Direct Differentiation of Bone Marrow Mononucleated Cells Into Insulin-Producing Cells Using 4 Specific Soluble Factors

Abstract

Bone marrow-derived stem cells are self-renewing and multipotent adult stem cells that differentiate into several types of cells. Here, we investigated a unique combination of 4 differentiation-inducing factors (DIFs), including putrescine (Put), glucosamine (GlcN), nicotinamide, and BP-1-102, to develop a differentiation method for inducing mature insulin-producing cells (IPCs) and apply this method to bone marrow mononucleated cells (BMNCs) isolated from mice. BMNCs, primed with the 4 soluble DIFs, were differentiated into functional IPCs. BMNCs cultured under the defined conditions synergistically expressed multiple genes, including those for PDX1, NKX6.1, MAFA, NEUROG3, GLUT2, and insulin, related to pancreatic beta cell development and function. They produced insulin/C-peptide and PDX1, as assessed using immunofluorescence and flow cytometry. The induced cells secreted insulin in a glucose-responsive manner, similar to normal pancreatic beta cells. Grafting BMNC-derived IPCs under kidney capsules of mice with streptozotocin (STZ)-induced diabetes alleviated hyperglycemia by lowering blood glucose levels, enhancing glucose tolerance, and improving glucose-stimulated insulin secretion. Insulin- and PDX1-expressing cells were observed in the IPC-bearing graft sections of nephrectomized mice. Therefore, this study provides a simple protocol for BMNC differentiation, which can be a novel approach for cell-based therapy in diabetes mellitus.

Graphical Abstract

Graphical Abstract.

Significance Statement.

Generating insulin-producing cells (IPCs) from stem cells is feasible and promising for treating diabetes. Our study investigated a unique combination of 4 differentiation-inducing factors, including putrescine, glucosamine, nicotinamide, and BP-1-102, to generate IPCs from bone marrow mononucleated cells (BMNCs): applying the 4 extrinsic factors generated glucose-responsive IPCs effectively in vitro and in vivo. We observed the hypoglycemic effect of the BMNC-induced IPCs grafted in diabetic mice. This study provides a simple protocol for differentiating BMNCs into IPCs, implicating the prospects of cell-based therapy in treating diabetes mellitus.

Introduction

Diabetes is a metabolic disorder characterized by chronic hyperglycemia due to the destruction or dysfunction of pancreatic beta cells. Islet transplantation is the most effective approach for reversing hyperglycemia. However, it is often hampered by immune rejection, autoimmunity against newly formed beta cells, or scarcity of donor islet cells.^1-3 Beta cell generation from stem or progenitor cells and trans-differentiation of mature cell populations^2,3 are potential alternatives to islet transplantation. Studies have demonstrated that embryonic stem cells (ESCs), induced pluripotent stem cells (iPSCs), and adult stem cells from the bone marrow, umbilical cord blood, pancreas, liver, and adipose tissues can differentiate into insulin-producing cells (IPCs).^4-6 However, ESCs and iPSCs have several limitations in clinical applications, such as production efficiency, high cost due to long exposure to exogenous soluble factors or small molecules, and safety concerns regarding tumorigenesis after transplantation.⁷ Therefore, adult stem cells are preferred for cell therapy in diabetes treatment.

Adult mouse bone marrow harbors cells capable of differentiating into pancreatic endocrine beta cells in vivo and serve as a source for cell-based diabetes treatment.⁸ Our previous studies have confirmed that BMNCs could differentiate into IPCs in vivo. This has been demonstrated by the neogenesis of green fluorescent protein (GFP)- and insulin-double-positive cells in streptozotocin (STZ)-induced diabetic chimeric mice, harboring BMNCs from the insulin promoter luciferase/GFP (MIP-Luc/GFP) transgenic mice.^9,10 These results demonstrate that BMNCs can differentiate into IPCs in response to the signals from damaged beta cells. Based on these findings, we developed a simple and reproducible IPC-generating priming protocol. In vitro-primed BMNCs regulate blood glucose levels in diabetic mice and improve fasting blood glucose levels and glucose tolerance.⁹

In this study, we assessed a unique combination of differentiation-inducing factors (DIFs) to develop a differentiation method for generating mature IPCs and apply it to our in vitro protocol. Certain extrinsic factors can facilitate the expression of transcription factors associated with beta cell differentiation and insulin production, enabling IPC generation. Interestingly, polyamines, which are ubiquitous aliphatic cations, exist at millimolar concentrations in pancreatic beta cells and are involved in regulating cellular proliferation and differentiation.^11-14 Intracellular polyamines stimulate insulin production and secretion.^11,12 Putrescine (Put), a representative polyamine, has been incorporated among extrinsic factors inducing stem cell differentiation from various sources to IPCs.^15-17 Glucosamine (GlcN) is a natural component of glycoproteins in connective tissues and gastrointestinal mucosal membranes and has been widely used for treating osteoarthritis. GlcN increases stem cell migration and proliferation and enhances cartilage differentiation from stem cells.^18,19 GlcN supplementation assists in differentiating adipose-derived stem cells (ADSCs) into glucose-responsive IPCs.²⁰ However, the combined effect of these factors on differentiation into IPCs has not been extensively studied.

This study investigated the combined effect of both metabolites, Put and GlcN, on the generation of IPCs from BMNCs, to optimize an efficient differentiation protocol for IPCs. Nicotinamide is a commonly used extrinsic inducer of endocrine differentiation to preserve islet viability and function.²¹ Additionally, the suppression of signal transducer and activator of transcription 3 (STAT3) signaling efficiently enhances the efficiency of reprogramming into beta cells induced by defined transcription factors.²² Therefore, we examined the combined effect of Put and GlcN with nicotinamide and STAT3 inhibitor BP-1-102 on in vitro differentiation into functional IPCs capable of alleviating hyperglycemia in a diabetic animal model.

Materials and Methods

The detailed methods are included in Supplementary material.

Differentiation Protocol of Murine BMNCs into IPCs

Whole BMNCs (5 × 10⁶ cells/well) (details of the culture are in Supplementary material) were seeded into 12-well non-coated plates and primed in suspension on a shaking (30 rpm) platform for 6 days in Connaught Medical Research Laboratories (CMRL) medium containing 10% fetal bovine serum (FBS) and antibiotics supplemented with 10 mM Put (Sigma-Aldrich, St. Louis, MI, USA), 10 mM GlcN (Sigma-Aldrich), 10 mM nicotinamide (Sigma-Aldrich), and 10 µM BP-1-102 (Sigma-Aldrich). The cells were collected for further experiments after differentiation for 6 days.

FACS Analysis

Flow cytometric analyses of BMNCs primed with DIFs were performed and analyzed using the BD FACS Canto II flow cytometer (ESM Methods).

RT-PCR and Quantitative Real-Time RT-PCR (qPCR)

Total RNA was isolated from cells using TRIzol reagent (Life Technologies, Waltham, MA, USA). According to the manufacturer’s instructions, cDNA was synthesized from 1 µg total RNA using reverse transcriptase (SuperScript Ⅱ; Promega, Madison, WI, USA). RT-PCR was performed to amplify each cDNA. The PCR products were analyzed and visualized (Supplementary material). qPCR was performed using SYBR Premix ExTaq (Takara, Shiga, Japan) and the ABI 7500 real-time PCR system (Applied Biosystems, Foster City, CA, USA). Each Ct value was subtracted from the GAPDH Ct value of the same sample (ΔCt) and then from the ΔCt value of each control set (ΔΔCt). Relative mRNA expression levels were calculated using the 2^−ΔΔCt method. The primer sequences are listed as ( Supplementary Supplementary Table S1).

Glucose-Stimulated Insulin Secretion (GSIS)

The cells were pre-incubated for 2 h in Krebs-Ringer Bicarbonate (KRB) buffer (see Supplementary material), followed by incubation in KRB buffer containing 2 or 20 mM glucose (Sigma-Aldrich) at 37 °C for 30 minutes. Insulin secretion was measured using the Mouse Ultrasensitive Insulin ELISA kit (ALPCO, Salem, NH, USA), according to the manufacturer’s instructions, and normalized to the protein content of the corresponding cell lysates.

Immunofluorescence Staining

Sectioned tissues or paraformaldehyde-fixed cells were washed thrice with phosphate-buffered saline (PBS), permeabilized with PBS containing 0.3% Triton X-100 (Sigma-Aldrich), and blocked with PBS containing 10% normal donkey serum (NDS; Jackson ImmunoResearch, West Grove, PA, USA) for 1 h at 20-22 °C. The cells were then incubated with primary antibodies (ESM Table 2) in 1% NDS overnight at 4 °C, washed thrice with PBS containing 1% NDS, and incubated with secondary antibodies in 1% normal goat serum (NGS) for 2 h at 20-22 °C. The cells were counterstained with diamidino-2-phenylindole (DAPI) (1:2500; Molecular Probes, Eugene, OR, USA) and observed using a confocal microscope (Leica Microsystems, Wetzlar, Germany) equipped with appropriate filters.

Transplantation of Primed BMNCs in Diabetic Mice

Diabetes was induced in male C57BL/6 mice (aged 8-12 weeks; n = 5-6 per group) using a single intraperitoneal injection of 150 mg/kg STZ (Sigma-Aldrich) (ESM Methods). Primed or non-primed BMNCs (1 × 10⁶ cells/mouse) were resuspended in Matrigel (Corning Life Sciences, New York, USA). The BMNC-containing Matrigel solution was aspirated into a 1 mL syringe connected to P50 polyethylene tubing and transferred to the head end. The recipient mouse was anesthetized, the left flank was shaved, and the kidney was exposed through a small lumbar incision. Capsulotomy was performed on the caudal outer surface, and the tip of the polyethylene tubing was inserted and advanced gently under the capsule. The kidney surface was kept moist with saline during the procedure. The BMNCs in the tubing were slowly injected, and the tube was removed after completely transferring the cells into the capsule. Random blood glucose levels, body weight, and food and water intake were assessed every 3-4 days.

Glucose Tolerance Test and In Vivo Glucose-Stimulated Insulin Secretion

Mice (n = 5-6 per group) were fasted overnight and blood glucose level was measured following intraperitoneal injection of glucose (1 g/kg body weight) at the indicated time points. The area under the curve (AUC) for glucose was then calculated. For in vivo GSIS, glucose (2 g/kg body weight) was intraperitoneally injected after overnight fasting. Blood samples were collected via the tail vein at baseline insulin levels (0 minutes) and 15 and 30 minutes after glucose loading. Plasma insulin levels were measured using Mouse Ultrasensitive Insulin ELISA (ALPCO), according to the manufacturer’s instructions.

Statistical Analysis

Data are presented as mean ± standard error of the mean unless specified otherwise. Statistical significance (P < .05) was determined using an unpaired Student’s t test.

Results

Identification of Putrescine as a DIF in IPCs

We examined the involvement of polyamine metabolites, spermidine, and spermine, in IPC differentiation using a 6-day priming protocol established in our previous study.⁹ We confirmed that BMNCs isolated from MIP-Luc/GFP mice primed with 100 µM spermidine or spermine for 6 days had a markedly increased number of GFP-expressing cells than untreated control cells had. Therefore, spermidine- or spermine-sensitized cells exhibited the characteristics of IPCs (ESM Fig. 1A). However, polyamines have strong cytotoxic effects in vitro because the polyamine oxidase in FBS converts them to aldehydes. Therefore, we investigated whether the precursor Put could replace the differentiation-inducing effects of spermidine and spermine. When commercially available human BMNCs were differentiated into IPCs in the presence of 10 mM Put, gene expression analyses of the primed BMNCs revealed induced mRNA expression of beta cell markers, including insulin (INS), PDX1, MAFA, NEUROD1, and NEUROG3, within 6 days (ESM Fig. 1B). BMNCs isolated from MIP-Luc/GFP mice and primed with Put had more GFP-expressing cells than those primed with spermidine or spermine (ESM Fig. 1A). This suggests that BMNCs, regardless of their origin (human or mouse), could differentiate into IPCs via sensitization with a single metabolite. Thus, Put can be considered a DIF for IPCs.

Glucosamine Supplementation Improves the Differentiation of BMNCs Into IPCs

O-GlcNAc is a post-translational modification of nuclear and cytosolic proteins. The enzyme responsible for O-GlcNAc modification, O-GlcNAc transferase, is highly expressed in pancreatic islets.²³ Studies have revealed that the transcription factor PDX1 is modified by O-GlcNAc. Also, an increase in protein O-GlcNAcylation correlates with an increase in the DNA-binding activity of PDX1 and insulin secretion. Further, O-GlcNAcylation downregulation attenuates insulin secretion in the pancreas.²⁴ In this study, we confirmed that O-GlcNAc levels were considerably higher in the pancreatic beta cell line MIN-6 than in the fibroblast NIH-3T3 cells (Supplementary Fig. S2A). Thus, we suggest that GlcN could increase O-GlcNAc levels by mimicking the hexosamine biosynthesis pathway.²⁵ When BMNCs isolated from MIP-Luc/GFP mice were primed in vitro with GlcN (10 mM) for 6 days, GFP was expressed via insulin promoter activation in GlcN-primed BMNCs (Supplementary Fig. S1A). We evaluated the in vivo maturation of the in vitro-primed BMNCs after a systemic infusion of GlcN-primed BMNCs (1 × 10⁶ cells) from MIP-Luc/GFP mice into the tail vein of hyperglycemic mice (ESM Methods). Immunofluorescence staining of the pancreatic tissues revealed GFP/PDX1 double-positive cells in the mice grafted with GlcN-primed cells. In contrast, no GFP-expressing cells were present in the pancreas of mice grafted with control cells (Supplementary Fig. S2B).

Establishment of the Optimal Condition for Differentiation of BMNCs-Derived IPCs Using the 4 Extrinsic Factors

qPCR revealed that co-treatment with Put and GlcN improved the differentiation of BMNCs into IPCs (Fig. 1). Nicotinamide promotes IPC maturation after the endocrine progenitor stage. Nicotinamide (10 mM) was added to the medium containing Put and GlcN to promote further differentiation. A combination of the 3 factors increased the transcript levels of Ins and mature beta cell markers (Pdx1, MafA, Nkx6.1, NeuroD1, and Neurog3) (Fig. 1). Suppressing STAT3 signaling efficiently enhanced the efficiency of reprogramming into beta cells induced by the defined transcription factors PDX1, NEUROG3, and MAFA.²² Therefore, BMNCs were treated with STAT3 inhibitor BP-1-102, alongside Put, GlcN, and nicotinamide, resulting in the most effective mRNA induction of beta cell markers (Fig. 1). GFP expression using BMNCs isolated from MIP-Luc/GFP mice also confirmed a synergistic increase in insulin promoter activation in response to the combination of the 4 factors (ESM Fig. 3).

Figure 1.

Synergistic effect of 4 DIFs in BMNC-derived IPCs. Beta cell-specific gene expression in IPCs on day 6 of differentiation from BMNCs treated with Put, GlcN, nicotinamide, and BP-1-102 was analyzed using qPCR. Results are presented as mean ± SEM from at least 3 independent experiments. Put, putrescine; GlcN, glucosamine. *Mean values are significantly different from that of non-treated BMNCs at *P < .05; **P < .01; ***P < .001. ^#Mean values are significantly different from that of BMNCs primed with Put + GlcN + nicotinamide at ^#P < .05.

Characterization of IPCs Following In Vitro Murine BMNC Culture With DIFs

Gene expression profiles for beta cell development, differentiation markers, and hormones were assessed using RT-PCR to determine whether the BMNCs primed with DIFs underwent differentiation.²⁶ Primed BMNCs underwent pancreatic endocrine differentiation, revealed by a gradual decrease in mRNA expression of definitive endodermal markers (Cxcr4, Sox17, and FoxA2) and an increase in mRNA expression of Pdx1, Nkx6.1, MafA, Neurog3, NeuroD1, and Ins until day 9. Primed BMNCs expressed somatostatin for the delta cells; however, glucagon for the alpha cells was not observed. This suggests that our protocol could be used to induce differentiation into Pax4-positive endocrine cells, including beta and delta cells, divergent from the Arx-positive alpha cell lineage (Fig. 2A). Thus, our differentiation protocol can be used to successfully differentiate BMNCs into endocrine progenitors. Next, to determine whether primed BMNCs produce PDX1 and synthesize insulin protein, BMNCs differentiated using the 4 extrinsic factors Put, GlcN, nicotinamide, and BP-1-102, were stained with anti-insulin and anti-PDX1 antibodies and visualized using confocal microscopy. Primed BMNCs were strongly stained with anti-insulin and anti-PDX1 antibodies (Fig. 2B). However, these proteins were rarely detected in the control cells. Furthermore, differentiation efficiency was evaluated using flow cytometry analysis of C-peptide- and PDX1-positive cells, revealing that the percentage of C-peptide- and PDX1-positive cells increased to 30.9% of total cells by day 6 (Fig. 2C).

Figure 2.

Characterization of IPCs following in vitro culture of murine BMNCs with 4 DIFs. (A) Expression of pancreatic developmental markers during differentiation was determined using PCR with reverse transcription. Data are representative of 3 independent experiments. (B) Immunofluorescence staining of cells using anti-insulin and anti-PDX1 antibodies, followed by DAPI staining for nuclei and observation using a confocal fluorescence microscope. Scale bar = 50 µm; original magnification, 1200×. (C) Representative FACS plots illustrating the expression of C-peptide and PDX1 in BMNCs-derived IPCs. The numbers mark the percentage of cells in each quadrant. Ins, insulin; Gcg, glucagon; Sst, somatostatin.

Differentiation of BMNCs Into Functional IPCs Using a Combination of 4 Extrinsic Factors

Insulin release from vehicle-treated control or DIF-treated (primed) BMNCs was measured using ELISA to determine whether the primed BMNCs were responsive to glucose challenge. Insulin release from the differentiated BMNCs was markedly higher in both basal and glucose-stimulated conditions than from control BMNCs (Fig. 3A). Insulin secretion in the primed cells was approximately 6-fold higher than that in the vehicle-treated cells under high-glucose conditions (control vs. primed cells, 1.48 ± 0.45 vs. 8.64 ± 0.67) and approximately 2.5-fold higher under low-glucose conditions (control vs. primed cells, 1.3 ± 0.37 vs. 3.45 ± 0.73). In contrast, control BMNCs did not significantly release insulin in the presence or absence of glucose (Fig. 3A). These results suggest that DIFs are indispensable in differentiating BMNCs into IPCs, and differentiated BMNCs are responsive to the glucose challenge. We performed qPCR to examine the mRNA expression of Glut2, Gck, and Syt in the differentiated cells to investigate whether the DIF-mediated insulin secretion is induced by the expression of insulin secretion machinery-associated genes. Consistent with the increased insulin release in primed cells, the mRNA levels of these genes were markedly increased in the primed cells (Fig. 3B). These results indicate that insulin secretion and dose-dependent glucose responsiveness of differentiated cells might be caused by an increase in the expression of molecules involved in the insulin secretion machinery.

Figure 3.

The functionality of BMNC-derived IPCs. (A) Ultrasensitive ELISA measured insulin secretion from differentiated IPCs on day 6 at low (2 mM) and high (20 mM) glucose concentrations. The results presented are those of 5 separate experiments. *Mean values significantly differ from that of non-treated control BMNCs at *P < .05 and ***P < .001. ^#Mean values significantly differ from that of primed cells at low glucose concentrations at ^#P < .001. (B) mRNA expression of Glut2, Gck, and Syt in differentiated IPCs was measured using qPCR. The results presented are those of at least 3 separate experiments. Data are mean ± SEM. *P < .05; **P < .01. Glut2, glucose transporter 2; Gck, glucokinase; Syt, synaptotagmin.

BMNC-Derived IPCs Improve Hyperglycemia in STZ-Induced Diabetic Mice

In C57BL/6 mice with STZ-induced hyperglycemia (Fig. 4A), blood glucose levels were measured every 3-4 days for 42 days after grafting BMNC-derived IPCs. The STZ-treated control mice grafted with non-primed BMNCs had elevated blood glucose levels, persistent weight loss, and a significant increase in food and water intake after hyperglycemia onset (Fig. 4B-D). However, the IPC-implanted mice exhibited an early decrease in blood glucose levels, ie, by day 18 of post-grafting (Fig. 4C). Aberrant food and water intake were decreased post-grafting (Fig. 4B). In addition, the IPC-grafted mice maintained their body weight; however, the control mice had a marginal decrease in body weight (Fig. 4D). These results suggest that BMNC-derived IPCs are functional in vivo and prevent hyperglycemia in mice. Mice implanted with BMNC-derived IPCs in hyperglycemic conditions had significantly lower fasting blood glucose levels and improved glucose tolerance 27 days post-grafting compared to the controls. The area under the glucose clearance curves was significantly reduced in IPC-grafted mice following the intraperitoneal glucose challenge (Fig. 4E). Further, we measured GSIS in vivo to determine the functional responsiveness of IPCs. GSIS analysis showed significantly higher (P = .017) plasma insulin at 15 minute post-glucose injection in IPC-transplanted mice than in STZ-treated control mice (Fig. 4F). Immunofluorescence staining of kidney sections from the nephrectomized mice confirmed the presence of insulin- and PDX1-expressing cells in IPC-transplanted kidney (Fig. 5A). Finally, to eliminate the possibility of endogenous pancreatic beta cell regeneration, we performed immunohistochemistry using anti-insulin antibody on the pancreatic tissues on day 42 from the STZ-treated mice grafted with BMNC-derived IPCs. The pancreas from mice grafted with IPCs almost completely lost pancreatic islets, comparable to the diabetic controls, and endogenous beta cell regeneration was unnoticed. As an additional assessment, we quantitatively evaluated the pancreatic insulin content in these animals, resulting in no significant difference between the groups (control vs. primed cells, 1.77 ± 1.46 vs. 2.20 ± 0.91; P = .598) (Fig. 5B,C). Therefore, IPCs responded to the glucose challenge in vivo by releasing insulin. Analysis of IPC graft sites 42 days after transplantation revealed no substantial tumor formation. Collectively, the engrafted IPCs remained differentiated and survived up to a month after transplantation. Our data suggest that BMNC-derived IPCs are responsible for reversing hyperglycemia in diabetic mice.

Figure 4.

Transplantation of primed BMNCs improves hyperglycemia in STZ-induced diabetic mice. (A) Summarized scheme of animal experiments. Male C57BL/6 mice were intraperitoneally injected with STZ. After 7 days, primed or non-primed BMNCs (1 × 10⁶ cells/mouse) were transplanted under the kidney capsule, followed by monitoring for 42 days. (B) Food (left panel) and water (right panel) intake was measured every 3-4 days. Changes in random feeding blood glucose levels (C) and body weight (D) of mice transplanted with primed BMNCs (filled circle, straight line) or comparable non-primed BMNCs (empty circle, dotted line) were measured on the indicated days. (E) Blood glucose levels from the intraperitoneal glucose tolerance test (ipGTT). Glucose (1 g/kg body weight) was injected after overnight fasting for 27 days post-grafting. AUC (right panel) was measured. (F) Plasma insulin levels at the indicated time points post-glucose injection (2 g/kg, body weight) following overnight fasting was measured using ultrasensitive ELISA for mouse insulin. Data are presented as mean ± SEM. *P < .05; **P < .01 for non-primed vs. primed BMNC-transplanted mice. STZ, streptozotocin.

Figure 5.

Analysis of grafts following kidney excision using immunofluorescence staining for insulin and PDX1 expression. (A) Representative immunofluorescent images of nephrectomized grafts stained with anti-insulin and anti-PDX1 antibodies, followed by DAPI staining for nuclei and observation under a confocal fluorescence microscope. Scale bar = 50 µm; original magnification, 1200×. (B) Representative immunohistochemical images of the pancreas for insulin harvested 42 days after transplantation. Images were acquired with an ECLIPSE Ci-L microscope. Scale bar = 100 µm; original magnification, 200×. (C) Pancreatic insulin content in mice was determined using high-range ELISA for mouse insulin from pancreatic tissues harvested 42 days after transplantation. All data are presented as mean ± SEM from 5-6 mice from each group.

Discussion

We explored the possibility of using BMNCs to generate IPCs under specific in vitro conditions to overcome the limitations in generating IPCs from progenitor cells. Pluripotent bone marrow cells are a safe and abundant source of adult stem cells.²⁷ Our study reveals that 4 DIFs can effectively generate glucose-responsive IPCs from BMNCs. In this study, we identified a unique combination of 4 DIFs: Put, GlcN, nicotinamide, and BP-1-102. We formulated an efficient 6-day protocol to generate IPCs from BMNCs without progressively changing the DIFs or culture media. Our simplified approach successfully differentiated BMNCs into definitive endoderm (cells expressing Cxcr4, Sox17, and FoxA2), followed by pancreatic endoderm (cells expressing Pdx1 and Nkx6.1) and endocrine progenitors (cells expressing Nkx6.1, Neurog3, and NeuroD1). Functional BMNC-derived IPCs were also generated under the aforementioned culture conditions.

The expression of beta cell development-associated genes and insulin production were confirmed using qPCR, immunofluorescence staining, and FACS analysis. Furthermore, we confirmed the functionality of the IPCs generated in vitro by measuring insulin secretion in response to a glucose challenge and the in vivo effect of the 4 DIFs by demonstrating alleviation of diabetes upon introducing the BMNC-derived IPCs under the kidney capsules of hyperglycemic mice. These results affirm that the bone marrow contains pluripotent cells capable of being programmed into functional IPCs in vitro and in vivo. A single treatment of Put or GlcN is involved in GSIS. Park et al.²⁸ demonstrated that Put’s micromolar concentrations in MIN-6 increased basal and glucose-stimulated insulin secretion owing to increased MafA expression, which is involved in insulin production. GlcN supplementation also potentiates the differentiation of ADSCs into glucose-responsive IPCs.²⁰ Adding GlcN at any stage during the 12-day stepwise differentiation significantly increased the expression of beta cell-specific genes and insulin secretory genes, thereby increasing insulin secretion. Contrarily, adding it at a later differentiation stage affected gene expression and insulin secretion, indicating that GlcN might affect endocrine lineage commitment and beta cell maturation. Here, we revealed that co-treatment with Put and GlcN substantially affects the expression of beta cell markers and insulin promoter activation using BMNCs derived from MIP-Luc/GFP mice.

Additionally, the specific in vitro culture conditions combined with nicotinamide and BP-1-102 enhance insulin secretion under basal and glucose-stimulated conditions. Moreover, we observed increased expression of genes associated with the insulin secretion machinery (Glut2, Gck, and Syt) in differentiated IPCs. Glucose transporters, such as GLUT2 and GCK, are essential regulators of glucose responses in beta cells. The expression of these regulators in BMNC-derived IPCs suggests that IPCs respond to glucose. SYT is also involved in exocytosis during insulin secretion when insulin granules are fused to the surface of the plasma membrane.²⁹ Therefore, the increase in insulin secretion and dose-dependent glucose responsiveness of IPCs differentiated using these factors are caused by increased expression of molecules involved in the insulin secretion machinery. Consistent with these findings, our study demonstrated that insulin secretion was drastically increased in IPCs differentiated in the presence of DIFs, including Put and GlcN.

Nicotinamide has been used for pancreatic endocrine differentiation.²¹ Jiang et al.³⁰ established that nicotinamide is a specific differentiation regulator of NEUROG3⁺ islet progenitor cells. Nicotinamide-regulated beta cell differentiation required a shorter period of 4-6 days. The transcriptome profile of differentiated IPCs on day 4 exhibited a correlation of 94% with the transcriptome of adult islets. Consistent with this finding, our study revealed a robust increase in the expression of insulin and other beta cell markers, as demonstrated using qPCR and RT-PCR. Thus, simultaneous exposure to Put, GlcN, and nicotinamide optimized insulin expression on day 6. STAT3 helps maintain cellular identities and differentiate various cell types, including those of the immune, nervous, and endocrine systems.^31,32 NEUROG3 is a critical downstream effector of STAT3-regulated differentiation of mammalian stem and progenitor spermatogonia.³³NEUROG3, a pro-endocrine gene, is an essential transcriptional regulator that determines pancreatic endocrine fate. Nonetheless, its expression is reduced at the mature beta cell stage.³⁴ A recent study by Miura et al.²² demonstrated that STAT3 inhibition and exogenous expression of PDX1, NEUROG3, and MAFA promote cellular reprogramming into beta cells. Particularly, the small molecule STAT3 inhibitor BP-1-102 and genetic deletion of STAT3 significantly increased the efficiency of beta cell formation in vivo and ameliorated hyperglycemia in diabetic mice. Similarly, IPC maturation from primed BMNCs up to day 9 was accompanied by increased insulin expression, reduced Neurog3 expression, and increased or maintained Pdx1, MafA, NeuroD1, and Nkx6.1 expression levels. Therefore, combining nicotinamide and BP-1-102 effectively promoted further maturation of BMNC-derived IPCs in our experimental system.

Conclusion

We formulated a simple and efficient protocol using a unique combination of 4 extrinsic factors—Put, GlcN, nicotinamide, and BP-1-102—to differentiate BMNCs into functional IPCs that secret physiologically active insulin. BMNCs primed with these factors in vitro followed sequential developmental pathways through intermediate physiological cells. Moreover, our in vitro and in vivo findings indicate that the bone marrow includes pancreatic progenitor cells capable of differentiating into functional endocrine cells. Approximately 30% of the cells were mature IPCs secreting insulin in response to glucose stimulation. Evidence suggests that producing IPCs from stem cells is feasible and promising. However, variations in stem cell sources and induction techniques pose a challenge in the absence of a standard protocol. Therefore, the hypoglycemic effect of the 4 DIF-primed BMNCs implanted in hyperglycemic mice may mitigate this challenge. Furthermore, it is essential to understand the mechanisms involved in the extrinsic factor-mediated differentiation process to facilitate the maturation and differentiation of BMNCs into functional beta-like cells. Therefore, our results suggest that combining the 4 DIFs established here can promote the differentiation of BMNCs into functional IPCs. These results can be utilized to develop a novel approach for cell-based therapy for diabetes mellitus.

Funding

This study was supported by grants from the Korea Health Technology R&D Project “Strategic Center of Cell and Bio Therapy for Heart, Diabetes & Cancer” through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare (MHW) (HI 17C 2085) and Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2020R1A6A1A03047972), Republic of Korea. A grant of Yangyoung Foundation, Republic of Korea, also supported this research. This research was supported by the BK21FOUR Program of the National Research Foundation of Korea (NRF) funded by the Ministry of Education (5120200513755).

Conflict of Interest

The authors declared no potential conflicts of interests.

Author Contributions

S.-A.L.: conception and design, collection and assembly of data, data analysis and interpretation, manuscript writing; S.K., S.Y.K., J.Y.P., J.N., H.S.P., H.L., Y.D.L., and H.L.: collection and/or assembly of data, data analysis and interpretation; S.K., H.S.J., and S.S.C.: data analysis and interpretation; K.S.P.: conception and design, financial support, administrative support, data analysis and interpretation, manuscript writing, final approval of the manuscript.

Data Availability

All data generated or analyzed during this study are included in this published article (and its supplementary information files).

Ethics Statement

All experiments involving mice were approved by the Institutional Animal Care and Use Committee of Seoul National University (Authorization No. SNU-200911-3-2).