In vitro -derived insulin-producing cells modulate Th1 immune responses and induce IL-10 in streptozotocin-induced mouse model of pancreatic insulitis

Gholmrez Dryor ,Esmeil Hshemi Shiri ,Zhr Amirghofrn , ,Eskndr Kmli-Srvestni , ,

a Autoimmune Diseases Research Center, Medical School, Shiraz University of Medical Sciences, Shiraz 71345-1798, Iran

b Department of Immunology, Medical School, Shiraz University of Medical Sciences, Shiraz 71345-1798, Iran

Keywords:Immunomodulation Insulitis Insulin-producing cells Mesenchymal stem cells Streptozotocin Th1 cell

ABSTRACT Background: Insulitis is defined by the presence of immune cells infiltrating in the pancreatic islets that might progress into the complete β-cell loss.The immunomodulatory properties of bone marrow-derived mesenchymal stem cells (BM-MSCs) have attracted much attention.This study aimed to evaluate the possible immunomodulatory effects of rat BM-MSCs and MSCs-derived insulin-producing cells (IPCs) in a mouse model of pancreatic insulitis.Methods: Insulitis was induced in BALB/c mice using five consecuti ve doses of streptozotocin.MSCs or IPCs were directly injected into the pancreas of mice and their effects on the expression of Th subsetsrelated genes were evaluated.Results: Both BM-MSCs and IPCs significantly reduced the expression of pancreatic Th1-related IFN- γ(P ＜ 0.001 and P ＜ 0.05,respectively) and T-bet genes (both P ＜ 0.001).Moreover,the expression of IL -10 gene was significantly increased in IPC-treated compared to BM-MSC-or PBS-treated mice (P ＜ 0.001 both comparisons).Conclusions: BM-MSCs and IPCs could successfully suppress pathologic Th1 immune responses in the mouse model of insulitis.However,the marked increase in IL-10 gene expression by IPCs compared to BM-MSCs suggests that their simultaneous use at the initial phase of autoimmune diabetes might be a better option to reduce inflammation but these results need to be verified by further experiments.

Introduction

Pancreatic insulitis,the autoimmune destruction of the insulinproducingβcells of the islets of Langerhans,is the initial phase of type 1 diabetes [1] .During inflammation,direct contact ofβcells with activated immune cells and their pro-inflammatory cytokines induceβcells apoptosis [2] .Several studies have proposed a direct role for Th1 cells in the pathogenesis of type 1 diabetes [3-5].It has been shown thatinterferon-γ(IFN-γ) gene expression under the control of the human insulin promoter is sufficient to induce diabetes in mice while its blockade in nonobese diabetic mice could prevent diabetes [6,7].Accordingly,immunosuppressive agents can prevent pancreaticβ-cell destruction [8] .However,such drugs cause various side effects that have limited their routine use [9,10].In recent years,stem cell-based therapies have been proposed as a promising approach for the treatment of inflammatory diseases [11] .Mesenchymal stem cells(MSCs) are immunomodulatory cells capable of preserving residualβ-cell mass from immune attack and providing a proper microenvironment for endogenousβ-cell regeneration [12,13].They also have the potential to differentiate into multiple cell types such as glucose-responsive insulin-producing cells (IPCs).We and several groups have successfully differentiated MSCs into functional IPCsin vitro[14-16].However,theinvivoimmunomodulatory properties of MSCs-derived IPCs have not been well addressed.Accordingly,in the present study,the immunomodulatory effects of IPCs were compared with those of bone marrow (BM)-MSCs in the mouse model of insulitis.

Materials and methods

MSCs isolation, culture, characterization, and differentiation toward IPCs

Four-week-old male Sprague-Dawley rats were purchased from the Center of Comparative and Experimental Medicine.BM-MSCs were isolated,cultured,and characterized by analyzing their surface markers using flow cytometry (CD34-,CD45-,CD106-,CD44 +,CD90 +,and CD105 +) and differentiating them toward adipocytes and osteocytes [16] .A hydrogel-based two-step method was used to generate IPCs [16] .Briefly,about 2 × 106BM-MSCs in DMEM complete medium were seeded on 1% agar coated dishes and maintained in a non-adherent state for 7 days.The media was changed every two days.Cellular clusters were then transferred into plates containing DMEM-high glucose (25 mmol/Lα-D glucose) medium supplemented with 10% FBS,1% antibiotic solution,1 mmol/L sodium pyruvate,10 mmol/L nicotinamide,and 0.1 mmol/L beta-mercaptoethanol for another 7 days [16] .

Induction of insulitis in mice

Multiple low doses of streptozotocin (STZ) were used to induce insulitis in 6-8 weeks old female BALB/c mice [17-19].Briefly,STZ powder was dissolved in 0.1 mol/L citrate buffer and was immediately injected intraperitoneally into 6 h fasted mice at a dose of 50 mg/kg.Similar injections were done for five consecutive days while control mice received only 0.1 mol/L citrate buffer.One week after the last STZ injection,tail vein blood was collected and fasting blood glucose level was determined using the Accu-Chek Active Blood Glucose Meter (Roche Diagnostics GmbH,Mannheim,Germany).Mice with two consecutive blood glucose levels of＞250 mg/dL were considered to have developed insulitis.Besides,the possible effects of STZ on the pancreas of treated mice was evaluated using hematoxylin-eosin (HE) staining.

MSC and IPC transplantation

To inject the cells into the pancreas,mice were anesthetized with a single intramuscular injection of ketamine (100 mg/kg)and xylazine (10 mg/kg).The pancreas was exposed through an abdominal incision (laparotomy) and prepared cell suspensions(2 × 106BM-MSCs or IPCs in 100 μL PBS) were directly injected into the pancreas using a 30 G needle (Insumed,Kraków,Poland).The control group received 100 μL PBS.

Survival of IPCs in the pancreas

IPCs were stained with DiD and DAPI (4 ′,6-Diamidino-2-phenylindole dihydrochloride) fluorescent dyes,which stain cell membrane and nucleic acids,respectively.Briefly,IPCs were trypsinized,counted,and 1 × 106cells in 1 mL PBS were used for staining.FiveμL of 1 mmol/L DiD solution was added to the cell suspension,cells were mixed well by gentle pipetting,and were incubated for 30 min at 37 °C.DiD-labeled IPCs were washed two times (400 ×g,5 min,23 °C) with PBS.After DiD labeling,the DAPI solution was added at a final concentration of 20 μg/mL.Cells were incubated for 30 min at 37 °C and subsequently,were washed with PBS (400 ×g,5 min,23 °C).To evaluate the staining protocol,a fraction of stained cells was transferred to a microscopic slide and the presence of double-stained IPCs was confirmed using fluorescent microscopy (Fig.1).To monitor the survival of IPCsinvivo,2 × 106cells were prepared in 100μL PBS and the mixture was directly injected into the pancreas of each insulitis mouse.Treated mice were sacrificed at 2,7,and 20 days after injection (5 mice at each time-point) prepared frozen sections were stained with HE and the presence of double-stained IPCs was evaluated using fluorescent microscopy (Olympus BX61,Olympus,Shinjuku-Ku,Tokyo,Japan).To do this,one frozen section from day 2 post-injection was used as the positive control and considered as 100% positive for DiD and DAPI.The mean intensity of both dyes was calculated and considered as the total fluorescent intensity.The fluorescent intensity of other frozen sections (2,7,and 20 days) was quantified in comparison with the positive control in three independent blind experiments by 3 individuals.

Analysis of pancreatic T-cell populations

Fifty days post cell injection,total RNA was extracted from mouse pancreas according to the instructions provided by the manufacturer (Parstous Biotechnology,Mashhad,Iran).The purity and concentration of RNA were determined using a UV spectrophotometer (Pico 100 μL Spectrophotometer,Picodrop Limited,Hunxton,UK) while its integrity was evaluated by running samples on 1.2% agarose gel.Afterward,cDNA was synthesized using the high capacity cDNA reverse transcription kit (ABI,Foster City,CA,USA).Subsequently,real-time PCR was done in triplicate in a total reaction volume of 20μL for each condition using the StepOne Real-Time PCR System (ABI).Briefly,2 μL cDNA,10 pmol/L of forward and reverse primers,and 10 μL 2 × SYBR Premix Ex Taq II(Takara,Tokyo,Japan) were used for each reaction.Genes of interest,primer sequences and primers annealing temperatures are listed in Table 1 .Every 40 cycles of PCR consisted of denaturation(5 s at 95 °C),annealing (15 s at 60 °C),and extension (30 s at 72 °C).The relative expression of the target gene was calculated using the 2-ΔΔCtcomparative method.Normalization was done againstβ-actin threshold cycle (Ct) values.The data were presented as the fold change of the target gene expression normalized byβ-actin and relative to the reference sample (e.g.normal mouse pancreas).

Table 1 Primer sequences of cytokines and transcription factors for evaluating T cell subsets in mouse pancreas.

Statistical analysis

Data were analyzed using SPSS version 25 (SPSS Inc.,Chicago,IL,USA) and GraphPad Prism version 7 (GraphPad Software Inc.,San Diego,CA,USA) software programs.Data of the results were represented as mean ± standard deviation (SD) of at least three independent experiments.Analysis of variance (ANOVA) followed by Tukey’s post-test was used to test the probability of significant differences between groups.APvalue of less than 0.05 was considered statistically significant.

Results

Rat BM-MSCs injection to the pancreas of normal mice had no adverse effect

Rat BM-MSCs were directly injected into the pancreas of five normal mice to examine any possible anomaly.For the sham group,five normal mice received PBS.Fourteen days after injection,animals were sacrificed and histological evaluation of the pancreas revealed no sign of inflammation nor pathologic changes compared to the pancreas of normal mice (Fig.2).

Confirmation of the mouse model of insulitis

HE staining revealed severe necrotic and degenerative changes in the pancreatic islets of STZ-treated mice (Fig.3).Considering high levels of blood glucose and the pathological changes of the pancreatic islets it can be conferred that lots of insulin-producingβcells have been destroyed.

Fig.1.IPCs labeling with fluorescent dyes DiD and DAPI.Light microscopy image (A),fluorescent images using DAPI (B) or DiD (C) filters,and merged images (D) of IPCs after labeling are shown (original magnification × 100).

IPCs survived in the pancreas of the mouse model of insulitis

Fluorescent microscopy revealed that IPCs were not rejected after transplantation for at least 20 days.The intensity of the fluorescent dyes was compared (Fig.4 A-D) and the results showed no significant differences between different time points (at day 2:79.5% ± 2.0%,day 7:63.1% ± 4.4%,day 20:65.7% ± 6.7%,respectively,P=0.17;Fig.4 E).Herby,IPCs survived in the pancreas of the mouse model of insulitis.

Effects of BM-MSCs and IPCs on pancreatic T cell subsets

Gene expression analyses of cytokines and transcription factors related to Th1 [IFN-γand T-box expressed in T cells (T-bet)],Th2 [IL-4and GATA binding protein (GATA)-3],Th17 [IL-17and retinoid-related orphan receptor (RORC)],and Treg [(IL-10and forkhead box P3 (FOXP3)] were done on the pancreas of mice at the end of the experiment.IL-4,IL-17,andFOXP3genes were not detectable (Data are not shown.).However,the expression levels ofIFN-γandT-betgenes were significantly reduced in the pancreas of IPC-or BM-MSC-treated mice compared to those of the PBS control group (forIFN-γgene:2.3 ± 0.8 fold,1.0 ± 0.3 fold,and 5.0 ± 2.6 fold,respectively,P=0.049 andP＜0.001,respectively,Fig.5 A;forT-betgene:3.9 ± 2.4 fold,1.7 ± 0.7 fold,and 56.8 ± 30.2 fold,respectively,bothP＜0.001,Fig.5 B).Of interest,IPC-treated mice had significantly higher expression levels ofIL-10compared to BM-MSC-or PBS-treated mice (2.2 ± 1.4 fold,0.01 ± 0.004 fold,and 0.02 ± 0.02 fold,respectively,P＜0.001 for both comparisons,Fig.5 C).Furthermore,the ratios of Th1/Treg related genes (IFN-γ/IL-10),as well as Th1/Th2 related transcription factors (T-bet/GATA-3) were significantly reduced in the pancreas of IPC-or BM-MSC-treated mice compared to those of the PBS control group (forIFN-γ/IL-10:1.4 ± 0.8 fold,72.5 ± 38.8 fold,and 273.0 ± 168.1 fold,respectively,P=0.002 andP=0.006,respectively,Fig.5 F;forT-bet/GATA-3:4.3 ± 6.5 fold,1.2 ± 0.3 fold,and 58.5 ± 32.2 fold,respectively,P=0.0013 andP=0.0001,respectively,Fig.5 G).Although IPCs-or BM-MSC-treated mice expressed higher pancreatic levels ofGATA-3but lower levels ofRORCgenes compared to PBS-treated mice,these differences were not statistically significant (Fig.5 D,E).

Fig.2.HE staining of pancreas tissue isolated from normal mice (A),normal mice received BM-MSCs (B) and normal mice received PBS (C) (original magnification × 100).

Fig.3.HE staining of pancreas tissue isolated from mice treated with citrate buffer (A) and mice treated with streptozotocin (B) (original magnification × 100).

Discussion

Fig.4.Tracking of IPCs in the pancreas of the mouse model of insulitis (A-D) (original magnification × 100).Light microscopy image of HE stained pancreatic tissue of diabetic mouse treated with labeled IPCs (A),fluorescent images using DAPI (B) or DiD (C) filters,and merged image (D) were prepared for tracking of injected IPCs in the pancreas.E:Comparison between pancreatic sections of mice at 2,7,and 20 days post-IPC injection.

Insulitis which is the infiltration of the immune cells into the islets of Langerhans during the early stage of type 1 diabetes promotes immune responses against pancreaticβcells and facilitates their apoptosis [20] .MSCs due to their immunomodulatory and anti-inflammatory properties are widely used in the cell therapy of inflammatory diseases [21,22].They have been shown to suppress the proliferation and IFN-γinduction by stimulated T cells [23,24].In a mouse model of arthritis,the injection of human adipose-derived stem cells at days 18 and 24 before the onset of disease decreased the number of Th1 and Th17 cells but increased the number of IL-10 producing lymphocytes [25,26].Similar results were found in human MSC-treated mouse models of colitis and sepsis [27,28].Human BM-MSCs and human adipose-derived stem cells have also successfully reduced the numbers of Th1 and Th17 cells but increased the numbers of Th2 cells and anti-inflammatory cytokines in both chronic and relapsing-remitting mouse models of multiple sclerosis [29,30].Furthermore,human MSCs have been shown to increase the frequency of Tregs,enhance the expression of IL-4,IL-10 and transforming growth factor (TGF)-β1,and reduce plasma levels of IFN-γ,IL-17A and tumor necrosis factor(TNF-α) [31,32].Besides,MSCs can transdifferentiate into mesodermal,ectodermal,and endodermal cell lineages [33] .It has been shown that transdifferentiated MSCs into specialized cell types such as IPCs could still maintain their immunosuppressive potentialinvitro[34] .However,invivoeffects of IPCs on the immune system have not been well acknowledged.Gene expression analysis can implicate the presence of immune cells within the pancreas of the STZ-induced mouse model of insulitis.It is worth mentioning that detecting cytokines at protein level by techniques such as ELISA,immunohistochemistry,and flow cytometry could confirm our results but due to the limitations,only the expression of T cells related genes was evaluated in the pancreas of mice.The present study showed that both BM-MSCs and IPCs can successfully modulate the expression of Th1-related genes within the inflamed pancreas.However,BM-MSCs more successfully suppressedIFN-γandT-betgene expressions compared to IPCs.Of interest,it was shown that IPCs can significantly increase the expression ofIL-10gene in the inflamed pancreas compared to BM-MSCs.In support,a study using 3D hanging drop culture method has generated IPCs from human MSCs that express significantly higherIL-10gene levels [35] .Accordingly,the enhanced ability of our IPCs in the induction ofIL-10gene within the pancreas of mice might be due to the initial 3D culture condition on 1% agar.IL-10 is an immunoregulatory cytokine that is mainly produced by Treg cells and protects the host from unwanted immune responses and autoimmunity [36] .IL-10 deficiency aggravates autoimmunity in several experimental models of autoimmune diseases [37-39].Li et al.[40] have shown that intraperitoneal injection of theadenoviralvector-mediatedIL-10(Ad-mIL-10) gene in non-obese diabetic mice can increase serum and pancreatic levels of IL-10 that protectβcell functions.Therefore,Th1 cell modulation and Th2 or Treg induction during insulitis might alleviate the inflammation.

Fig.5.Effects of BM-MSCs and IPCs on the expression of T cell subsets-related genes in the pancreas of a streptozotocin-induced mouse model of insulitis.The expression of IFN- γ (A) and T-bet (B) genes were significantly reduced in BM-MSCs or IPCs-treated mice but IL-10 (C) gene expression was increased in IPC-treated mice compared to the control PBS and BM-MSC-treated mice.However,the expression of GATA-3 (D) and RORC (E) genes were not significantly affected by BM-MSCs or IPCs.Th1/Treg (IFN- γ/ IL-10)(F) and Th1/Th2 (T-bet/ GATA-3) (G) ratios were also significantly reduced in the pancreas of treated mice compared to those of the control PBS.* P ＜ 0.05,** P ＜ 0.01,*** P＜ 0.001.RQ:relative quantification.

In conclusion,our protocol of IPC induction can generate much higher levels of IL-10 in the inflamed pancreas compared to the BM-MSCs induction.Accordingly,it can be concluded that the simultaneous use of BM-MSCs and IPCs (generated with our protocol) might be a better approach for the treatment of insulitis at the initial stages but these results need further verifications.

Acknowledgments

We thank Mrs.Farzaneh Taki (Autoimmune Diseases Research Center,Shiraz University of Medical Sciences) for her help to accomplish the goals of the present study and Ms.Yasamin Kamali-Sarvestani for English editing.

CRediTauthorshipcontributionstatement

GholamrezaDaryabor:Data curtion,Formal analysis,Writingoriginal draft,Writing -review &editing.EsmaeilHashemiShiri:Methodology.ZahraAmirghofran:Writing -review &editing.EskandarKamali-Sarvestani:Conceptualization,Funding acquisition,Supervision,Writing -review &editing.

Funding

This study was supported by a grant from Shiraz University of Medical Sciences (No.94-7616).

Ethicalapproval

This study was approved by the Ethics Committee of Shiraz University of Medical Sciences (No.1394.S676).

Competinginterest

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.