOD mice is considered to change the transport of HSCs and peripheral dendritic cells, which is conducive to the occurrence of T1DM (78, 83). The results demonstrated the beneficial effects of AHSCT in patients with recent-onset T1DM by increasing the concentration of C-peptide and inducing insulin independence, and the safety and good tolerability of AHSCT compared with conventional intensive insulin therapy was also certified (77). Another clinical trial in Ning's research also proved that AHSCT was safe without a reduction in the diversity of T-cell receptor (TCR) repertoires, and TCR repertoires tended to be more stable after AHSCT (85). The clinical trial data also showed significant direct correlations between HSPC levels and the coefficient of variation of glucose levels or time in hypoglycemia, which were weaker in patients with long-standing diabetes than in those with short-term diabetes (86). HSC transplantation improves glycated hemoglobin levels in a time-dependent manner (87).

ESC and iPSC Transplantation in T1DM
ESCs and iPSCs are PSCs that can regenerate the islet b-cells and immune cells through differentiating, which helps to increase the mass of b-cells. D'Amour et al. firstly proved that ESC-derived bcells could be successfully generated through the in-vitro recapitulation of pancreatic islets and b-cell physiological development by stepwise application of specific factors (88,89). iPSCs, reprogrammed from somatic cells, have a similar ability to differentiate and proliferate like ESCs. iPSCs collected from the umbilical cord at birth have the potential for self-renewable multipotency and can differentiate into various lineages such as islets (31,59,90). Hence, iPSCs provide a promising platform to produce insulin-secreting cells in vitro. However, the utilization of ESCs and iPSCs is less due to law restrictions in many countries, so there is little clinical research on ESCs and iPSCs.

Immunomodulatory Ability of ESCs and iPSCs
Haque et al. characterized autoantigen-specific naturally occurring Treg-like iPSC-Tregs and proved that adoptive transfer of ovalbumin (OVA)-specific iPSC-Tregs greatly suppressed autoimmunity in the mouse model preventing the b-cells from destruction. These tissue-associated Tregs can effectively inhibit the migration and activity of the pathogenic immune cells and accumulate in the diabetic pancreas causing T1DM by downregulating the production of proinflammatory cytokine IFN-g and suppressing the expression of ICAM-1 (95). Another report declared that pancreatic endoderm derived from hESCs can generate functional insulin-producing cells in vivo regardless of the presence of innate lymphoid cell elements, and the combination of CTLA4Ig and anti-CD40L mAbs can block hESC-PE graft rejection in immunocompetent mice, while regulatory T cells were not needed for the tolerance during hESC-PE transplantation (96).
To increase the function of stem cells, it is very important to sustain the regeneration and differentiation ability of stem cells and prevent the programmed death of stem cells in vivo (30,97). Several approaches and conditions, including far-red light, genetic engineering, biological material scaffolds, nanofiber tubular, combination treatment with insulin or other drugs, microcapsules, and co-transplantation with more than one type of stem cells, were utilized to promote the survival, differentiation, and immunomodulatory ability of stem cells in vivo and in vitro. These preclinical attempts tried to derive pancreas islet cells, increase the number and function of Tregs, ameliorate the function of islets, and prevent b-cells more effectively (Table 1). Also, stem cells have been used in human clinical trials, which showed that stem cell transplantation had beneficial effects on T1DM, with no obvious adverse reactions ( Table 2). Recently, an allogeneic, gene-edited, immune-evasive, stem cell-derived therapy for the treatment of T1DM was approved in Canada for clinical trial application (CTA)   (CRISPR Therapeutics and ViaCyte, Inc. to start clinical trial of the first gene-edited cell replacement therapy for the treatment of T1DM, retrieved on November 16, 2021). This CRISPR therapeutics offered novel b-cell replacement therapies to address unmet T1DM needs. All these efforts are aimed at better promoting the effectiveness of stem cells, which proved to be a more viable option for the treatment of T1DM, lessening the suffering of the patients.

CONCLUDING REMARKS
In recent research, stem cell therapy has demonstrated itself as a rapidly expanding and potentially limitless source of b-cells to arrive at a cure for T1DM by reconstitution of immunotolerance and differentiation into islet b-cell clusters. As the immunosuppression affected the effect of transplantation of stem cells, stem cell intervention before transplantation could help preserve b-cells and remodel the immune response. However, several challenges, such as the ethical problem of autologous and allogeneic stem cells used to preserve the function of b-cells, still need resolution. Although research into b-cell replacement derived from stem cells is increasing every year, we must make more efforts in the future on the intervention with stem cell transplantation, which can help achieve remission of T1DM by b-cell replacement.