Abstract
Porcine mesenchymal stem cells (MSCs) are similar to human MSCs, hence considered a valuable model for assessing potential for cell therapy. Porcine adipose-derived MSCs (AD-MSCs) and endometrial stromal MSCs (EMSCs) displayed fibroblast-like morphology and were positive for MSC markers CD73, CD90, and CD105 and negative for hematopoietic markers CD34 and CD45. The EMSCs had similar or slightly higher growth rate compared to AD-MSCs, and similar percentage of cells of both EMSCs and AD-MSCs were at G0/G1 and G2/M phases; however, EMSCs had significantly (P < .05) higher percentage of cells at S phase of cell cycle than AD-MSCs. Transdifferentiation ability to cardiomyocyte-like cells was confirmed in differentiated cells by the expression of lineage-specific marker genes such as DES, ACTA2, cTnT, and ACTC1 by real-time quantitative polymerase chain reaction (RT-qPCR). Furthermore, cardiomyocyte-specific protein markers cTnT and ACTC1 were expressed in completely differentiated cells. Endodermal differentiation capacity of EMSCs to pancreatic β cell-like cells was evident with the changes in morphology and the expression of β-cell–specific marker genes such as PDX1, GLUT2, SST, NKX6.1, PAX4, and NGN3 as analyzed by RT-qPCR. The differentiated cells secreted insulin and C-peptide upon glucose challenge and also they expressed insulin, PDX1, PAX4, NGN3, and GLUT2 at protein level as assessed by immunostaining confirming the successful differentiation to β cell-like cells. Porcine EMSCs possess all the characteristics of MSCs and are suitable model for studying molecular mechanisms of cellular differentiation.
Keywords
Introduction
In the recent years, stem cell therapy is gaining more importance in treating many diseases including heart disease and vastly tested for application in regenerative medicine. Porcine organs and cells display similarities to human both anatomically and genetically and hence have gained importance in various fields of biomedical research. 1 Mesenchymal stem cells (MSCs) are adult stem cells with a capacity of self-renewal and differentiation to multilineage. According to International Society for Cellular Therapy (ISCT), MSCs are plastic-adherent multipotent cells with a potential to differentiate into osteocytes, adipocytes, and chondrocytes. 1 Although there are no specific cell surface makers that are universally applicable to all MSCs isolated from different adult tissues, according to ISCT, the MSCs should express CD73, CD90, and CD105 while negative for CD11b, CD14, CD34, CD45, CD79α, and HLA-DR. 2
In porcine, MSCs are successfully isolated from bone marrow, 3 adipose tissues, 4 umbilical cord Wharton jelly, 5 and umbilical cord blood. 6 Bone marrow-derived MSCs (BM-MSCs) are most commonly studied MSCs for their strong stem cell characteristics, 7 despite the invasive procedure and repetitive collection. It was previously reported that porcine adipose tissue is a good source of adult stem cells with a great potential for application in cell therapy and regenerative medicine. 8 Alternatively, endometrium is a highly regenerative tissue consisting of upper functionalis comprising epithelial cells and lower basalis harboring stromal cells, endothelial cells, leukocytes, and blood cells. 9 It has been previously reported that basalis stromal region in porcine possesses MSCs 10 having potential to differentiate into mesenchymal lineages as well as to transdifferentiate into immature neuron-like cells with functional electrophysiological properties. 11
Porcine MSCs are known to differentiate into cardiomyogenic lineage both in vivo and in vitro. 1 5-Azacytidine (5-AZA), a DNA methylation inhibitor, is commonly used to induce cardiomyogenic differentiation in MSCs. 12,13,7 DNA methylation is generally associated with gene expression, chromatin modification, X chromosome inactivation, endogenic gene silencing, and genomic imprinting. 14 Nevertheless, DNA methylation also plays a major role in maintaining pluripotency and self-renewal of stem cells. Hypomethylation marks the active state whereas hypermethylation inactive state of genes. When pMSCs were treated with 5-AZA, it resulted in 30% to 50% of cells with cardiomyocyte-like characteristics expressing lineage-specific markers such as T-trophonin, desmin, and connexin-3. 15 Although the use of 5-AZA has shown to induce cardiac-like phenotypes in MSCs by activating a number of genes, its effect may not be specific, indicating that the treatment with demethylating agents may not be sufficient to reprogram adult MSCs to the cardiogenic lineage, and there might be other factors supplied by the cardiac niche that influence cardiac-lineage commitment. 16 Hence, 5-AZA alone cannot efficiently induce MSCs into cardiomyocytes 17 and requires additional growth factors and cytokines for successful differentiation. In recent studies, angiotensin-II was found to initiate cardiomyocyte differentiation in rat BM-MSCs, 18 which when further treated with transforming growth factor-β1 (TGFβ1) transformed the cells efficiently into cardiomyocyte-like cells. 19 Therefore, in this study, we have used 5-AZA, angiotensin-II, and TGFβ1 for the differentiation of porcine endometrial stromal mesenchymal stem cells (EMSCs) into cardiomyocyte-like cells.
During diabetes, insulin secretion is interrupted, and this insulin deficiency can be restored either by transplantation of ex vivo generated β cells or regeneration of endogenous β cells. 20 Generally, patients with either type 1 or type 2 diabetes mellitus are unable to maintain their normal plasma glucose level, 21,22 and if untreated, this may lead to severe complications such as diabetic nephropathy, diabetic retinopathy, neuropathy, cardiovascular disease, and cataract. The major goal in the treatment of diabetes mellitus is to maintain the normal plasma glucose level, and therefore insulin therapy has emerged as most promising along with other medications such as dietary maintenance and regular exercise depending on the patient’s underlying health condition. Although insulin therapy is more promising, it does not match the precision of functioning β cells. 23 Therefore, an immediate alternative approach could be islet transplantation. However, the shortage of donor islets made this approach unattractive. 24 In this context, there is a growing interest in stem cell therapy which could replace the dysfunctional β cells to cure diabetic patients from insulin deficiency. 23 Previously, it was reported that adult tissue-derived MSCs can be differentiated to β cells and can be transplanted in vivo. 25–26 However, many of these differentiation protocols involve gene transfection and/or genetic manipulation hindering the clinical application of these differentiated cells. Previous study shows MSCs from human endometrium upon induction differentiated into glucose-responsive, insulin-producing cells and were similar to endogenous islets. 27,28
The present study was designed to characterize and compare the porcine EMSCs and adipose-derived mesenchymal stem cells (AD-MSCs) based on their basic stem cell characteristics and mesenchymal lineage differentiation ability. Further, we assessed the efficiency of EMSCs to transdifferentiate into cardiomyocyte-like cells and insulin-producing β cell-like cells.
Materials and Methods
All animal samples were collected after the approval of Research Ethical Committee of Gyeongsang National University, Animal Center for Biomedical Experimentation under set guidelines (GNU-140305-P0016).
Chemicals and Media
All chemicals were procured from Sigma Chemicals Company (St Louis, Missouri) and media, serum, and buffers from Gibco (Invitrogen, Burlington, Ontario, Canada) unless otherwise specified.
Isolation and Culture of Porcine EMSCs and AD-MSCs
Porcine (Sus scrofa) uterus (n = 6) and adipose tissue from nonpregnant 6-month-old female were collected at local slaughter house and transported to laboratory on ice-cold Dulbecco phosphate-buffered saline (DPBS) within 1 hour of collection. The EMSCs were isolated following previously reported protocol. 11 The AD-MSCs were isolated from porcine abdominal subcutaneous adipose tissues as reported previously. 4 Following the isolation, cells were expanded under in vitro culture conditions. Once the cells reached confluence, they were subcultured into 1:4 ratio in advanced Dulbecco modified Eagle’s medium (ADMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin (Pen-Strep; 10 000 IU and 10 000 µg/mL, respectively). All further experiments were carried out using cells from passage 3 or later. For transdifferentiation of EMSCs into cardiomyocyte-like cells and insulin-producing β cell-like cells, the CD105+ cells were sorted using flow cytometer and used.
Flow Cytometric Analysis
Both EMSCs and AD-MSCs were characterized for their mesenchymal markers employing flow cytometry (BD FACS Calibur, Becton Dickinson, NJ) as reported previously. 29,11 Briefly, the cultured cells at ∼80% confluence were fixed with 3.7% formaldehyde and labeled (1 × 105 cells per marker) with fluorescein isothiocyanate (FITC)-conjugated CD34, CD45, and CD90 and FITC-unconjugated CD73 and CD105 for 30 minutes in dark. Unconjugated primary antibodies were treated with FITC-conjugated secondary antibodies for 30 minutes in the dark. The FITC-conjugated mouse immunoglobulin G1 (IgG1) served as isotype-matched negative control.
For analysis of the percentage of cells expressing insulin after differentiation into insulin-producing β cell-like cells, the cells were fixed with 3.7% formaldehyde and labeled (1 × 105 cells) with insulin primary antibody for 30 minutes followed by FITC-conjugated secondary antibody for another 30 minutes in the dark. The FITC-conjugated rabbit IgG served as isotype-matched negative control. The details of antibodies are listed in Table 1.
Details of Antibodies Used for Flow Cytometric Analysis.
Abbreviation: IgG, immunoblobulin G.
A total of 10 000 labeled cells were acquired and measured using BD FACS Calibur flow cytometer, and the results were analyzed by CellQuest Pro software, Version 5.1.
The cell cycle was analyzed by fixing 1 × 106 cells/mL in 70% ethanol at 4°C for overnight. The cells were then washed twice with DPBS and stained with propidium iodide solution (10 μg/mL) for 15 minutes. DNA content was then measured by flow cytometer and categorized as G0/G1, S, or G2/M phase of the cell cycle.
Cell Proliferation Assay
The proliferation rate of EMSCs and AD-MSCs was analyzed by 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay according to manufacturer’s protocol. Briefly, 1 × 103 cells at passage 4 were seeded into 12-well plates in 1 mL ADMEM supplemented with 10% FBS and 1% Pen-Strep. The MTT assay was performed by adding 200 µL of 12 mmol/L MTT stock solution into each well containing 1 mL media and incubated at 37°C for 4 hours. After removing all MTT solutions, 100 µL of dimethyl sulfoxide (DMSO) was added to dissolve the insoluble formazan formed from viable cells after metabolizing MTT and incubated for 10 minutes. The product formed in each well was transferred into 96-well plate, and subsequently the absorbance was measured at 570 nm using Versa Max microplate reader (Molecular Devices, LCC, SJ, CA, USA).
Differentiation Into Adipocytes, Osteocytes, and Chondrocytes
In vitro differentiation of EMSCs and AD-MSCs into adipocytes and osteocytes were carried out using previously published protocol. 11 Briefly, adipogenesis was induced with media containing 1 µmol/L dexamethasone, 10 µmol/L insulin, 100 µmol/L indomethacin, and 500 µmol/L isobutylmethylxanthine for 21 days. Adipocyte differentiation was confirmed by the formation of lipid droplets by staining with Oil Red O and the expression of adipocyte lineage-specific marker genes. Osteogenesis was induced with media containing 10 mmol/L sodium β–glycerophosphate, 0.05 mmol/L ascorbic acid, and 0.1 µmol/L dexamethasone for 21 days. The mineralization was detected by staining with von Kossa and alizarin red. Further, osteogenesis was confirmed by the expression of osteogenic lineage-specific marker genes. Both EMSCs and AD-MSCs were induced to chondrogenic differentiation with StemPro osteocyte/chondrocyte differentiation basal medium (StemPro chondrogenesis supplement; Gibco) for 21 days, media were changed twice a week. Differentiated cells were stained with Alcian blue (1%) and safranin O to confirm the formation of proteoglycans. Additionally, the chondrogenesis was confirmed by the expression of chondrogenic lineage-specific marker genes.
Induction and Differentiation to Cardiomyocyte-Like Cells
The EMSCs were differentiated into cardiomyocyte-like cells using 5-AZA, angiotensin-II (A9525, Sigma), and TGFβ1 (PHG9204, Gibco) following previously reported protocol with modification. 18,7,19 Briefly, cells at 60% confluence were induced with 0.1 µmol/L angiotensin-II and 10 µmol/L 5-AZA for 24 hours in ADMEM without FBS. After 24 hours induction, cells were washed 2 times with DPBS and treated with 10 ng/mL TGFβ1 in ADMEM containing 10% FBS for 20 days, and fresh media with TGFβ1 was replaced for every 3 days. Untreated cells maintained in ADMEM containing 10% FBS were used as control cells.
Induction and Differentiation to Insulin-Producing β Cell-Like Cells
For differentiation into insulin-producing β cell-like cells, a previously reported protocol for human adult stem cells 30,28 with minor modification was used. Briefly, the induction was carried out in 4 different steps.
Step 1: EMSCs at 70% confluence were cultured in 25 mmol/L DMEM containing 10% FBS and supplemented with 1 µmol/L retinoic acid for 24 hours.
Step 2: The induced cells were cultured for another 2 days in 25 mmol/L DMEM containing 10% FBS and dissociated with 0.25% trypsin–EDTA solution.
Step 3: Dissociated cells were plated on to a 6-well plate precoated with Geltrex LDEV-free membrane matrix and cultured in 5.6 mmol/L DMEM containing 10% FBS supplemented with 10 mmol/L nicotinamide, 10 ng/mL epidermal growth factor, and 300 nmol/L indolactam V for 9 days. The media were changed for every 3 days.
Step 4: In the final step, cells were cultured in 5.6 mmol/L DMEM containing 10% FBS supplemented with 10 mmol/L exendin-4 and 50 ng/mL Activin A for 7 days. Control EMSCs were maintained in 25 mmol/L DMEM containing 10% FBS.
Dithizone Staining of Differentiated Insulin-Producing β Cell-Like Cells
Differentiated β cell-like cells were stained with diphenylthiocarbazone (dithizone) to assess their insulin-producing ability. Briefly, 1 mg of dithizone was dissolved in 1 mL DMSO and further diluted to 1:10 in PBS to prepare working solution. Both the control and the differentiated cells (105) were treated with working solution of dithizone for 30 minutes at 37°C, washed thrice with DPBS, and observed under phase-contrast microscope for insulin granules (Nikon, Japan).
Glucose Challenge Assay for Insulin and C-Peptide Secretion
For estimation of insulin and C-peptide secretion from β cell-like cells, both control and differentiated β cell-like cells were washed twice with Krebs buffer (Krb) and incubated in low-glucose (5.6 mmol/L) Krb for 30 minutes, and the supernatant was collected. Then, the cells were washed twice in Krb buffer followed by incubation in high-glucose (25 mmol/L) Krb for 30 minutes, and the supernatant was collected. These steps were repeated 3 times, and the supernatant was stored at −20 C. The supernatant was assessed for insulin (Neoscientific, Massachusetts) and C-peptide (Sigma, Missouri) using enzyme-linked immunosorbent assay/enzyme immunoassay (ELISA/EIA) kits following manufacturer’s instructions. The absorbance was measured at 450 nm using ELISA plate reader (Molecular Devices). For C-peptide, the values were normalized to the DNA content. The DNA was extracted from the equal number of cells seeded at same time and same treatment using DNeasy tissue kit (Qiagen, Hilden, Germany) according to manufacturer’s protocol.
RNA Extraction, Complementary DNA Synthesis, and Real-Time Quantitative Polymerase Chain Reaction Analysis
Total RNA was isolated from control undifferentiated and differentiated cells using an RNeasy mini kit (Qiagen) and quantified using UV BIO spectrophotometer OPTIZEN 3220 (Mecasys Co, Ltd, Korea), and quality was assessed using RNA gel. The reverse transcription of purified total RNA (2 μg) to complementary DNA (cDNA) was carried out using Omniscript reverse transcription kit (Qiagen) with random hexamer primer (Invitrogen) in a 20 μL reaction mixture at 37°C for 1 hour.
Real-time quantitative polymerase chain reaction (RT-qPCR) analysis was performed using Rotor Gene Q (Qiagen) machine with 50 ng cDNA quantified with Rotor-Gene 2X SYBR Green mix (Qiagen) supplemented with 1 µL each of forward and reverse primers at a final concentration of 400 nmol/L (Table 2). The RT-qPCR protocol comprised of predenaturation at 95°C for 10 minutes; 40 PCR cycles at 95°C for 10 seconds, 60°C for 6 seconds and 72°C for 6 seconds; melting curve from 60°C to 95°C by 1°C per second; and cooling at 40°C for 30 seconds according to manufacturer’s protocol. Rotor-Gene Q Series Software (Qiagen) was used to determine melting curves, amplification curves, and cycle threshold values (Ct values). The expression level of genes was normalized to HMBS expression. All products were run in triplicate and confirmed by agarose gel electrophoresis for nonspecific amplification with negative control.
RT-PCR and RT-qPCR Primer Sequences Specific to Porcine MSCs and Differentiated Cells.
Abbreviations: RT-PCR, real-time polymerase chain reaction; RT-qPCR, real-time quantitative polymerase chain reaction; MSCs, mesenchymal stem cells.
Immunocytochemistry of Differentiated Cells
Both control and differentiated cells into cardiomyocyte-like cells and insulin-producing β cell-like cells were analyzed by immunocytochemical staining following previously described protocol. 31 Briefly, both differentiated and control cells were washed with DPBS, fixed with 3.7% formaldehyde for 50 minutes, and permeabilized by 0.2% Triton X-100 supplemented with 2% bovine serum albumin (BSA) for 15 minutes. After blocking with 2% BSA in DBPS for 1 hour, the cells were incubated in primary antibody (1:100) at 4°C for 1 hour. After incubation with primary antibodies, cells were washed 3 times with DPBS, followed by incubation with conjugated secondary antibodies (1:100). Slides were then counterstained with 1 µg/mL 4′,6-diamidino-2-phenylindole for 5 minutes at room temperature (RT). Finally, slides were mounted with Vectashield (Vector Laboratories, Inc, California) and observed under fluorescence microscope (Leica, Wetzlar, Germany). The details of antibodies are presented in Table 3.
Details of Western and Immunofluorescence Antibodies Specific to Porcine MSCs.
Abbreviations: GAPDH, glyceraldehyde 3-phosphate dehydrogenase; HRP, horse radish peroxidase; IgG, immunoglobulin G; MSCs, mesenchymal stem cells.
Western Blot Analysis
The total protein was extracted with RIPA buffer (PIERCE, Illinois) containing protease inhibitor, quantified using BCA protein assay kit (PIERCE). A total of 20 µg protein from differentiated cells into cardiomyocyte-like cells, insulin-producing β cell-like cells, and untreated control EMSCs were separated on sodium dodecyl sulfate polyacrylamide gel electrophoresis at 100 V and wet transferred electrically onto a polyvinylidene fluoride (PVDF) membrane (Biorad, California) for 2 hours at 200 V. The PVDF membrane was then blocked with 5% BSA in Tris-buffered saline (TBS; 1 M Tris [pH 7.5], 5 N NaCl) for 1 hour at RT followed by a wash with TBST-0.1% (Tween 20). The protein blot was incubated with antibodies specific to cardiomyocyte or pancreatic lineage marker proteins (Table 3) in 0.1% TBST–5% BSA overnight at 4°C. Followed by 3 times washing with 0.1% TBST, the blot was incubated with horse radish peroxidase-conjugated secondary antibodies in 0.1% TBST–5% BSA for 1 hour at RT. The expression of proteins was detected using enhanced chemiluminescence (Supersignal West Pico chemiluminescent substrate, PIERCE) and then exposed to X-ray films (FUJI Photo Film Co, Ltd, Tokyo, Japan). Final data are representative of 2 individual cell culture experiments from different donors (n = 2).
Statistical Analysis
The statistical significance of difference among groups was analyzed by 1-way analysis of variance using SPSS 21.0 (SPSS Inc, Chicago, Illinois) followed by Tukey test. Values are presented as mean ± standard error of the mean. Differences were considered to be significant when P < .05.
Results
Cell Morphology and Cell Surface Markers
Both porcine AD-MSCs and EMSCs were successfully isolated from adipose and endometrium stromal tissues, and these cells exhibited fibroblast morphology upon forming colonies (Figure 1A). Both AD-MSCs and EMSCs were positive for MSC markers CD73 (100 [0.63] and 99 [0.45]), CD90 (88 [0.54] and 97 [0.58]), CD105 (77 [0.61] and 83 [0.53]), respectively, as analyzed by flow cytometer (Figure 1B). However, hematopoietic markers CD34 (7 [0.42] and 3 [0.54]) and CD45 (5 [0.29] and 1 [0.16]) were not detected in both MSC types.
A, Morphological characteristics of porcine adipose tissue- and endometrial stromal–derived mesenchymal stem cells; scale bar = 100 µm. B, CD markers analysis by flow cytometry; color-filled histogram represents specific surface marker and open histograms refers to isotype controls. EMSCs were strongly positive (>94%) for CD73, CD90, and CD105 and negative (<2%) for CD34 and CD45. C, Analysis of cell proliferation of EMSCs and AD-MSCs by MTT assay. D, Cell cycle analysis of ADMSCs and EMSCs evaluated by flow cytometry. EMSCs indicates endometrial stromal mesenchymal stem cells; AD-MSCs, adipose-derived mesenchymal stem cells; MTT, 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide.
Cell Proliferation and Cell Cycle Analysis
Cell proliferation assessed by MTT assay showed similar growth patterns for both AD-MSCs and EMSCs; however, EMSCs had significantly (P < .05) higher proliferation rate compared to AD-MSCs (Figure 1C).
The cell cycle analysis revealed that there was no significant difference (P < .05) between AD-MSCs (75.6 [2.8] and 10.9 [2.3]) and EMSCs (73.9 [2.5] and 7.2 [1.8]) at G0/G1 and G2/M phases. However, at S phase, AD-MSCs (13.4 [1.4]) had slightly lower percentage of cells compared to EMSCs (18.7 [1.7]; Figure 1D).
Differentiation Into Adipocytes, Osteocytes, and Chondrocytes
Mesenchymal lineage differentiation potential of AD-MSCs and EMSCs was carried out using lineage-specific induction media and stained for cytochemical changes.
After induction to adipogenic lineage, both AD-MSC- and EMSC-differentiated cells revealed the accumulation of lipid droplets compared to control cells as confirmed by Oil Red O staining. However, AD-MSCs showed higher differentiation potential to adipocytes with more accumulation of lipid droplets compared to EMSCs (Figure 2A). The expression of adipocyte lineage-specific marker genes such as fatty acid-binding protein (FABP), lipoprotein lipase (LPL), and adipocyte protein (AP2) were found to be significantly (P < .05) higher in differentiated EMSCs compared to undifferentiated control cells (Figure 2B). Osteogenic differentiation of both AD-MSCs and EMSCs showed accumulation of calcium and extracellular mineralization matrix which was confirmed by alizarin red and von Kossa staining (Figure 2A). Both showed similar differentiation potential to osteogenic lineage. The expression of osteogenic lineage-specific marker genes such as osteonectin (ON), Runt-related transcription factor 2 (RUNX2), and biglycan (BG) were found to be significantly (P < .05) higher in differentiated EMSCs compared to undifferentiated control cells (Figure 2C). Furthermore, both AD-MSCs and EMSCs were successfully differentiated into chondrocytes upon induction, and cells tend to form aggregation at the final stage of differentiation. The differentiated cells were positively stained for Alcian blue (stains sulfate glycosaminoglycan) and safranin O (cartilage). The differentiated AD-MSCs showed higher safranin O (cartilage) staining compared to differentiated EMSCs, whereas Alcian blue staining was noted higher in differentiated EMSCs (Figure 2A). The expression of chondrocyte lineage-specific marker genes such as SRY (sex determining region Y)-Box9 (SOX9), aggrecan (ACAN), collagen, type X, α1 (COL10A) were found to be significantly (P < .05) higher in differentiated EMSCs compared to undifferentiated control cells (Figure 2D). Overall, EMSCs showed higher differentiation potential to mesenchymal lineages compared to AD-MSCs.
A, Mesenchymal differentiation potential of EMSCS to adipocytes, osteocytes, and chondrocytes compared to AD-MSCs; Oil Red O staining of lipid droplets; alizarin red and von Kossa staining of mineralization of calcium deposit from differentiated cells; alcian blue and safranin O staining of chondrogenic differentiated cells; scale bar = 50 μm. B- D, RT-qPCR results showing adipocyte, osteocyte, and chondrocyte lineage-specific marker genes expression pattern in EMSCs and AD-MSCs; * indicates a significant difference (P < 0.05). CN indicates control undifferentiated cells; DF, differentiated cells; EMSCs, endometrial stromal mesenchymal stem cells; AD-MSCs, adipose-derived mesenchymal stem cells; RT-qPCR, real-time quantitative polymerase chain reaction.
Cardiomyogenic Differentiation Potential of EMSCs
To evaluate the transdifferentiation potential of porcine EMSCs into cardiomyocyte-like cells, EMSCs were treated with 5-AZA and angiotensin-II for 24 hours followed by 20 days incubation in TGFβ1-supplemented media. After 1 week of culture, EMSCs displayed elongated sticklike morphology, and within 2 to 3 weeks, all cells were found to be enlarged and connected with neighboring cells. The cells further changed their morphology into cluster-like aggregates and acquired myotube-like structures at the end of culture (Figure 3A). The RT-qPCR analysis of cardiomyocyte-specific genes such as desmin (DES), α-smooth muscle actin (ACTA2), cardiac troponin-T (cTnT), and α-cardiac actin (ACTC1) showed significantly (P < .05) increased expression in differentiated cells compared to untreated control cells (Figure 3B). The cardiomyocyte-specific proteins such as cTnT and ACTC1 showed increase in expression in a week-specific manner. At third week of differentiation, both proteins were highly expressed in differentiated cells demonstrating efficient transformation of EMSCs to cardiomyocyte-like cells (Figure 3C). Nevertheless, the immunofluorescence staining of differentiated EMSCs showed strong expression of cardiomyocyte-specific proteins DES, ACTA2, cTnT, and ACTC1 (Figure 4). The control EMSCs (CNT) displayed no morphological changes and negative for cardiomyocyte lineage-specific markers.
A, In vitro differentiation of EMSCs into cardiomyocyte-like cells. Significant changes in morphology were observed between undifferentiated control and differentiated cells; scale bar = 100 μm. B, RT-qPCR results showing a significant increase in cardiomyocyte-specific genes expression during differentiation compared to undifferentiated control cells (CNT); GAPDH was used as a housekeeping gene; * indicates a significant difference (P < .05) in cardiomyocyte-specific genes expression pattern at different time interval. C, Western blot results showing an increase in α-cardiac actin and cardiac troponin-T upon progression of differentiation; GAPDH was used as internal control protein. EMSCs indicates endometrial stromal mesenchymal stem cells; RT-qPCR, real-time quantitative polymerase chain reaction; GAPDH, glyceraldehyde 3-phosphate dehydrogenase.
Immunocytochemical analysis of EMSCs differentiated into cardiomyocyte-like cells showing positive expression of cardiac troponin T, α-cardiac actin, desmin, and α-smooth muscle actin; scale bar = 100 µm. EMSCs indicates endometrial stromal mesenchymal stem cells.
Differentiation of EMSCs Into Pancreatic β Cell-Like Cells
Upon stepwise induction to pancreatic β cell-like cells, the morphology did not change during step 1; however, at step 2, the cells started clustering and formed aggregates. By the end of step 3, cells formed spheroid bodies and finally at step 4 the cells formed islet-like clusters with distinct morphology similar to islet cells (Figure 5A). Most of the spheroid clusters were positively stained with DTZ cytochemical solution that reacted with pancreatic β cell-like cells (Figure 5B).
A, In vitro differentiation of porcine EMSCs into pancreatic β cell-like cells. Morphological changes were observed during differentiation when compared to undifferentiated control cells (CNT); scale bar = 100 µm. B, Differentiated β cell-like cells were positive for dithizone staining demonstrating the presence of insulin-containing vesicles; scale bar = 100 µm. C, RT-qPCR analysis of pancreatic β-cell lineage-specific marker genes expression revealing a significant increase in differentiated EMSCs when compared to undifferentiated control cells (CNT). GAPDH was used as internal control gene. * indicates a significantly (P < .05) higher levels of mRNA expression in differentiated β cell-like cells compared to untreated EMSCs control. D, Western blot showing an increase in PDX1 expression upon differentiation; β-tubulin was used for internal control protein. E, Densitometric analysis of (D); *, **, *** indicates a significant difference (P < .05) in PDX1 protein expression among different steps of induction. EMSCs indicates endometrial stromal mesenchymal stem cells; RT-qPCR, real-time quantitative polymerase chain reaction; GAPDH, glyceraldehyde 3-phosphate dehydrogenase.
PDX1, also known as insulin promoter factor 1, expression was significantly (P < .05) higher in cells at step 3 of differentiation compared to that in cells at step 1, 2, or 4 of differentiation. However, PDX1 expression was lower in control EMSCs (Figure 5D and E).
The RT-qPCR analysis showed a significant (P < .05) increase in messenger RNA expression of pancreatic β-cell–specific genes such as pancreatic and duodenal homeobox 1 (PDX1), glucose transporter 2 (GLUT2), somatostatin (SST), NKX6.1, paired box 4 (PAX4), and neurogenin-3 (NGN3; NEUROG3) in differentiated cells compared to untreated control EMSCs (Figure 5C). Immunocytochemical analysis showed that the expression of PDX1, INS, PAX4, NGN3, and GLUT2 were positive in differentiated pancreatic β cell-like cells, whereas no such staining was detected in control cells (Figure 6). The flow cytometric analysis revealed that around 42% of cells were positive for insulin expression in differentiated cell populations when compared to control undifferentiated cells (Supplementary Figure 1).
Immunocytochemical analysis of differentiated EMSCs into pancreatic β cell-like cells; scale bar = 100 µm. EMSCs indicates endometrial stromal mesenchymal stem cells.
The secretion of insulin and C-peptide in response to glucose challenge in differentiated cells were significantly (P < .05) higher than undifferentiated control cells (Figure 7A). Nevertheless, differentiated pancreatic β cell-like cells were positive for insulin protein as evaluated by immunocytochemistry (Figure 7B).
A, Evaluation of functional properties of differentiated β cell-like cells. Quantification of insulin and C-peptide levels upon glucose stimulation of differentiated EMSCs; * indicates significant differences (P < .05) in insulin and C-peptide levels in differentiated cells compared to control undifferentiated EMSCs. B, Immunocytochemical analysis showing insulin protein expression in differentiated EMSCs; scale bar = 100 µm. EMSCs indicates endometrial stromal mesenchymal stem cells.
Discussion
When pMSCs were treated with 5-AZA, 30% to 50% of cells showed cardiomyocyte-like characteristics and also expressed T-cardiotrophonin, desmin, and connexin-3. 15
The multilineage differentiation is a functional property of MSCs in which undifferentiated self-renewal cells are differentiated to specialized adult cells by expressing phenotypic differentiation markers and transcription factors. 32 The plasticity of adult stem cells are well demonstrated in BM-MSCs, where they migrate via bloodstream to damaged tissues like heart, muscle, liver, nervous system, pancreas, and skin to transdifferentiate to become cardiomyocytes, skeletal muscle cells, hepatocytes, neurons, β cells, and keratinocytes, respectively. 33–34 In porcine, umbilical cord matrix, BM-MSCs, and AD-MSCs were also reported to possess transdifferentiation capacity. 5,35
The human EMSCs possess high capacity to transdifferentiate into cardiomyogenic cells (76%-96%) compared to BM-MSCs (0.3%). 36,37 When pMSCs were treated with 5-AZA, 30% to 50% of cells showed cardiomyocyte-like characteristics and also expressed T-trophonin, desmin, and connexin-3, the markers of cardiomyocyte. 15 Although the exact mechanism of 5-AZA in cardiomyogenic transdifferentiation is unknown, it was found to upregulate cardiac troponin-T which is involved in cell proliferation and differentiation. 38 Angiotensin II was known to be involved in many cardiovascular diseases, vascular hypertrophy, 39 and also known to differentiate bone marrow-derived mononuclear cells to smooth muscle-like cells after vascular injury to repair damaged tissue. 40,41 Also in AD-MSCs, angiotensin-II through TGFβ1-Smad2 pathway can differentiate them into contractile smooth muscle-like cells. Further, a combination of angiotensin-II and 5-AZA more efficiently converted rat BM-MSCs to cardiomyocytes. 18 In this study, the transdifferentiation ability to cardiomyocyte-like cells was confirmed in differentiated cells by the expression of lineage-specific genes such as DES, ACTA2, cTnT, and ACTC1 by RT-PCR and immunostaining. Also, the Western blotting demonstrated the expression of cardiomyocyte-specific protein markers such as cTnT and ACTC1 in differentiated EMSCs.
There are controversial reports on beating cells following 5-AZA treatment. In 1999, Makino et al 42 reported the formation of beating cells from murine MSCs, while similar beating cells were not formed in other studies. 15,43 In the present study, we did not observe any beating cells though the expression of genes and proteins specific to cardiomyocyte lineage were highly upregulated. Therefore, the current protocol is not sufficient toward obtaining mature cardiomyocytes in vitro and requires additional factors. However, the observed morphological changes in this study were more eminent after induction proving that pretreatment of EMSCs for in vivo transplantation experiments will facilitate efficient differentiation toward cardiomyogenic lineage. Therefore, more detailed studies are necessary to analyze the molecular and functional changes during the course of transdifferentiation.
The present study also demonstrated the differentiation potential of EMSCs to pancreatic β cell-like cells. The current differentiation protocol does not involve any gene transfection or genetic modification and rather rely on defined culture conditions. Several factors might be involved in our transdifferentiation protocol of EMSCs to β cell-like cells, such as Geltrex LDEV-free reduced growth factor basement membrane matrix coating and treatment with indolactam V during initial stage, and exendin-4 and activin A at final stages. The role of small molecules, indolactam V, in differentiation of embryonic stem cells to pancreatic lineage cells was reported earlier. 44 Exendin-4, a 39 amino acid agonist of glucagon-like peptide 1, stimulates β-cell formation from ductal progenitor cells 45 and was utilized in our present study. Previously, human EMSCs were successfully differentiated to insulin-producing cells by employing indolactam V and exendin-4, and upon differentiation, these cells exhibited glucose-dependent C-peptide and insulin secretion 28 ; our results are also in agreement with these previous studies. The expression of PDX1, also known as insulin promoter factor 1, is necessary during endodermal commitment of embryonic development and also important transcription factor for β-cell maturation. 46 During step 3 of our differentiation, PDX1 expression was increased indicating the efficient transformation of EMSCs toward β cell-like cells.
Proinsulin composed of 3 chains, a carboxy-terminal A chain, an amino-terminal B chain connected with a short 31 amino acid polypeptide called C-peptide, or C-chain. 47 In the endoplasmic reticulum, endopeptidases cleaves the proinsulin and release insulin and C-peptide, which are packaged into secretory granules in Golgi body and released to cell cytoplasm. 48 The glucose challenge assay is considered gold-standard assay for assessing β-cell function, 48,27 and in the present study the differentiated cells exhibited glucose concentration-dependent insulin and C-peptide secretion, proving successful differentiation to mature pancreatic β cell-like cells. Our results are in agreement with previous reports from human EMSCs and BM-MSCs. 27,28,48
In conclusion, due to shortage of organs for transplantation in humans, xenotransplantation is gaining much more attention, and pigs being similar to humans in organ morphology have to be carefully evaluated for their suitability. From the perspective of therapeutic application, there is a need to understand the lineage-specific differentiation of these MSCs and their transformation to functional mature cells. A detailed study on the knowledge of molecules involved in transdifferentiation and their mechanism may provide new alternatives for cell therapy using pMSCs.
Supplementary Material
Supplementary_figure - CD105+ Porcine Endometrial Stromal Mesenchymal Stem Cells Possess Differentiation Potential Toward Cardiomyocyte-Like Cells and Insulin-Producing β Cell-Like Cells In Vitro
Supplementary_figure for CD105+ Porcine Endometrial Stromal Mesenchymal Stem Cells Possess Differentiation Potential Toward Cardiomyocyte-Like Cells and Insulin-Producing β Cell-Like Cells In Vitro by Raghavendra Baregundi Subbarao, Sharath Belame Shivakumar, Yong-Ho Choe, Young-Bum Son, Hyeon-Jeong Lee, Imran Ullah, Si-Jung Jang, Sun-A Ock, Sung-Lim Lee and Gyu-Jin Rho in Reproductive Sciences
Footnotes
Authors’ Note
Raghavendra Baregundi Subbarao and Sharath Belame Shivakumar contributed equally to this manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by a grant from the
Supplementary Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
