Insulin-producing cells could not mimic the physiological regulation of insulin secretion performed by pancreatic beta cells
- Qiping Shi†1,
- Simin Luo†1,
- Haiying Jia1,
- Lie Feng1,
- Xiaohua Lu1,
- Lixin Zhou2 and
- Jiye Cai2Email author
© Shi et al.; licensee Springer. 2013
Received: 31 January 2013
Accepted: 13 February 2013
Published: 19 February 2013
The aim of this study was to compare the difference between insulin-producing cells (IPCs) and normal human pancreatic beta cells both in physiological function and morphological features in cellular level.
The levels of insulin secretion were measured by enzyme-linked immunosorbent assay. The insulin gene expression was determined by real-time quantitative polymerase chain reaction. The morphological features were detected by atomic force microscopy (AFM) and laser confocal scanning microscopy.
IPCs and normal human pancreatic beta cells were similar to each other under the observation in AFM with the porous structure features in the cytoplasm. Both number of membrane particle size and average roughness of normal human beta cells were higher than those of IPCs.
Our results firstly revealed that the cellular ultrastructure of IPCs was closer to that of normal human pancreatic beta cells, but they still could not mimic the physiological regulation of insulin secretion performed by pancreatic beta cells.
KeywordsInsulin-producing cells Human normal pancreatic beta cells Atomic force microscopy
Diabetes is caused by absolute or relatively insufficient insulin secretion. Hitherto, there is no cure for diabetes. Treatment with insulin prolongs survival and improves glycemic control, and current standard diabetes treatment regimens with insulin replacement remain away from ideal. Transplantation of either isolated islets or the whole pancreas provides another mode for insulin replacement  but is often accompanied by many undesirable side effects [2–4]. Insulin-producing cells (IPCs) from pluripotent stem cells offer the potential to treat diabetes by providing a source for injured pancreatic beta cells without the limitations of current therapeutic modalities. To be successful, yet, IPCs must possess physiologically appropriate regulation of insulin secretion [5, 6], including sensing circulating glucose concentrations and secreting insulin in response to physiological glucose concentrations appropriately without risk of neoplastic transformation [7, 8]. Nowadays, unresolved obstacles associated with differentiation of stem cells into IPCs include maturation of the insulin secretory pathways and mechanisms responsible for sensing ambient glucose concentrations as well as lack of sufficient development of the insulin processing machinery [9, 10].
Atomic force microscopy (AFM) has been widely used in cell biology studies, especially of both cellular and subcellular structures and topographical morphology [11, 12], because of its ability to image biological samples at nanometer resolutions. Differences in cell morphology can likely reveal the reason why there is great difference in cellular function. Thus, we compared the differences in morphology and function between normal human pancreatic beta cells and IPCs derived from human adipose-derived stem cells (hADSCs). Moreover, we examined the relationship between cell morphology and function. At the molecular level, we found that although IPCs had a similar distribution of membrane proteins to normal pancreatic beta cells, they still could not mimic the physiological regulation of insulin secretion performed by normal pancreatic beta cells. We propose that the difference in physiological function between these two kinds of cells is due to the difference in the nanostructure of their cell membranes.
Isolation and differentiation of MSCs from human adipose tissue
Human adipose tissue was obtained from four donors, two males and two females. Informed consent was obtained from participating donors according to procedures approved by the Ethics Committee at the Chinese Academy of Medical Sciences. Experiments were performed according to the ethical standards formulated in the Helsinki Declaration. The isolated and differentiated procedure was described by Shi et al. . In order to authenticate the phenotypes of mesenchymal stem cells (MSCs), flow cytometric analysis of hADSCs was performed using antibodies for CD59, CD34, CD44, CD45, CD105, CD13, and HLA-DR (BD Biosciences, Franklin Lakes, NJ, USA).
Culture of normal human pancreatic beta cells
Normal human pancreatic beta cells were obtained commercially (HUM-CELL-0058, Wuhan Pricells Biotechnology & Medicine Co., Ltd., Wuhan, China). Expansion medium contained MED-0001 and 5 ng/mL rhEGF, 5 μg/mL rhinsulin, 5 μg/mL transferrin, 10 nM T3, 1.0 μM epinephrine, 5 μg/mL hydrocortisone, 10% fetal bovine serum (all expansion media were from Wuhan Pricells Biotechnology & Medicine Co., Ltd.). The cells were cultured in complete medium in T-25 tissue culture flasks that have been coated with collagenase at 37°C in 5% CO2. Culture medium was changed every other day.
Both normal human pancreatic beta cells and IPCs were preincubated in Dulbecco's phosphate-buffered saline (D-PBS, without glucose), low-glucose Dulbecco's modified Eagle's medium (DMEM; 5.5 mM, Gibco, Grand Island, NY, USA), or high-glucose DMEM (25 mM, Gibco) for 1 h or 30 min. The buffers from six wells of cells were collected separately. The amount of insulin in the buffer of each well was determined by ultrasensitive insulin enzyme-linked immunosorbent assay (ELISA) and normalized by the number of cells in each well.
Quantitative gene expression analysis
Sequences of primers for real-time qRT-PCR
Sample preparation for AFM
To detect the morphological changes of beta cells and IPCs before and after glucose stimulation, cells were separated into five groups: glucose-free culture medium group (D-PBS), 30-min low-glucose stimulation group, 1-h low-glucose stimulation group, 30-min high-glucose stimulation group, and 1-h high-glucose stimulation group. Cell samples were preincubated for 1 h or 30 min in D-PBS, low-glucose DMEM (Gibco), or high-glucose DMEM (Gibco). They were then washed in distilled water twice before being fixed with 2.5% glutaraldehyde for 20 min. The samples were washed in distilled water three times again, then air-dried for AFM scanning.
An Autoprobe CP AFM (Veeco, Plainview, NY, USA) was used in contact mode to detect the immobilized IPCs and normal human pancreatic beta cells at room temperature. Silicon nitride tips (UL20B, Park Scientific Instruments, Sunnyvale, CA, USA) were employed in all AFM measurements. An optical microscope was used to help select the desired cells and direct the position of the AFM tip. Single-cell imaging was repeated for six cells, and each cell was scanned for three times. All images were analyzed by the instrument-equipped software (Image Processing Software Version 2.1) to gain information on the topography. ‘Ra’ denotes the average roughness in the analytical area. All parameters were directly generated by the software IP2.1.
LCSM and observation
Cells were fixed in 2.5% glutaraldehyde for 30 min, washed in PBS, and then permeabilized in 0.1% Triton X-100 at room temperature for 30 min. After, cells were washed in PBS thoroughly. Cells were then incubated with 1 μM phalloidin-rhodamine (Biotium, Inc., Hayward, CA, USA) at 4°C overnight to label F-actin. After several washes in PBS, the labeled cells were scanned by LCSM (510 Meta Duo Scan, Carl Zeiss, Oberkochen, Germany) using 545-nm (He-Ne) excitation. Emission was detected above 600 nm.
All data were presented as mean values ± standard deviation taken from ten different cells. The morphologic parameters between the different groups were compared using t test (via SPSS 11). Differences with P values less than 0.05 were considered to be statistically significantly.
Morphology and phenotypes of cultured hADSCs
Isolated hADSCs exhibited a spindle shape, began to appear in culture, and reached 90% confluence in about 10 to 12 days. The second passage of hADSCs expanded rapidly and developed a uniform morphology that resembled that of fibroblasts. FACS analysis of hADSCs at the third passage showed that these cultured cells were positive for CD13 (98.88%), CD44 (98.9%), CD59 (98.4%), and CD105 (71.24%). In addition, expression of HLA-DR (0.98%) was not detected. Furthermore, hADSCs exhibited low expression of hematopoietic lineage markers CD45 (1.03%) and CD34 (2.88%).
Differentiation of IPCs
Insulin secretion of cells (μU/mL)
Normal human pancreatic β cells
9.25 ± 1.14
9.65 ± 1.12
23.43 ± 4.12
25.81 ± 2.57
0.46 ± 0.04
1.01 ± 0.11
1.20 ± 0.13
1.50 ± 0.23
Morphology of cells as observed by AFM
Characteristic of cells
Normal human pancreatic β cells
55.46 ± 4.84
73.45 ± 2.08*
34.71 ± 1.57
40.78 ± 1.09*
505.39 ± 12.01
421.46 ± 19.25*
Morphological features of three groups of cells
Normal human pancreatic β cells
107.05 ± 10.77
30.50 ± 1.61
H-glucose (30 min)
135.05 ± 6.46*
41.88 ± 2.38*
H-glucose (1 h)
138.26 ± 11.76*
49.41 ± 7.42*
L-glucose (30 min)
115.81 ± 46.86*
30.76 ± 1.29
L-glucose (1 h)
129.99 ± 15.33*
36.58 ± 2.99*
Particle size (nm)
215 ± 7.9
152 ± 5.7
H-glucose (30 min)
345 ± 9.35*
225 ± 7.9*
H-glucose (1 h)
360 ± 8.0*
233 ± 10.4*
L-glucose (30 min)
221 ± 12.94*
160 ± 7.90
L-glucose (1 h)
229 ± 14.74*
169 ± 9.62
Observation of cytoskeleton in human normal pancreatic beta cells and IPCs
We successfully extracted hADSCs from human adipose tissue according to the method reported in the literature [14, 15] and characterized the phenotypes of hADSCs through flow cytometry. After that, we used a simple chemical method not involving insulin to differentiate hADSCs into IPCs in vitro. In order to assess the function of IPCs, we tested the glucose-induced insulin secretion of IPCs and beta cells in vitro. Our data show that regardless of whether they were stimulated for 30 min or 1 h, the beta cells could release a certain amount of insulin after stimulation with high or low glucose concentrations. However, only when stimulated for 1 h in low glucose concentrations did IPCs secrete a little bit of insulin. The results indicate that IPCs can secrete insulin in response to glucose stimulation, similar to, but not as well as beta cells. Even though we only compared beta cells and one kind of IPC which was derived from one source using one differentiation method, our results made evident the difference in physiological function between these IPCs and beta cells. This evidence led to the question: ‘What were the reasons for the difference between IPCs and beta cells?’ We conjectured that these differences were due to the differences in cellular structure. To confirm our hypothesis, we first used AFM to detect cell surface ultrastructure of beta cells and IPCs.
AFM images indicated the changes in morphological properties of IPCs and beta cells stimulated by glucose. The morphologies of IPCs and beta cells were similar to each other, as observed via AFM. They all were polygonal and contained visibly porous features in the cytoplasm. AFM is a common method used to observe cell morphology. However, few studies have reported that these porous structures existed naturally on the cell surface [16–20]. Pores on the cell surface generally appeared after treatment with some drugs [21, 22]. Nevertheless, the pores observed after drug treatment were not the same as the porous structures we detected. The porous structures in the IPCs and beta cells were organized and well distributed around the nuclei. The pores that appear after drug treatment are dispersed and isolated. Kim et al. deemed that these isolated holes on the cell surface after drug treatment might be one form of cell apoptosis . Additionally, we speculated that these uniform holes arranged in the cytoplasmic membrane might be dependent onto the type of cells. Beta cells are endocrine cells, while IPCs are artificially synthesized copies of endocrine cells. The pores in the cytoplasmic membrane might be indicative of the exocytosis process through which the hormone is released into the extracellular space.
Simultaneously, we used AFM to compare the cell membrane particle size and Ra of the membrane surface before or after glucose stimulation of IPCs and beta cells. Our results revealed that both membrane particle size and Ra of beta cells were larger than those of IPCs. When both two groups of endocrine cells were stimulated by glucose, the membrane particle size and Ra were higher than those not stimulated, except for IPCs that were stimulated for 30 min with low glucose concentration. The magnitude of cellular Ra, as well as the types, structure, and quantity of membrane protein molecules, directly influenced the inclines and declines of the membrane surface . We speculated that the reason for the lower membrane particle size and Ra in IPCs might be due to their lower membrane protein content. The cell membrane accomplishes its biological function through membrane liquidity, and exocytosis is one of the functions that depend on membrane liquidity [24, 25]. IPCs and beta cells secreted insulin through exocytosis. In the meantime, their plasma membranes were replenished via membrane liquidity. We inferred that the change in membrane liquidity might cause the increase in cell membrane particle size and Ra after glucose stimulation.
Beta cells secrete insulin through exocytosis. In beta cells, actin filaments form a dense network under plasma membrane. This actin network acts as a barricade, preventing passive diffusion of insulin follicles to the plasma membrane. Thus, the actin network ultimately lessens insulin secretion via reduction of exocytosis . On the contrary, F-actin depolymerization can increase exocytosis, which increases insulin secretion. We proposed that the pores we observed that were located in the cytoplasmic membrane were one of the characteristics of insulin exocytosis, and increased evidence of porous structures may be related to the enhancement of insulin exocytosis.
To prove that exocytosis had been enhanced after glucose stimulation of IPCs and beta cells, we demonstrated that without glucose stimulation, the actin network underneath the plasma membrane was continuous and dense. After glucose stimulation, the actin network depolymerized and became discontinuous. After F-actin depolymerization, inhibition of exocytosis was relieved and insulin secretion increased. Interestingly, in the IPCs group, the cortical actin network did not depolymerize in low glucose concentrations after 30 min of stimulation. The actin network became discontinuous and depolymerized only after low-glucose stimulation for 1 h.
In conclusion, our data proved that only normal human pancreatic beta cells could release insulin after low- and high-glucose stimulation for 30 min and 1 h. The cellular ultrastructure and function of IPCs resembled closely those of the normal human pancreatic beta cells. However, there is still so much diversity between the two groups of cells. These differences might provide a future research direction to figure out how to optimize differentiation into IPCs. In our study, we only tested the difference between one kind of IPC and normal human pancreatic beta cells. Therefore, our results are not enough to elucidate the relationship between cellular ultrastructure and function. In order to explore the relationship between cellular structure and cell function, we need to study the links between cell function and more cell membrane proteins, as well as analyze various types of endocrine cells by looking for the common cellular surface ultrastructure.
This work was funded by the Guangdong Provincial Science and Technology Project of China (2010B031600105, 2011B031800066), granted from the Guangdong Provincial Medical Scientific Research Foundation (B2011161), and supported by the National Natural Science Foundation of China (973 program projects, 2010CB833603) and the Fundamental Research Funds for the Central Universities.
- Venstrom JM, McBride MA, Rother KI, Hirshberg B, Orchard TJ, Harlan DM: Survival after pancreas transplantation in patients with diabetes and preserved kidney function. JAMA 2003, 290: 2817–2823. 10.1001/jama.290.21.2817View ArticleGoogle Scholar
- Campbell PM, Senior PA, Salam A, LaBranche K, Bigam DL, Kneteman NM, Imes S, Halpin A, Ryan EA, Shapiro AMJ: High risk of sensitization after failed islet transplantation. Am J Transplant 2007, 7: 2311–2317. 10.1111/j.1600-6143.2007.01923.xView ArticleGoogle Scholar
- Korsgren O, Nilsson B, Berne C, Felldin M, Foss A, Kallen R, Lundgren T, Salmela K, Tibell A, Tufveson G: Current status of clinical islet transplantation. Transplantation 2005, 79: 1289–1293. 10.1097/01.TP.0000157273.60147.7CView ArticleGoogle Scholar
- Ryan EA, Lakey JR, Rajotte RV, Korbutt GS, Kim T, Imes S, Rabinovitch A, Elliott JF, Bigam D, Kneteman NM, Warnock GL, Larsen I, Shapiro AJ: Clinical outcomes and insulin secretion after islet transplantation with the Edmonton protocol. Diabetes 2001, 50: 710–719. 10.2337/diabetes.50.4.710View ArticleGoogle Scholar
- Porat S, Dor Y: New sources of pancreatic beta-cells. Curr Diab Rep 2007, 7: 304–308. 10.1007/s11892-007-0049-8View ArticleGoogle Scholar
- Rolletschek A, Kania G, Wobus AM: Generation of pancreatic insulin-producing cells from embryonic stem cells—“proof of principle”, but questions still unanswered. Diabetologia 2006, 49: 2541–2545. 10.1007/s00125-006-0441-yView ArticleGoogle Scholar
- Fujikawa T, Oh SH, Pi L, Hatch HM, Shupe T, Petersen BE: Teratoma formation leads to failure of treatment for type I diabetes using embryonic stem cell derived insulin-producing cells. Am J Pathol 2005, 166: 1781–1791. 10.1016/S0002-9440(10)62488-1View ArticleGoogle Scholar
- Kroon E, Martinson LA, Kadoya K, Bang AG, Kelly OG, Eliazer S, Young H, Richardson M, Smart NG, Cunningham J, Agulnick AD, D'Amour KA, Carpenter MK, Baetge EE: Pancreatic endoderm derived from human embryonic stem cells generates glucose-responsive insulin-secreting cells in vivo. Nat Biotechnol 2008, 26: 443–452. 10.1038/nbt1393View ArticleGoogle Scholar
- Fellous TG, Guppy NJ, Brittan M, Alison MR: Cellular pathways to β-cell replacement. Diabetes Metab Res Rev 2007, 23: 87–99. 10.1002/dmrr.692View ArticleGoogle Scholar
- Santana A, Ensenat-Waser R, Arribas MI, Reig JA, Roche E: Insulin-producing cells derived from stem cells: recent progress and future directions. J Cell Mol Med 2006, 10: 866–883. 10.1111/j.1582-4934.2006.tb00531.xView ArticleGoogle Scholar
- Bushell GR, Cahill C, Clarke FM, Gibson CT, Myhra S, Watson GS: Imaging and force-distance analysis of human fibroblasts in vitro by atomic force microscopy. Cytometry 1999, 36: 254–264. 10.1002/(SICI)1097-0320(19990701)36:3<254::AID-CYTO16>3.0.CO;2-4View ArticleGoogle Scholar
- Domke J, Dannohl S, Parak WJ, Muller O, Aicher WK, Radmacher M: Substrate dependent differences in morphology and elasticity of living osteoblasts investigated by atomic force microscopy. Colloids Surf B Biointerfaces 2000, 19: 367–379. 10.1016/S0927-7765(00)00145-4View ArticleGoogle Scholar
- Shi P, Luo S, Jin H, Cai J, Jia H, Feng L, Lu X: Insulin-producing cells from human adipose tissue-derived mesenchymal stem cells detected by atomic force microscope. Appl Microbiol Biotechnol 2012, 94: 479–486. 10.1007/s00253-012-3904-8View ArticleGoogle Scholar
- Linscheid P, Seboek D, Nylen ES, Langer I, Schlatter M, Becker KL, Keller U, Müller B: In vitro and in vivo calcitonin I gene expression in parenchymal cells: a novel product of human adipose tissue. Endocrinology 2003, 144: 5578–5584. 10.1210/en.2003-0854View ArticleGoogle Scholar
- Timper K, Seboek D, Eberhardt M, Linscheid P, Christ-Crain M, Keller U, Müller B, Zulewski H: Human adipose tissue-derived mesenchymal stem cells differentiate into insulin, somatostatin, and glucagon expressing cells. Biochem Biophys Res Commun 2006, 341: 1135–1140. 10.1016/j.bbrc.2006.01.072View ArticleGoogle Scholar
- Wozniak MJ, Kawazoe N, Tateishi T, Chen GP: Monitoring of mechanical properties of serially passaged bovine articular chondrocytes by atomic force microscopy. Micron 2009, 40: 870–875. 10.1016/j.micron.2009.06.002View ArticleGoogle Scholar
- Cross SE, Jin YS, Tondre WR, Wong R, Rao JY, Gimzewski JK: AFM based analysis of human metastatic cancer cells. Nanotechnology 2008, 19: 384003. 10.1088/0957-4484/19/38/384003View ArticleGoogle Scholar
- Azeloglu EU, Bhattacharya J, Costa KD: Atomic force microscope elastography reveals phenotypic differences in alveolar cell stiffness. J Appl Physiol 2008, 105: 652–661. 10.1152/japplphysiol.00958.2007View ArticleGoogle Scholar
- Parpura V, Haydon PG, Henderson E: Three-dimensional imaging of living neurons and glia with the atomic force microscope. J Cell Sci 1993, 104: 427–432.Google Scholar
- Luo S, Shi Q, Zha Z, Ping Y, Lin H, Liu N, Wu H, Jin H, Cai JY: Morphology and mechanics of chondroid cells from human adipose-derived Stem cells detected by atomic force microscopy. Mol Cell Biochem 2012, 365: 223–231. 10.1007/s11010-012-1263-5View ArticleGoogle Scholar
- Wang M, Ruan YX, Chen Q, Li SP, Wang QL, Cai JY: Curcumin induced HepG2 cell apoptosis-associated mitochondrial membrane potential and intracellular free Ca2+ concentration. Eur J Pharmacol 2011, 650: 41–47. 10.1016/j.ejphar.2010.09.049View ArticleGoogle Scholar
- Kim KS, Cho CH, Park EK, Jung MH, Yoon KS, Park HK: AFM-detected apoptotic changes in morphology and biophysical property caused by paclitaxel in Ishikawa and HeLa cells. PLoS One 2012, 7: e30066. 10.1371/journal.pone.0030066View ArticleGoogle Scholar
- Brammer KS, Oh S, Cobb CJ, Bjursten LM, van der Heyde H, Jin S: Improved bone-forming functionality on diameter-controlled TiO(2) nanotube surface. Acta Biomater 2009, 5: 3215–3223. 10.1016/j.actbio.2009.05.008View ArticleGoogle Scholar
- Kubina M, Lanza F, Cazenave JP, Laustriat G, Kuhry JG: Parallel investigation of exocytosis kinetics and membrane fluidity changes in human platelets with the fluorescent probe, trimethylammonio-diphenylhexatriene. Biochim Biophys Acta 1987, 901: 138–146. 10.1016/0005-2736(87)90265-3View ArticleGoogle Scholar
- Hirano K: Change in membrane fluidity of sand dollar egg cortices caused by Ca2+-induced exocytosis: microscopic analysis with fluorescence anisotropy. Dev Growth Differ 1991, 33(5):451–458. 10.1111/j.1440-169X.1991.00451.xView ArticleGoogle Scholar
- Olofsson CS, Håkansson J, Salehi A, Bengtsson M, Galvanovskis J, Partridge C, SörhedeWinzell M, Xian X, Eliasson L, Lundquist I, Semb H, Rorsman P: Impaired insulin exocytosis in neural cell adhesion molecule−/− mice due to defective reorganization of the submembrane F-actin network. Endocrinology 2009, 150(7):3067–3075. 10.1210/en.2008-0475View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.