Background
As a major risk factor for renal failure in patients with end-stage renal disease [
1], diabetic nephropathy (DN) often leads to diabetic mortality. The occurrence and development of DN might involve complex pathophysiological interactions of inflammatory, metabolic and hemodynamic factors [
2]. All these factors induce injuries of glomeruli, tubular epithelial cells, interstitial fibroblasts and vascular endothelial cells. Recent advances showed that oxidative stress played key roles in the pathogenesis of DN. Glucose-dependent pathways of advanced glycation end products (AGEs) were essential in the development of diabetic nephropathy. Accumulation of enhanced AGEs in kidney could directly modulate the expression of key components of renin-angiotensin system [
3]. And AGEs interacted with their corresponding receptors (RAGE) in glomerular endothelial cells for up-regulating protein kinase C (PKC), suppressing protein kinase A (PKA) and activating oxidative stress response [
4], a salient feature of microangiopathy in DN [
5]. Also using PKC inhibitor or PKA agonist may arrest the onset and progression of DN.
Strict controls of glucose level and blood pressure have remained standard treatment for DN [
6,
7]. However, the efficacy has been unsatisfactory. Recent studies confirmed that pancreatic and islet transplantation significantly reduced the risks of diabetic macrovascular complications and effectively reversed diabetic microvasculopathy [
8‐
10]. However, due to an acute shortage of pancreas donor, pancreas and islet transplantation failed to meet clinical demands, greatly limiting its clinical application. Various biologically active products secreted by pancreatic islet cells, such as C-peptide and glucagon-like peptide 1 (GLP-1), have been efficacious for diabetic vasculopathy. With the potentials of self-renewal and cellular differentiation, stem cell technology featuring in vitro proliferation and directional differentiation for preparing functional islets might increase the supply of islet cells [
11,
12].
Stem cells derived from early human embryonic pancreas had strong capacities of in vitro proliferation and directional differentiation. Specifically, human embryonic pancreas-derived progenitor cells were utilized for preparing progenitor cell-derived islets via induced directional differentiation. The progenitor cell-derived islets were transplanted into liver for evaluating treatment efficacy and elucidating mechanisms in DN rats. Our work might provide preliminary evidence for treating DN with progenitor cell-derived islets.
Methods
Animal modeling
The study protocol was approved by the Animal Ethics Committee of China-Japan Friendship Hospital. And 8-week-old male Wistar rats (250–300 g) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China) and were allowed to adapt to the housing environment and diet for 1 week. Then the diabetic model was induced by administration of 12 g/L streptozocin (freshly diluted with 0.1 mol/L of citrate buffer at pH 4.5) intraperitoneally (30 mg/kg) twice within 1 week interval. Blood glucose of caudal vein >16.7 mmol/L for 3 consecutive days after second STZ-injection was considered as a standard of diabetic model. Successful diabetic modeling was confirmed if blood glucose exceeded 16.7 mmol/L for 3 consecutive days. Diabetic animals were fed with a normal diet for 6 consecutive months until an onset of severe kidney disease. Successful modeling of DN was confirmed if 24 h urinary output was >twofolds, urinary albumin >50-folds and urinary protein >threefolds in DN rats as compared to normal control rats. DN animals were divided randomly into three groups of progenitor cell (n = 6, transplantation of progenitor cell-derived islets into liver via portal vein), insulin-treated (n = 6, 2.5 IU/day insulin glargine, sc) and DN (n = 6, no treatment). And healthy Wistar rats of the same age were used as normal controls.
Expansion and culturing of human fetal pancreatic progenitor cells
The study protocol was approved by the Clinical Research Ethics Committee of our hospital. Progenitor cells were isolated from aborted human fetal pancreas at gestational week 8. And human fetal pancreases were harvested after obtaining informed patient consents. As previously described [
13], fetal pancreases were digested with collagenase XI at 37 °C for 15 min. Islet-like tissue rich in endocrine progenitor cells were collected and cultured in DMEM/F12 containing basic fibroblast growth factor, epidermal growth factor, leukemia inhibitor factor (Peprotech, NJ, USA) and 5% fetal bovine serum for stem cells.
Cells were expanded in growth factor enriched medium for about 30–45 days and the cells between passages 6–9 were used in this study. The marker expression of pancreatic endocrine progenitor cells was evaluated by immunofluorescent staining. The progenitor cells were differentiated for 3 weeks in culture medium of M199 containing 15% fetal bovine serum, glp-1 (10 nM) and nicotinamide (5 mM). The endocrine hormones of insulin and glucagon were measured by immunofluorescent staining. And the insulin release upon glucose stimulation was measured by enzyme-linked immunosorbent assay (ELISA).
Transplantation of progenitor cell-derived islets
Differentiated cells were re-suspended and cultured overnight for inducing islet-like clusters. Progenitor cell-derived islet suspension was slowly injected into liver via portal vein (approximately 1000 islets equivalent per animal). Rats in the “Insulin Group” received 2.5 IU/day insulin glargine (sc).
Functional assessments of progenitor cells in vivo
For glucose monitoring via tail vein, animals were placed into metabolic cages (Suzhou Fengshi Laboratory Animal Equipment Co., Ltd, Jiangsu, China) before and every 2 weeks after transplantation. A 24-h urinary measurement was made for each animal of each group. Urinary albumin was measured by ELISA (Assay Max Rat Albumin ELISA kit, Gentaur, Belgium) and urinary protein by BCA assay (BCA protein assay kit, Beyotime, Shanghai, China).
At week 16 post-transplantation, the serum level of human C-peptide level was measured by ELISA (DRG International Inc., NY, USA).
Evaluations of cellular immunogenicity
Expression of HLA molecules
HLA classes I & II molecules on progenitor cells were detected by flow cytometry. In brief, single cell solution were blocked with 0.1% BSA in PBS, and then incubated with mouse anti-HLA classes I & II antibodies at 4 °C for 40 min, washed thrice with 0.1% BSA/PBS and incubated with Alexa488-conjugated donkey anti-mouse IgG respectively at 4 °C for 30 min. Fluorescence was detected by flow cytometry (Beckman coulter, CA, USA).
Activation of lymphocyte by progenitor cells
A total of 5 × 106 progenitor cells undergone frozen-thaw thrice and then ultrasonicated for cell lysate. Rat lymphocytes were isolated by Ficoll, seeded in 24-well plate at a density of 1 × 106/well and incubated with cell lysate for 24 or 48 h. PHA (10 μg/mL) and PMA (10 ng/mL) were used as positive controls. Supernatant was collected for measuring IL-2 concentration by ELISA (R&D, MN, USA).
Histopathological and immunohistochemical stains
Dissected liver and kidney tissues were fixed for 48 h in 10% neutral buffered formalin and followed by conventional tissue processing and paraffin-embedding. Each paraffin-embedded sample was sectioned into 3-µm thick slices and stained by hematoxylin & eosin (H&E). And kidney sections were stained with periodic acid-Schiff (PAS) reagent. For immunohistochemical staining, liver sections were incubated with primary anti-human mitochondrial antigen, anti-human C-peptide and anti-human glucagon antibodies while kidney sections with primary anti-fibronectin (anti-FN), anti-RAGE, anti-PKC β, anti-PKA, anti-iNOS and anti-SOD1 antibodies. Subsequent incubation was made with corresponding horseradish peroxidase (HRP)-conjugated secondary antibodies and staining with 3,3′-diaminobenzidine (DAB) peroxidase substrate solution.
Transmission electron microscopy
Renal cortex was fixed for 1 h at room temperature in 2.5% glutaraldehyde (prepared in 0.1 M phosphate buffer, pH 7.4). Specimens were processed routinely for electron microscopy. Morphology of glomerular podocytes was assessed under a TEM JEO1010. And the thickness of glomerular basement membrane (GBM) was measured and averaged (n > 25).
Real-time polymerase chain reaction
Total RNA from isolated glomeruli were extracted with RNAeasy Mini (Qiagen, Germany). Two micrograms of total RNA template, Oligo dT primer and AMV reverse transcriptase (Invitrogen, Thermo Fisher Scientific Co., USA) were used for synthesizing first-strand cDNA. Real-time fluorescent quantitative polymerase chain reaction (PCR) were performed with SYBR green PCR reagent kit (TOYOBO, Osaka, Japan) on Applied Biosystems 7300 Real-Time PCR System (Life Technologies Corporation, Carlsbad, CA). And β-actin was utilized as a house keeping gene for normalizing mRNA expression using the 2−ΔΔCt formula.
Western blot
Glomerular protein was extracted with SDS protein lysate (KeyGEN, Nanjing, China), separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a PVDF membrane. PVDF membrane was blocked with 5% skim milk containing 0.05% Tween-20 and incubated with individual primary antibodies, including anti-FN (Santa Cruz Biotechnology, Dallas, TX), anti-RAGE, anti-PKC (Sigma-Aldrich, Shanghai, China), anti-PKA (Abcam, UK), anti-iNOS (Abcam, UK), anti-SOD1 (Santa Cruz, USA) and anti-actin (Sigma-Aldrich, Shanghai, China). Then HRP-conjugate secondary antibody was utilized and color development achieved with enhanced chemiluminescence (ECL) reagent (EMD Millipore, Billerica, MA, USA).
Statistical analysis
SPSS13.0 software was used for statistical analysis of all data. Data were presented as mean ± standard deviation (x ± s). And comparisons among multiple groups were performed using ANOVA. P < 0.05 was deemed as statistically significant.
Discussion
Insulin producing cells derived from progenitor cells were reported effectively in reducing blood glucose in diabetic animals [
14‐
16]. In this study, progenitor cell-derived islet transplantation improved significantly not only blood glucose but also DN in DN animals.
Fetal pancreatic progenitor cells were isolated from 8th gestational week. It was reported that first-trimester human fetal pancreas had lower immunogenicity than second-trimester pancreas [
17]. The immunogenicity of pancreatic progenitor cells and islet-like cell clusters from first-trimester was much lower than that of first-trimester human fetal pancreas tissue [
18]. Moreover, these cells had a distinctively lower MHC I & II expression relative to second-trimester pancreatic progenitor cells, even after IFNγ challenge [
18]. Our results indicated that progenitor cells isolated from 8th gestational week exhibited low immunogenicity in vitro and in vivo. Transplanted progenitor cell-derived islets could survive longer in liver and effectively reduced DN without immunosuppressants.
As an early sign of DN, microalbuminuria gradually develops into refractory proteinuria and ultimately renal failure [
19,
20]. Reduction of urinary albumin is a marker for improved DN [
21]. Our data show that progenitor cell-derived islets could survive in liver and urinary protein and albumin decreased in diabetic rats.
Structural and functional changes of glomerular filtration barrier, composed of podocytes, GBM and endothelial cells, are the pathophysiological basis of proteinuria in DN. Adjacent processes between podocytes alternate with each other and form a 30–40-nm slit covered with a 4–6 nm membrane layer of GPSD. This is the final barrier for plasma protein passing through renal glomerular vasculature and maintaining the structural and functional integrity of glomerular filtration barrier [
22]. Progenitor cell-derived islets could reduce GBM thickness in diabetic rats and it explained a lower level of urinary albumin.
Mostly distributed in glomerular mesangial matrix, FN is a major non-collagenous glycoprotein within glomerular ECM. As a result, any change of FN fully reflects in glomerular ECM. Islet transplantation inhibited the glomerular accumulation of FN in DN rats. Progenitor cell-derived islet could up-regulate the expression of GPSD molecules. Thus progenitor cell-derived islets might reverse DN through improving the structural integrity of glomerular podocytes, GPSD and GBM.
As confirmed by in vivo experiments and clinical trials, AGEs significantly increased in types I & II DM. And AGEs interact with its receptor RAGE for promoting the pathogenesis of DN. A combination of AGEs and RAGE promotes ROS generation, activates the renin-angiotensin system and interacts with such signaling molecules as microtubule-affinity-regulating-kinase (MARK), nuclear factor-kappa B (NF-κB) and PKC. And PKC family includes at least 12 subunits [
23] and PKC-β subunit is a major subunit involved in pathological changes of DN [
24]. Our previous study found that bonding of AGEs-RAGE contributed to DN by an up-regulation of PKC-β and a down-regulation of PKA, thereby promoting DN via oxidative stress [
4]. Moreover, the glomerular accumulation of FN was also correlated with an up-regulation of RAGE and a down-regulation of PKA [
25]. In this study, progenitor cell-derived islets could reduce DN by lowering RAGE and PKC but increasing PKA in glomeruli.
Previous studies suggest that PKC and PKA signaling pathways are closely correlated with oxidative stress. Two major causes of oxidative stress are an excessive generation of ROS and insufficient antioxidants. And iNOS expression was higher and SOD1 expression was lower in diabetic rats than controls. Progenitor cell-derived islet transplantation reduced oxidative stress, lowered RAGE and PKC and elevated PKA in DN rats.
Although insulin treatment lowered and maintained stable blood glucose in our rat model, its therapeutic effect was insignificant, suggesting that simple control of blood glucose failed to restore structural and functional renal lesions in DN rats.
The efficacy of progenitor cell-derived islets may be due to insulin. Immunohistochemical staining showed that progenitor cell-derived islets secreted insulin, C-peptide and human glucagon in liver. The findings were similar to those of previous reports [
14,
26]. And C-peptide was efficacious for diabetic microangiopathy, especially DN [
27‐
30]. Dosing of C-peptide could significantly improve renal size, morphology and function in DN rats [
25,
31‐
33]. In humans, short and long-term treatments of C-peptide affected renal regulatory and physiological functions in type I DM [
34‐
36]. With a restoration of endogenous insulin after transplantation, mechanisms behind C-peptide-mediated microvascular improvements may be due to specific binding of G-protein coupled receptors to renal tubular and mesangial cells [
37]. Besides C-peptide, glucagon secretion from A-cells of pancreatic islets offered supports for beta cell function. Excessive digestion of isolated islets depleted alpha cells and impaired islet function after transplantation. However, it is unclear whether or not glucagon is involved in other mechanisms of direct improvements in DN rats.
Beside insulin replacement, indirect effects of pancreatic progenitor cannot be ruled out in this study. For example, it may have immune modulation and paracrine action on liver. Recent years, immune modulation of several kinds of stem cells is accumulated. A phase 1/phase 2 study showed that T2D patients achieve improved metabolic control and reduced inflammation markers after receiving autologous mononuclear cells which briefly co-cultures with adherent cord blood-derived multipotent stem cells. The reason is stem cells modulating immune function of monocytes and balancing Th1/Th2/Th3 cytokine production [
38]. Mesenchymal stem cell was also well known by the moderating immune response of type 1 diabetes [
39]. In addition, allogeneic adipose-derived mesenchymal stem cell (ADMSCs) was reported to ameliorate experimental autoimmune diabetes via downregulation of the CD4(+) Th1-biased immune response and expansion of regulatory T cells (Tregs) in the pancreatic lymph nodes. In vitro, ADMSCs induced the expansion/proliferation of Tregs in a cell contact-dependent manner mediated by programmed death ligand 1 [
40]. Although there is no data about the effect of pancreatic progenitors on Tregs, it is valuable to do further investigation.
We suppose that one of the other mechanisms of pancreatic progenitor cells maybe ameliorates insulin resistance by paracrine effect. As shown in MSC, glucose uptake in peripheral tissues, including skeletal muscle and adipose tissue, was elevated in MSC-treated mice. Furthermore, enhanced glucose uptake in these tissues was associated with improved insulin signaling as assessed by Akt phosphorylation and the expression of GLUT-4 [
41]. In summary, pancreatic progenitor cells may improve DN by multi-ways.
Authors’ contributions
YJ collection of data, data analysis and interpretation, manuscript writing; WZ collection and/or assembly of data, data analysis and interpretation, manuscript writing; SX collection of data, data analysis and interpretation, administrative support; HL collection and/or assembly of data, administrative support; WS conception and design, collection and/or assembly of data; HL, LP and QF collection and/or assembly of data, data analysis and interpretation; LC and JL conception and design, manuscript editing. All authors read and approved the final manuscript.