Original Paper Accepted after revision: March 27, 2014 Published online: December 3, 2014
Cells Tissues Organs DOI: 10.1159/000362500
In vitro Differentiation of Human Adipose Tissue-Derived Stem Cells into Islet-Like Clusters Promoted by Islet Neogenesis-Associated Protein Pentadecapeptide Lili Ren a, b Lijuan Chen b Hui Qi b Furong Li b Feili Gong a
a
Department of Immunology, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, and b Clinical Medical Research Center, Shenzhen People’s Hospital, The Second Clinical Medicine College, JiNan University, Shenzhen, China
Abstract Human adipose tissue-derived stem cells (hASCs) are considered an ideal tool for the supply of insulin-producing cells to treat diabetes mellitus, with high differentiation efficiency. Islet neogenesis-associated protein (INGAP) is an initiator of islet neogenesis, and the peptide sequence comprising amino acids 104–118, named INGAP pentadecapeptide (INGAP-PP), has been shown to increase β-cell mass in animals and human pathological states. Here, we report a novel 4-step method to promote hASCs to differentiate into islet-like clusters (ILCs) more efficiently by adding INGAP-PP. The hASCs were isolated, purified and differentiated using a 4-step protocol including trichostatin A, INGAP-PP/scrambled peptide (Scrambled-P), dexamethasone, nicotinamide, glucagon-like peptide-1, transforming growth factor β1 and exendin-4. Results showed that ILCs in the INGAP-PP group were more similar to the fresh islets with regard to both size and morphology and expressed significantly higher levels of both insulin
© 2014 S. Karger AG, Basel 1422–6405/14/0000–0000$39.50/0 E-Mail [email protected] www.karger.com/cto
and C-peptide than those in the Scrambled-P group. Moreover, the ILCs from the INGAP-PP group secreted higher levels of insulin and C-peptide than those from the Scrambled-P group in response to both a low (5.6 mM) and high (25 mM) glucose challenge and secreted 6 times more hor-
Abbreviations used in this paper
BM-MSCs CK19 DMEM ESCs GLP-1 hASCs ILCs INGAP-PP PBS PDX-1 PE Scrambled-P SVF TSA
bone marrow mesenchymal stem cells cytokeratin 19 Dulbecco’s modified Eagle’s medium embryonic stem cells glucagon-like peptide-1 human adipose tissue-derived stem cells islet-like clusters islet neogenesis-associated protein pentadecapeptide phosphate-buffered saline pancreatic and duodenal homeobox-1 phycoerythrin scrambled peptide stromal vascular fraction trichostatin A
Prof. Feili Gong Department of Immunology, Tongji Medical College Huazhong University of Science and Technology Wuhan 430030 (China) E-Mail flgong @ 163.com Prof. Furong Li Clinical Medical Research Center, Shenzhen People’s Hospital The Second Clinical Medicine College, JiNan University Shenzhen 518020 (China) E-Mail frli62 @ 163.com
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Key Words Adipose tissue-derived stem cells · Differentiation · Transplantation · Diabetes · Islet neogenesis-associated protein
Introduction
Diabetes mellitus is a disease often associated with deficient insulin secretion caused by a decrease in the insulin-producing β-cell mass [Butler et al., 2003]. Many attempts have been made to reestablish the β-cell mass through pancreas or islet transplantation. However, both approaches are limited by the shortage of available organs and the long-term need for immunosuppressive therapy [Shapiro et al., 2000; Atkinson and Eisenbarth, 2001]. Previous studies have reported several sources of insulin-secreting cells. Pluripotent embryonic stem cells (ESCs) could be a potential good source of insulin-producing cells, but their usage is restricted by ethical concerns and risks of teratoma formation [Fujikawa et al., 2005]. Pancreatic epithelium cells have also been used to generate islet-like clusters (ILCs), which partially reversed hyperglycemia in streptozotocin-induced diabetic rats [Ramiya et al., 2000]. However, pancreatic epithelium cells are difficult to obtain and to proliferate efficiently in vitro. Bone marrow mesenchymal stem cells (BM-MSCs) are another alternative source of insulinpositive cells [Oh et al., 2004]. BM-MSCs from rats [Choi et al., 2005] and humans [Moriscot et al., 2005] can differentiate into insulin-secreting cells in vitro and correct hyperglycemia in diabetic mice. Clinical applications of BM-MSCs are limited by availability because harvesting bone marrow involves a highly invasive procedure and only a limited quantity of bone marrow can be acquired from a single donor. Recent reports have indicated that human adipose tissues have sufficient multipotent stem cells. Human adipose tissue is available in large quantities as a waste product of cosmetic liposuction, which involves a less invasive procedure than retrieving human bone marrow. Similar to BM-MSCs, human adipose tissue-derived stem cells (hASCs) [Zuk et al., 2001] are capable of differentiating into adipogenic, chondrogenic, osteogenic, myogenic, 2
Cells Tissues Organs DOI: 10.1159/000362500
neurogenic [Zuk et al., 2002], endothelial [Planat-Benard et al., 2004], hepatic [Seo et al., 2005] and pancreatic [Timper et al., 2006] lineages. Such differentiation can be induced by endogenous or pharmacological compounds either in vivo or in vitro, providing functional β-cells able to correct the decreased β-cell mass and normalize blood glucose levels in various animal models. Among these compounds, a pentadecapeptide comprising the amino acid sequence 104–118 of endogenous islet neogenesisassociated protein (INGAP) represents an attractive therapeutic alternative. INGAP, a member of the Reg-related protein family [Taylor-Fishwick et al., 2003], is an initiator of islet neogenesis in animal models [Rafaeloff et al., 1997; Rosenberg et al., 2004; Barbosa et al., 2006]. INGAP is also found in the pancreas in human pathological states involving islet neogenesis [Vinik et al., 1997], suggesting its early normal presence and possible role in pancreas development and patterning. A pentadecapeptide corresponding to amino acids 104–118 of INGAP (Ile-GlyLeu-His-Asp-Pro-Ser-His-Gly-Thr-Leu-Pro-Asn-GlySer) had similar biological effects to the intact molecule and has a unique insertion of 5 amino acids (Ser-His-GlyThr-Leu) compared with regenerating protein/pancreatic stone protein, a related family of genes. This insertion precedes a potential glycosylation site at position 126 and is hence a core of potential biologic activity [Borelli et al., 2005]. Previous studies have shown that INGAP pentadecapeptide (INGAP-PP) can stimulate an increase in β-cell mass in mice, rats, hamsters and dogs [Rafaeloff et al., 1997; Rosenberg et al., 2004; Barbosa et al., 2006; Pittenger et al., 2007]. Normal neonatal and adult rat islets cultured with INGAP-PP show increased β-cell size and insulin secretion in response to glucose [Rosenberg et al., 2004]. In the present study, we demonstrate a 4-step differentiation protocol for hASCs to differentiate into functional ILCs in defined medium. Our 16-day protocol is based on a previous in vitro differentiation strategy [Seeberger et al., 2006; Okura et al., 2009; Song et al., 2010] including trichostatin A (TSA), INGAP-PP/scrambled peptide (Scrambled-P), dexamethasone, nicotinamide, glucagonlike peptide-1 (GLP-1), transforming growth factor β1 and exendin-4. When transplanted, these ILCs can restore normoglycemia in streptozotocin-induced diabetic mice. Thus, the present findings present a promising step toward cell replacement therapy for diabetes mellitus by autologous transplantation of ILCs differentiated from a patient’s own hASCs.
Ren /Chen /Qi /Li /Gong
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
mones under the high-glucose challenge. Real-time PCR and immunocytochemistry showed that ILCs of the INGAPPP group expressed human pancreatic endocrine hormones and transcription factors. Transplantation of ILCs into diabetic rats partially reversed diabetes and prolonged their life span. In conclusion, the INGAP-PP protocol can efficiently induce hASCs to differentiate into ILCs in vitro, and thus hASCs could be a promising source of cells for transplantation to treat diabetes mellitus. © 2014 S. Karger AG, Basel
Culture of hASCs Human adipose tissue samples were obtained from two donors who provided informed consent. The ethics council of Shenzhen People’s Hospital and Tongji Medical College approved this study. The hASCs were isolated and cultured in low-glucose Dulbecco’s modified Eagle’s medium (DMEM) as described previously by Zuk et al. [2001]. Briefly, to isolate the stromal vascular fraction (SVF), adipose tissue was washed extensively with phosphate-buffered saline (PBS), and extracellular matrix was digested at 37 ° C for 30 min with 0.075% type I collagenase (Sigma, St. Louis, Mo., USA). Enzyme activity was neutralized with DMEM containing 10% FBS, and the tissue was centrifuged at 1,200 rpm for 10 min to obtain the SVF pellet. The pellet was resuspended in 0.16 M NH4Cl for 5 min to lyse red blood cells and then centrifuged at 1,200 rpm for 10 min. The collected SVF pellet was filtered through 100-μm nylon mesh to remove cellular debris and incubated overnight at 37 ° C in 5% CO2 in control medium (DMEM containing 10% FBS, 1% antibiotic/antimycotic solution).
Table 1. List of primers used in this study Gene
Primers
Amplicon Annealing length, bp temperature, ° C
Insulin
GCTGGTAGAGGGAGCAGATG AGCCTTTGTGAACCAACACC
243
58
Glucagon
ACAAGGCAGCTGGCAACGTTCCCT 343 CCTTCCTCCGCCTTTCACCAGCCA
58
Oct3/4
CTT GCT GCA GAA GTG GGT GGA GGA A CTG CAG TGT GGG TTT CGG GCA
169
59
PDX-1
CCTTTCCCATGGATGAAGTC TTCA- 199 ACATGACAGCCAGCTC
54
Fluorescence-Activated Cell Sorting Analysis The fourth passage of hASCs cultured in stem-cell medium were detached, fixed and labeled with anti-CD73-phycoerythrin (PE), CD90-PE (Becton-Dickinson, San Diego, Calif., USA), CD105-PE (DiNonA Inc., Seoul, Korea) and CD45-FITC (BD Pharmingen) antibodies. Labeled cells (1 × 106) in 0.5 ml of 1% paraformaldehyde/PBS were analyzed using a FACSCalibur (Becton-Dickinson). Mouse anti-pancreatic and duodenal homeobox-1 (PDX-1) and mouse anti-cytokeratin 19 (CK19; Chemicon) were used in the intracellular staining protocol. PE-conjugated rabbit anti-mouse IgG (Jackson ImmunoResearch) was used as secondary antibody. Mouse isotype antibodies were used as controls. Western Blot Analysis Western blot analysis was performed according to standard protocols. Lysate samples of hASCs/ESCs containing 10 μg of protein were subjected to Western blot analysis. The following primary antibodies were used: rabbit polyclonal anti-Oct3/4 (1:200), anti-REX-1 (1:100), anti-Nanog (1:200) and goat polyclonal antiactin (1:100). All antibodies were from Santa Cruz Biotechnology (Santa Cruz, Calif., USA). Blots were incubated with horseradish peroxidase-conjugated secondary antibody (1: 2,000; Cell Signaling Technology Inc., Danvers, Mass., USA), and bands were detected using an Amersham ECLTM Western Blotting Analysis System (GE Healthcare, Buckinghamshire, UK). Quantitative Real-Time PCR Total RNA was extracted using a TRI reagent kit (Qiagen) according to the user manual. cDNA was synthesized from 1 μg of RNA using random primers and a RevertAidTM First Strand Synthesis Kit (Fermentas). Quantitative real-time PCR was performed with the QuantiTect SYBR Green RT-PCR Kit (Qiagen) and the ABI PRISM 6700. All gene expression values were normalized to those of β-actin calculated using the 2–ΔΔ method [Moriscot et al., 2005]. Briefly, the first Δ presents the difference between the β-actin threshold cycle and the gene of interest threshold cycle. A sample of human islet was used as a reference. The difference in Δ values
Differentiation of hASCs
Nkx2.2 CGG CGA GTG CTT TTC TCC AA GCG CTT CAT CTT GTA GCG G
165
56
FoxA2
GCG ACC CCA AGA CCT ACA G GGT TCT GCC GGT AGA AGG G
162
56
Pax-6
TGC GAC ATT TCC CGA ATT CT GAT GGA GCC AGT CTC GTA ATA CCT
81
59
REX-1
GCG TCA TAA GGG GTG AGT TTT AGA ACA TTC AAG GGA GCT TGC
134
53
Nanog
TGA TTT GTG GGC CTG AAG AAA 60 A GAG GCA TCT CAG CAG AAG ACA
51
CK19
CAGCAGCGAGCAGGTGTTGA CCAAGGTAGATCTGTGCTTAGC
241
58
Nestin
CAGGAGAAACAGGGCCTACA GTGTCTCAAGGGTAGCAGGC
294
58
β-Actin AGAGCTACGAGCTGCCTGAC AGCACTGTGTTGGCGTACAG
181
58
Nkx2.2 = NK homeobox gene-2.2.
between the ILCs and the islet represents the second Δ. The 2–ΔΔ formula is applied to calculate actual changes in expression levels of genes of interest. The sequences of the primers, the annealing temperatures and the expected amplicon sizes are listed in table 1. INGAP-PP Synthesis Synthesis of the INGAP-PP (NH-Ile-Gly-Leu-His-Asp-ProSer-His-Gly-Thr-Leu-Pro-Asn-Gly-Ser-COOH) was performed on a 431A Applied Biosystems peptide synthesizer using Fmoc solid-phase methodology on p-hydroxymethylphenoxymethyl polystyrene resin at GL Biochem Ltd. (Shanghai, China; ≥95% purity and a molecular weight of 1,501.63). To evaluate the specificity of the INGAP-PP effect, a pentadecapeptide with the same 15 amino acids arranged in a completely different order (Ser-SerThr-Gly-Gly-Gly-Asp-Ile-Pro-Pro-His-Leu-Leu-His-Asn), called Scrambled-P, was used as control. This ‘scrambled peptide’ had been previously reported to be inactive by Rosenberg et al. [2004]. All peptides were reconstituted in isotonic saline to 250 μg/ml.
Cells Tissues Organs DOI: 10.1159/000362500
3
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Materials and Methods
3 days
Stage 1 High-glucose DMEM/F12 medium, 55 nM TSA, 10 mg/l insulin 5 days
Stage 2 induced group (250 ng/ml INGAP-PP)
control group (250 ng/ml Scrambled-P) 5 days
Stage 3 Dexamethasone (10 mM) Nicotinamide (10 mM) 3 days
Stage 4 TGF-DŽ1 (10 ng/ml) GLP-1 (10 nM) Exendin-4 (10 mM)
Fig. 1. Experimental schema of the differentiation of hASCs. The timeline and the induction treatments of the INGAP-PP protocol are illustrated in the figure. The number of days to the right of each arrow indicates the length of the treatment in each stage. At stage 2, the cells were randomly divided into two groups, i.e. the INGAPPP and Scrambled-P groups, the latter serving as control. TGF-β1 = Transforming growth factor β1.
Isolation of Human Islets Human islets were isolated essentially as described elsewhere [Dominici et al., 2006]. Briefly, islets were isolated from pancreata obtained from organ donors after a cold ischemia time of 4.2 ± 1.6 h. The pancreatic duct was perfused with a cold enzyme mixture containing Liberase HI (Roche, Indianapolis, Ind., USA). Tissue was then transferred to a Ricordi chamber and separated by gentle mechanical agitation and enzymatic digestion at 37 ° C. Islets were purified with the use of discontinuous gradients of Ficolldiatrizoic acid in an aphaeresis system (model 2991, Cobe Laboratories, Lakewood, Colo., USA). The discontinuous Ficoll gradient used solution densities of 1.108, 1.096 and 1.037 g/ml layered upon each other before the separation step. During centrifugation, islets migrated to the interfaces between 1.037 and 1.096 g/ml, and between 1.096 and 1.108 g/ml. Islets collected by Ficoll were handpicked under the microscope and maintained in CMRL 1066 medium at 37 ° C in 5% CO2 for a week before use.
Insulin and C-Peptide Quantitation by ELISA To measure secreted insulin and C-peptide levels, 50 handpicked ILCs from each of the INGAP-PP and Scrambled-P groups and fresh islets were preincubated for 90 min at 37 ° C in Hanks’ buffered salt solution supplemented with 2.5 mM glucose. For induced insulin release, cells were further incubated in Hanks’ buffered salt solution supplemented with 5.6 mM glucose or 25 mM glucose for 15 min at 37 ° C. An ELISA kit (Mercodia) was used to determine the amount of secreted insulin and C-peptide.
Differentiation of hASCsA 4-step, 16-day protocol was performed to induce differentiation of hASCs into ILCs. The schema of the experiment is illustrated in figure 1. Briefly, sixth-passage hASCs were expanded to 90% confluency after being cultured for 3 days in high-glucose DMEM (Gibco) supplemented with 55 nM TSA, 10 mg/l insulin and 1% FBS. Afterwards, the cells were randomly divided into two groups and cultured in DMEM with 250 ng/ml INGAP-PP (INGAP-PP group) or 250 ng/ml Scrambled-P (Scrambled-P group) for 5 days. Subsequently, the cells were cultured in low-glucose DMEM supplemented with 10 mM nicotinamide and 10 mM dexamethasone for 5 days. Finally, cells were cultured in low-glucose DMEM supplemented with 10 ng/ml transforming growth factor β1, 10 nM GLP-1 and 10 nM exendin-4 for another 3 days. All cultures were maintained in a 5% CO2 incubator at 37 ° C. Culture medium was changed every 3–4 days until ILCs were assessed.
Immunocytochemistry and Semiquantitative Immunoassay The hASCs and ILCs were fixed with 4% paraformaldehyde for 30 min followed by incubation with mouse anti-insulin (Sigma),
4
Cells Tissues Organs DOI: 10.1159/000362500
Cell Transplantation Diabetes was induced in 8-week-old male Sprague-Dawley rats by streptozotocin (Sigma; 55 mg/kg body weight in citrate buffer, pH 4.4, i.p.). Then, 48 h after injection, animals with blood glucose levels ≥16.65 mM for 2 consecutive days were transplanted with ILCs induced by INGAP-PP (1,000 IEQ/100 μl high-glucose DMEM) into the left subrenal capsule (n = 10). As controls, freshly isolated human islets (1,000 IEQ/100 μl high-glucose DMEM) or saline without cells were transplanted. Two transplantation groups were administered tacrolimus (50 ng/kg/day, p.o.) to avoid immune rejection. To evaluate the effects of INGAP-PP protein on remission of diabetes, we also injected purified INGAP-PP (250 μg/0.5 ml, i.p.) into diabetic rats twice a day for 20 days. Blood glucose levels of all rats were measured every 3 days. Glucose tolerance tests were performed after injection of glucose (2 g/kg body weight,
Ren /Chen /Qi /Li /Gong
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Sixth passage of hASCs
rabbit anti-C-peptide (AbD Serotec), rabbit anti-PDX-1 (Chemicon), rabbit anti-CK19 (Chemicon), rabbit anti-glucagon (Chemicon), rabbit anti-somatostatin (Chemicon) and mouse anti-nestin (Chemicon) antibodies. FITC-labeled rabbit anti-mouse IgG and PE-labeled goat anti-rabbit IgG (Jackson ImmunoResearch) were used as secondary antibodies. Nonimmune serum was used as negative control. Nonspecific staining was excluded using isotypematched irrelevant primary antibodies. These samples were analyzed using a fluorescence microscope (Eclipse TE300, Nikon) and a confocal laser-scanning microscope (FV500, Olympus). A semiquantitative immunofluorescence imaging analysis was performed to determine the amount of insulin-positive cells using the HPIAS-100 color image analyzing system (Qianping, Wuhan, China). Randomly selected (n = 25) but representative areas of each sample were analyzed for the labeling index, defined as the ratio of the positive signal area to the measured area.
Statistical Analysis The results are presented as the mean ± SEM, and statistical comparisons between the INGAP-PP group and the Scrambled-P group were performed using Student’s unpaired t test. p < 0.05 indicated a significant difference.
Results
Characterization of hASCs Twenty-four hours after initial plating of the washed SVF pellet, spindle-shaped primary hASCs grew slowly and reached 90% confluency within 14 days. After the second passage, cells proliferated much more quickly and were subcultured at 1:3 ratios every 3 days. At the end of the fourth passage, hASCs developed a uniform fibroblast-like morphology (fig. 2a). Fluorescence-activated cell sorting analysis of the fourth-passage cells showed that purified hASCs expressed high levels of the cell surface markers CD73 (99.1%), CD90 (97.9%) and CD105 (97.2%). These markers are highly expressed in MSCs [Oh et al., 2004]. hASCs had extremely low expression levels of CK19 and PDX-1, a distinctive feature of pancreatic stem cells. We confirmed that these cells are not hematopoietic stem cells because they are CD45 negative (fig. 2b). To further characterize the hASCs, we examined the expression of markers of undifferentiated ESCs including Oct3/4, Nanog and REX-1. Real-time PCR analysis and Western blot showed that undifferentiated hASCs expressed Oct3/4, Nanog and REX-1, both at the mRNA and protein level (fig. 2c, d). Taken together, these results indicate that hASCs not only have MSC characteristics but also some ESC characteristics and might be an alternative stem cell source, free of the ethical issues impeding the use of ESCs. Differentiation of hASCs with the INGAP-PP Protocol The sixth-passage hASCs were induced according to a 4-step protocol described in detail in Materials and Methods. The cells expanded rapidly in expansion medium and reached 100% confluency in about 3 days. At the end of stage 2, the spindle-shaped cells shrank to a cobblestone-like morphology (fig. 3a). Cell aggregations were observed at about 4 days after initial treatment (fig. 3b). Before step 2 of differentiation, the reconstitutDifferentiation of hASCs
ing cells were equated at 4 ×105 cells/ml in the INGAPPP group and the Scrambled-P group. When the last differentiation step was finished, the production of ILCs in the INGAP-PP group (fig. 3d) comprised 87 ± 12 ILCs/106 cells, significantly higher than the 45 ± 7 ILCs/106 cells produced in the Scrambled-P group (fig. 3c; p < 0.01). Moreover, most of the ILCs in the INGAP-PP group (fig. 3d) were more regular in morphology and more compact in structure than the ILCs in the Scrambled-P group (fig. 3c). Also, there was a significant difference in the size of the ILCs between the two groups (163 ± 26.2 μm for INGAP-PP, 77.4 ± 11.9 μm for Scrambled-P; p < 0.01). Furthermore, figures 3e and f show DTZ-stained ILCs from the INGAP-PP group and freshly isolated islets, which revealed that ILCs from the INGAP-PP group did express insulin, although less than native islets. Expression of Cell Markers in hASCs and ILCs during the Differentiation The immunocytochemistry results (fig. 4) and quantitative real-time PCR results (fig. 5) showed that ILCs from the INGAP-PP group and fresh islets did not express Nanog, Oct3/4 and REX-1 anymore, which are highly expressed by undifferentiated hASCs. At the early stage, the morphology of hASCs changes from spindle-like to epithelial-like cells (fig. 4a, b), and cells start to express the nestin gene (fig. 4a), which is the marker to isolate pancreatic precursor cells and duct epithelium cells, suggesting that there is differentiation from hASCs to ILCs. High-level expression of CK19 (fig. 4b) suggested a possible transition from mesenchymal cells to epithelial cells. Both the quantitative real-time PCR and immunocytochemistry showed pancreas hormone expressions (insulin, glucagon, somatostatin) of the ILCs from the INGAP-PP group, with fresh human islets as control. Quantitative real-time PCR results showed that by adding the INGAP-PP, the differentiation of hASCs to ILCs was significantly promoted in the INGAP-PP group compared to the Scrambled-P group. The expression levels (percentage islet expression) of the endocrine transcription factors (PDX-1, NK homeobox gene-2.2, FoxA2 and Pax-6) in ILCs from the INGAP-PP group were 76.2 ± 2.1, 87.6 ± 3.7, 65.5 ± 2.9 and 63.7 ± 3.2%, respectively, while they were 39.7 ± 2.1, 30 ± 4.7, 18.6 ± 0.5 and 15.9 ± 0.37%, respectively, in the Scrambled-P group. The pancreas hormone expression levels (insulin, glucagon, somatostatin) of the ILCs from the INGAP-PP group were significantly higher than those in the ScramCells Tissues Organs DOI: 10.1159/000362500
5
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
i.p.). Normal rats as well as diabetic rats transplanted with fresh islets were used as positive controls. Diabetic rats were used as negative control. Every group contained 10 rats. Statistical analyses were performed using the SPSS 11.5 software.
Relative mRNA level (%)
120
c
a
Undifferentiated ESCs Undifferentiated hASCs
100 80 60 Oct3/4 40
Undifferentiated ESCs Undifferentiated hASCs
20 0
REX-1 Nanog
Oct3/4
REX-1
Nanog
DŽ-Actin
d
99.1%
97.1%
b
CD105-PE
1.1%
CK19-PE
CD45-FITC
97.5% 0.4%
CD90-PE
PDX-1-PE
Fig. 2. Characterization of the hASCs. a Image of the sixth-passage hASCs. Undifferentiated hASCs turned into a uniform spindleshaped morphology. Scale bar = 50 μm. b Fluorescence-activated cell sorting analysis of different cell markers in hASCs. Sixth-passage hASCs were stained with PE- or FITC-labeled antibodies and analyzed by fluorescence-activated cell sorting. Undifferentiated hASCs were positive for CD73, CD90 and CD105 and negative for
CD45, PDX-1 and CK19. c Quantitative real-time PCR analysis presents the relative expression of undifferentiated hASCs compared to undifferentiated ESCs for Oct3/4, Nanog and REX-1. d Western blot analysis for Oct3/4, Nanog and REX-1. Undifferentiated hASCs and ESCs expressed Oct3/4, Nanog and REX-1 at the protein level.
bled-P group (15.7 ± 3.9, 4.1 ± 1.2 and 5.8 ± 1.9%, respectively, compared to 3.1 ± 0.4, 0.47 ± 0.11 and 1.1 ± 0.21%, respectively). Insulin expression was enhanced from 5–10% of the islets in previous research to about 10–20% in the present
study. Semiquantitative images for insulin labeling results showed 18.9 ± 3.2% (percentage islet expression) in the INGAP-PP group and 5.1 ± 1.1% (percentage islet expression) in the Scrambled-P group, which coincides with the results of quantitative real-time PCR.
6
Cells Tissues Organs DOI: 10.1159/000362500
Ren /Chen /Qi /Li /Gong
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
CD73-PE
2.4%
a
b
c
d
e
f
spindle-shaped hASCs changed into cobblestone-like shapes at the end of stage 2. b Cells started to aggregate after stage 1 and stage 2 inductions. c ILCs formed in the Scrambled-P group. d ILCs formed in the INGAP-PP group. e DTZ-stained ILCs formed in the INGAP-PP group. f DTZstained fresh islet. ×100. Scale bars = 50 μm.
ILCs Produced Using the INGAP-PP Protocol Released Higher Amounts of Insulin and C-Peptide in Response to Glucose To determine the glucose response of ILCs, insulin and C-peptide secretion in the presence of low (5.6 mM)
and high (25 mM) glucose concentrations were measured using ELISA. Islets and ILCs from both the INGAP-PP and Scrambled-P groups could all secrete insulin and C-peptide in response to glucose, but the levels of secretion varied. ILCs from the INGAP-PP group se-
Differentiation of hASCs
Cells Tissues Organs DOI: 10.1159/000362500
7
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Fig. 3. Images of hASCs or ILCs at different stages of the INGAP-PP protocol. a The
Nestin
50.0 μm
Insulin
50.0 μm
Insulin
e
c
50.0 μm
h
50.0 μm
C-peptide
50.0 μm
f
Merge
g
50.0 μm
Insulin/glucagon
50.0 μm
i
50.0 μm
Insulin
Insulin/somatostatin
8
20.0 μm
Merge
d
j
b
50.0 μm
Cells Tissues Organs DOI: 10.1159/000362500
k
50.0 μm
(For figure 4l and legend see next page.)
Ren /Chen /Qi /Li /Gong
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
a
4
PDX-1
CK19
***#
1.0
tive image for insulin labeling. Semiquantitative immunofluorescence analysis was performed to estimate the labeling index of insulin-producing cells derived from hASCs by different treatment. Twenty-five areas were chosen each time. Each value (n = 3 experiments) represents the mean ± SEM. ** p < 0.001, *** p < 0.01 versus Scrambled-P. # p < 0.05 versus hASCs.
0.7 0.6 0.5 0.4 0.3
90
** #
0.2 0.1
l
80
#
0
INGAP-PP group Scrambled-P group
hASCs
INGAP-PP
Scrambled-P
Islets
***
60
**
**
-6
**
70
Fo xA 2
** **
50 40 30
4
1 XRE
3/ ct O
og an N
Pa x
2 2.
ratios between rats transplanted with ILCs induced by INGAP-PP and normal islets. Because of the poor induction results in the Scrambled-P group, the transplantation step excluded the ILCs from the Scrambled-P group. Streptozotocin-induced diabetic rats often have a blood glucose level higher than 16.7 mM. Blood glucose levels of rats transplanted with INGAP-induced ILCs fell to 7.8 ± 0.8 mM and stayed at this level for the entire course of the experiment. Diabetic rats transplanted with normal islets, as a control, showed normal blood glucose levels after 1–2 weeks. Nontransplanted diabetic rats remained hyperglycemic and subsequently died. Interestingly, rats injected with INGAP-PP also showed Cells Tissues Organs DOI: 10.1159/000362500
9
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Remission of Diabetes by Transplantation of INGAP-Induced Cells To evaluate whether ILCs could reverse diabetes in rats, we compared the blood glucose levels and survival
CK 19 N es tin
** kx
**
1
0
**
N
10
PD X-
20
creted a significantly higher amount of insulin and Cpeptide when compared to ILCs from the Scrambled-P group. The levels of insulin and C-peptide release were about 10% of that of normal islets. Moreover, the ILCs from the INGAP-PP group secreted 6–7 times more insulin and C-peptide under high-glucose than low-glucose conditions, similar to the change in the islets group (fig. 6a, b).
Differentiation of hASCs
0.8
In su lin Gl uc ag So on m at os ta tin
vious immunocytochemical results, endocrine hormone (insulin, glucagon, somatostatin) expression levels in ILCs of the INGAP-PP group were compared to those in islets. ESC markers (REX-1, Oct3/4 and Nanog) were undetectable in both ILCs and islets. Human islet precursor cell markers (nestin and CK19) were detected, and while most endocrine transcription factors assessed [NK homeobox gene-2.2 (Nkx2.2), Pax-6, FoxA2] were found in islets and ILCs, the expression of one such factor that has also been proposed as a progenitor cell marker (PDX-1) was somewhat maintained in both ILCs and islets at a similar level. Data are presented as means ± SEMs of 3 independent experiments. ** p < 0.01, *** p < 0.001.
100 Expression/DŽ-actin (% islet expression)
Fig. 5. Real-time PCR. Confirming the pre-
0.9
Labeling index
Fig. 4. Immunocytochemical analysis of genes related to pancreatic development. a–j Gene expression levels in the INGAP-PP group. a Nestin (green). b CK19 (red). c PDX-1 (red). d Insulin (green). e Merge of c and d. f C-peptide (red). g Insulin (green). h Merge of f and g. i Double staining of insulin (green) and glucagon (red). j Double staining of insulin (green) and somatostatin (red). a, b Cells harvested on day 7. ×400. c–h Cells harvested on day 16. ×100. k Insulin expression of islet (green). l Semiquantita-
Insulin (mIU/l)
40
45.6
5,000
30 20
** 7.8
10 0
a –10
0.74
* 0.18
6,000
**
INGAP-PP group Scrambled-P group Islets
C-peptide (ng/l)
50
4.8
5.6 mM
*
1.6 25 mM
INGAP-PP group Scrambled-P group Islets
**
4,860.5
4,000 3,000 2,000 1,000 0
b –1,000
**
812.1
654.7
114.3 5.6 mM
*
25 mM
Fig. 6. Secretion of insulin (a) and C-peptide (b) in response to 5.6 or 25 mM glucose in INGAP-PP and Scram-
bled-P groups and normal human islets (n = 3). Each value represents the mean ± SEM. * p < 0.05, ** p < 0.01 (Student’s t test).
Discussion
Insulin-producing cell transplantation has been shown to be the most promising therapy to cure insulin-dependent diabetes. Though ESCs, pancreatic epithelium stem cells, BMSCs and hASCs are all capable of differentiating into insulin-secreting cells or even ILCs [Tang et al., 2004; Choi et al., 2005; Moriscot et al., 2005], the efficiency of differentiation was too low to produce adequate functional cells for clinical applications.
10
Cells Tissues Organs DOI: 10.1159/000362500
To overcome this obstacle, we developed a 4-step strategy to induce ILC formation based on a modified protocol used by previous studies [Vinik et al., 1997; Rosenberg et al., 2004; Timper et al., 2006]. Our strategy consisted of 4 stages, each involving different kinds of reagents to induce hASC differentiation. TSA, a histone deacetylase inhibitor, was added to the culture medium to provide better access to the chromatin. Histone acetylation decreases the ability of histones to bind DNA, causing a more ‘open’ chromatin state. Inhibition of histone deacetylase prevents histone deacetylation and therefore can alter gene expression patterns and help trans-acting factors to bind their target DNA sequences. INGAP-PP and transcription factors exendin-4 and GLP-1 all play important roles in pancreas development. Sequentially stimulating hASCs with these reagents helped efficient differentiation into functional ILCs. By examining the molecular marker of differentiated cells, we found that hASCs might differentiate into functional ILCs through a mesenchymal-to-epithelial transition. This transition has been reported to be present during the differentiation of human islet precursor cells into insulin-producing cells [Gershengorn et al., 2004]. Our results also demonstrated that the expression levels of early developmental-stage pancreatic genes, such as CK19, Nestin and PDX-1, were similar in the INGAP-PP and Scrambled-P groups. In contrast, genes related to pancreatic function, such as insulin and C-peptide, were expressed at significantly higher levels in the INGAP-PP Ren /Chen /Qi /Li /Gong
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
a decrease in blood glucose level after 5 days of injections. However, the glucose level increased again 2 weeks after INGAP-PP injections were discontinued (fig. 7a). Diabetic rats transplanted with INGAP-PP-induced ILCs or normal islets survived the whole 69-day process, while diabetic rats without treatment suffered from hyperglycemia and died within 51 days. All of the diabetic rats who received INGAP-PP injections also survived until the end of the experiment (fig. 7b) Intraperitoneal glucose tolerance tests showed that Sprague-Dawley rats transplanted with ILCs induced by INGAP-PP had similar kinetics of glucose clearance to the group transplanted with normal islets, although the blood glucose levels were higher at all time points. In contrast, untreated diabetic rats revealed no reaction to the high glucose concentration (fig. 7c).
40
Blood glucose (mM)
35 30
**
25
** **
**
** ** ** ** **
** **
**
20 15 10 5 0 0
a
3 6
9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 Days after transplantation
1.0 0.9 0.8
Diabetic rats Diabetic rats + INGAP-PP-induced cells Diabetic rats + INGAP-PP injection Diabetic rats + normal islets Normal rats
Survival ratio
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
b
Fig. 7. Remission of diabetes by transplantation of ILCs into diabetic rats. a Blood
10
20 30 40 50 Days after transplantation
40 35 Blood glucose (mM)
**
30
**
70
60
**
**
25
**
20 15 10 5 0
c
0
20
40 60 80 Time after glucose loading (min)
100
120
group than in the Scrambled-P group. These results strongly suggest an essential role of INGAP-PP in stimulating the maturation and insulin secretion ability of β-cells rather than initiating pancreatic lineage differentiation.
The result that ILCs from the INGAP-PP group are more similar to fresh islets is consistent with the previous conclusion that INGAP-PP promotes maturation of pancreas and insulin secretion [Borelli et al., 2005]. Although levels of insulin and C-peptide secreted by them
Differentiation of hASCs
Cells Tissues Organs DOI: 10.1159/000362500
11
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
glucose levels of transplanted rats. ILCs and normal islets were transplanted under the kidney capsule of streptozotocin-induced diabetic rats (blood glucose levels ≥16.7 mM), and blood glucose levels were measured every 3 days. The arrow shows that the kidney containing the graft was removed 60 days after transplantation. b Survival rates of transplanted rats. c Intraperitoneal glucose tolerance test. The tests were performed 4 weeks after transplantation. Each value represents the mean ± SEM. ** p < 0.01 (Student’s t test). n = 10/group.
0
were revealed to be only 10–20% of those of the normal islets by quantitative PCR and immunocytochemistry and about 10% by the glucose-stimulating test, it was a promising improvement compared with the approximately 2–5% secretion level reported previously [Lumelsky et al., 2001]. ILC transplantation into diabetic rats did indeed decrease the blood glucose levels and extend the life span of the diabetic rats. However, disease remission was not as complete as with fresh islet transplantation. This was probably due to the fact that the amount of ILCs was not sufficient to compensate the function of destroyed pancreas. Consistent with in vitro glucose stimulation test results, intraperitoneal glucose tolerance tests showed that ILCs upregulated insulin secretion and reversed hyperglycemia. In addition, we have to say that the effects we got in vivo experiments are much more satisfactory than those in vitro, the reason may lie in the continuous differentiation of ILCs. Interestingly, INGAP-PP injection alone led to a decrease in blood glucose level in diabetic rats 5–6 days after the injections. However, the blood glucose level increased
2 weeks after termination of the injections. These results may imply that although INGAP-PP could promote islet neogenesis, the regeneration islets were subsequently destroyed in diabetic rats. In summary, induced production of ILCs or insulinproducing cells represents an area of increasing interest in regenerative medicine. With our 4-stage INGAP-PP protocol, we enhanced the differentiation efficiency of hASCs into insulin/C-peptide-positive ILCs, the transplantation of which partially reversed diabetes in an animal model. Thus, our studies suggest hASCs as an alternative, autologous source for cell replacement therapy in diabetes.
Acknowledgements This research was supported by the 973 special plan of China (No. 2007CB516811) and the key Science and Technology plan of Shenzhen (No. 200901001). We thank colleagues in our laboratory for technical assistance and advice. We appreciate ChunYan Deng for providing technical support for flow cytometry analysis.
References
12
Cells Tissues Organs DOI: 10.1159/000362500
Fujikawa, T., S.H. Oh, L. Pi, H.M. Hatch, T. Shupe, B.E. Petersen (2005) Teratoma formation leads to failure of treatment for type I diabetes using embryonic stem cell-derived insulin-producing cells. Am J Pathol 166: 1781– 1791. Gershengorn, M.C., A.A. Hardikar, C. Wei, E. Geras-Raaka, B. Marcus-Samuels, B.M. Raaka (2004) Epithelial-to-mesenchymal transition generates proliferative human islet precursor cells. Science 306: 2261–2264. Lumelsky, N., O. Blondel, P. Laeng, I. Velasco, R. Ravin, R. McKay (2001) Differentiation of embryonic stem cells to insulin-secreting structures similar to pancreatic islets. Science 292: 1389–1394. Moriscot, C., F. Fraipont, M.J. Richard, M. Marchand, P. Savatier, D. Bosco, M. Favrot, P.Y. Benhamou (2005) Human bone marrow mesenchymal stem cells can express insulin and key transcription factors of the endocrine pancreas developmental pathway upon genetic and/or microenvironmental manipulation in vitro. Stem Cells 23: 594–603. Oh, S.H., T.M. Muzzonigro, S.H. Bae, J.M. LaPlante, H.M. Hatch, B.E. Petersen (2004) Adult bone marrow-derived cells trans-differentiating into insulin-producing cells for the treatment of type I diabetes. Lab Invest 84: 607–617.
Okura, H., H. Komoda, Y. Fumimoto, C.M. Lee, T. Nishida, Y. Sawa, A. Matsuyama (2009) Transdifferentiation of human adipose tissue-derived stromal cells into insulin-producing clusters. J Artif Organs 12: 123–130. Pittenger, G.L., D.A. Taylor-Fishwick, R.H. Johns, N. Burcus, S. Kosuri, A.I. Vinik (2007) Intramuscular injection of islet neogenesisassociated protein peptide stimulates pancreatic islet neogenesis in healthy dogs. Pancreas 34: 103–111. Planat-Benard, V.P., J.S. Silvestre, B. Cousin, M. Andre, M. Nibbelink, R. Tamarat, M. Clergue, C. Manneville, C. Saillan-Barreau, M. Duriez, A. Tedgui, B. Levy, L. Penicaud, L. Casteilla (2004) Plasticity of human adipose lineage cells toward endothelial cells: physiological and therapeutic perspectives. Circulation 109: 656–663. Rafaeloff, R, G.L. Pittenger, S.W. Barlow, X.F. Qin, B. Yan, L. Rosenberg (1997) Cloning and sequencing of the pancreatic islet neogenesis associated protein (INGAP) gene and its expression in islet neogenesis in hamsters. J Clin Invest 99: 2100–2109. Ramiya, V.K., M. Maraist, K.E. Arfors, D.A. Schatz, A.B. Peck, J.G. Cornelius (2000) Reversal of insulin-dependent diabetes using islets generated in vitro from pancreatic stem cells. Nat Med 6: 278–282.
Ren /Chen /Qi /Li /Gong
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Atkinson, M.A., G.S. Eisenbarth (2001) Type 1 diabetes: new perspectives on disease pathogenesis and treatment. Lancet 358: 221–229. Barbosa, H., S. Bordin, L. Stoppiglia, K. Silva, M. Borelli, H. Del Zotto, et al. (2006) Islet Neogenesis Associated Protein (INGAP) modulates gene expression in cultured neonatal rat islets. Regul Pept 136: 78–84. Borelli, M.I., L.F. Stoppiglia, L.F. Rezende, L.E. Flores, H. Del Zotto, A.C. Boschero, et al. (2005) INGAP-related pentadecapeptide: its modulatory effect upon insulin secretion. Regul Pept 131: 97–102. Butler, A.E., J. Janson, S. Bonner-Weir, R. Ritzel, R.A. Rizza, P.C. Butler (2003) Beta-cell deficit and increased beta-cell apoptosis in humans with type 2 diabetes. Diabetes 52: 102–110. Choi, K.S., J.S. Shin, J.J. Lee, Y.S. Kim, S.B. Kim, C.W. Kim (2005) In vitro trans-differentiation of rat mesenchymal cells into insulinproducing cells by rat pancreatic extract. Biochem Biophys Res Commun 330: 1299–1305. Dominici, M., K. LeBlanc, I. Mueller, I. SlaperCortenbach, F. Marini, D. Krause, R. Deans, A. Keating, D. Prockop, E. Horwitz (2006) Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy 8: 315–317.
Differentiation of hASCs
Song, C. K., J.H. Duck, J.Y. Lee (2010) Adipose tissue derived stem cells for regeneration and differentiation into insulin-producing cells. Curr Stem Cell Res Ther 5: 190–194. Tang, D.Q., L.Z. Cao, B.R. Burkhardt, C.Q. Xia, S.A. Litherland, M.A. Atkinson, L.J. Yang (2004) In vivo and in vitro characterization of insulin-producing cells obtained from murine bone marrow. Diabetes 53: 1721–1732. Taylor-Fishwick, D.A., S. Rittman, H. Kendall, L. Roy, W. Shi, Y. Cao (2003) Cloning genomic INGAP: a Reg-related family member with distinct transcriptional regulation sites. Biochim Biophys Acta 1638: 83–89.
Cells Tissues Organs DOI: 10.1159/000362500
Timper, K., D. Seboek, M. Eberhardt, P. Linscheid, M. Christ-Crain, U. Keller, B. Müller, H. Zulewski (2006) Human adipose tissuederived mesenchymal stem cells differentiate into insulin, somatostatin, and glucagons expressing cells. Biochem Biophys Res Commun 341: 1135–1140. Vinik, A, R. Rafaeloff, G. Pittenger, L. Rosenberg, W. Duguid (1997) Inducement of pancreatic islet neogenesis. Horm Metab Res 29: 278– 293. Zuk, P.A., N Zhu, H. Mizuno, J. Huang, J.W. Futrell, A.J. Katz, P. Benhaim, H.P. Lorenz, M.H. Hedrick (2001) Multilineage cells from human adipose tissue: implications for cellbased therapies. Tissue Eng 7: 211–228. Zuk, P.A., M. Zhu, P. Ashjian, J.I. De Ugarte, D.A. Huang, H. Mizuno, Z.C. Alfonso, J.K. Fraser, P. Benhaim, M.H. Hedrick (2002) Human adipose tissue is a source of multipotent stem cells. Mol Biol Cell 13: 4279–4295.
13
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Rosenberg, L., M. Lipsett, J.W. Yoon, M. Prentki, R. Wang, H.S. Jun, et al. (2004) A pentadecapeptide fragment of islet neogenesis-associated protein increases beta-cell mass and reverses diabetes in C57BL/6J mice. Ann Surg 240: 875–884. Seeberger, K.L., J.M. Dufour, A.M. Shapiro, J.R. Lakey, R.V. Rajotte, G.S. Korbutt (2006) Expansion of mesenchymal stem cells from human pancreatic ductal epithelium. Lab Invest 86: 141–153. Seo, M.J., S.Y. Suh, Y.C. Bae, J.S. Jung (2005) Differentiation of human adipose stromal cells into hepatic lineage in vitro and in vivo. Biochem Biophys Res Commun 328: 258–264. Shapiro, A.M., J.R. Lakey, E.A. Ryan, G.S. Korbutt, E. Toth, G.L. Warnock, N.M. Kneteman, R.V. Rajotte (2000) Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Eng J Med 343: 230–238.
Cells Tissues Organs DOI: 10.1159/000362500
In vitro Differentiation of Human Adipose Tissue-Derived Stem Cells into Islet-Like Clusters Promoted by Islet Neogenesis-Associated Protein Pentadecapeptide Lili Ren a, b Lijuan Chen b Hui Qi b Furong Li b Feili Gong a
a
Department of Immunology, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, and b Clinical Medical Research Center, Shenzhen People’s Hospital, The Second Clinical Medicine College, JiNan University, Shenzhen, China
Abstract Human adipose tissue-derived stem cells (hASCs) are considered an ideal tool for the supply of insulin-producing cells to treat diabetes mellitus, with high differentiation efficiency. Islet neogenesis-associated protein (INGAP) is an initiator of islet neogenesis, and the peptide sequence comprising amino acids 104–118, named INGAP pentadecapeptide (INGAP-PP), has been shown to increase β-cell mass in animals and human pathological states. Here, we report a novel 4-step method to promote hASCs to differentiate into islet-like clusters (ILCs) more efficiently by adding INGAP-PP. The hASCs were isolated, purified and differentiated using a 4-step protocol including trichostatin A, INGAP-PP/scrambled peptide (Scrambled-P), dexamethasone, nicotinamide, glucagon-like peptide-1, transforming growth factor β1 and exendin-4. Results showed that ILCs in the INGAP-PP group were more similar to the fresh islets with regard to both size and morphology and expressed significantly higher levels of both insulin
© 2014 S. Karger AG, Basel 1422–6405/14/0000–0000$39.50/0 E-Mail [email protected] www.karger.com/cto
and C-peptide than those in the Scrambled-P group. Moreover, the ILCs from the INGAP-PP group secreted higher levels of insulin and C-peptide than those from the Scrambled-P group in response to both a low (5.6 mM) and high (25 mM) glucose challenge and secreted 6 times more hor-
Abbreviations used in this paper
BM-MSCs CK19 DMEM ESCs GLP-1 hASCs ILCs INGAP-PP PBS PDX-1 PE Scrambled-P SVF TSA
bone marrow mesenchymal stem cells cytokeratin 19 Dulbecco’s modified Eagle’s medium embryonic stem cells glucagon-like peptide-1 human adipose tissue-derived stem cells islet-like clusters islet neogenesis-associated protein pentadecapeptide phosphate-buffered saline pancreatic and duodenal homeobox-1 phycoerythrin scrambled peptide stromal vascular fraction trichostatin A
Prof. Feili Gong Department of Immunology, Tongji Medical College Huazhong University of Science and Technology Wuhan 430030 (China) E-Mail flgong @ 163.com Prof. Furong Li Clinical Medical Research Center, Shenzhen People’s Hospital The Second Clinical Medicine College, JiNan University Shenzhen 518020 (China) E-Mail frli62 @ 163.com
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Key Words Adipose tissue-derived stem cells · Differentiation · Transplantation · Diabetes · Islet neogenesis-associated protein
Introduction
Diabetes mellitus is a disease often associated with deficient insulin secretion caused by a decrease in the insulin-producing β-cell mass [Butler et al., 2003]. Many attempts have been made to reestablish the β-cell mass through pancreas or islet transplantation. However, both approaches are limited by the shortage of available organs and the long-term need for immunosuppressive therapy [Shapiro et al., 2000; Atkinson and Eisenbarth, 2001]. Previous studies have reported several sources of insulin-secreting cells. Pluripotent embryonic stem cells (ESCs) could be a potential good source of insulin-producing cells, but their usage is restricted by ethical concerns and risks of teratoma formation [Fujikawa et al., 2005]. Pancreatic epithelium cells have also been used to generate islet-like clusters (ILCs), which partially reversed hyperglycemia in streptozotocin-induced diabetic rats [Ramiya et al., 2000]. However, pancreatic epithelium cells are difficult to obtain and to proliferate efficiently in vitro. Bone marrow mesenchymal stem cells (BM-MSCs) are another alternative source of insulinpositive cells [Oh et al., 2004]. BM-MSCs from rats [Choi et al., 2005] and humans [Moriscot et al., 2005] can differentiate into insulin-secreting cells in vitro and correct hyperglycemia in diabetic mice. Clinical applications of BM-MSCs are limited by availability because harvesting bone marrow involves a highly invasive procedure and only a limited quantity of bone marrow can be acquired from a single donor. Recent reports have indicated that human adipose tissues have sufficient multipotent stem cells. Human adipose tissue is available in large quantities as a waste product of cosmetic liposuction, which involves a less invasive procedure than retrieving human bone marrow. Similar to BM-MSCs, human adipose tissue-derived stem cells (hASCs) [Zuk et al., 2001] are capable of differentiating into adipogenic, chondrogenic, osteogenic, myogenic, 2
Cells Tissues Organs DOI: 10.1159/000362500
neurogenic [Zuk et al., 2002], endothelial [Planat-Benard et al., 2004], hepatic [Seo et al., 2005] and pancreatic [Timper et al., 2006] lineages. Such differentiation can be induced by endogenous or pharmacological compounds either in vivo or in vitro, providing functional β-cells able to correct the decreased β-cell mass and normalize blood glucose levels in various animal models. Among these compounds, a pentadecapeptide comprising the amino acid sequence 104–118 of endogenous islet neogenesisassociated protein (INGAP) represents an attractive therapeutic alternative. INGAP, a member of the Reg-related protein family [Taylor-Fishwick et al., 2003], is an initiator of islet neogenesis in animal models [Rafaeloff et al., 1997; Rosenberg et al., 2004; Barbosa et al., 2006]. INGAP is also found in the pancreas in human pathological states involving islet neogenesis [Vinik et al., 1997], suggesting its early normal presence and possible role in pancreas development and patterning. A pentadecapeptide corresponding to amino acids 104–118 of INGAP (Ile-GlyLeu-His-Asp-Pro-Ser-His-Gly-Thr-Leu-Pro-Asn-GlySer) had similar biological effects to the intact molecule and has a unique insertion of 5 amino acids (Ser-His-GlyThr-Leu) compared with regenerating protein/pancreatic stone protein, a related family of genes. This insertion precedes a potential glycosylation site at position 126 and is hence a core of potential biologic activity [Borelli et al., 2005]. Previous studies have shown that INGAP pentadecapeptide (INGAP-PP) can stimulate an increase in β-cell mass in mice, rats, hamsters and dogs [Rafaeloff et al., 1997; Rosenberg et al., 2004; Barbosa et al., 2006; Pittenger et al., 2007]. Normal neonatal and adult rat islets cultured with INGAP-PP show increased β-cell size and insulin secretion in response to glucose [Rosenberg et al., 2004]. In the present study, we demonstrate a 4-step differentiation protocol for hASCs to differentiate into functional ILCs in defined medium. Our 16-day protocol is based on a previous in vitro differentiation strategy [Seeberger et al., 2006; Okura et al., 2009; Song et al., 2010] including trichostatin A (TSA), INGAP-PP/scrambled peptide (Scrambled-P), dexamethasone, nicotinamide, glucagonlike peptide-1 (GLP-1), transforming growth factor β1 and exendin-4. When transplanted, these ILCs can restore normoglycemia in streptozotocin-induced diabetic mice. Thus, the present findings present a promising step toward cell replacement therapy for diabetes mellitus by autologous transplantation of ILCs differentiated from a patient’s own hASCs.
Ren /Chen /Qi /Li /Gong
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
mones under the high-glucose challenge. Real-time PCR and immunocytochemistry showed that ILCs of the INGAPPP group expressed human pancreatic endocrine hormones and transcription factors. Transplantation of ILCs into diabetic rats partially reversed diabetes and prolonged their life span. In conclusion, the INGAP-PP protocol can efficiently induce hASCs to differentiate into ILCs in vitro, and thus hASCs could be a promising source of cells for transplantation to treat diabetes mellitus. © 2014 S. Karger AG, Basel
Culture of hASCs Human adipose tissue samples were obtained from two donors who provided informed consent. The ethics council of Shenzhen People’s Hospital and Tongji Medical College approved this study. The hASCs were isolated and cultured in low-glucose Dulbecco’s modified Eagle’s medium (DMEM) as described previously by Zuk et al. [2001]. Briefly, to isolate the stromal vascular fraction (SVF), adipose tissue was washed extensively with phosphate-buffered saline (PBS), and extracellular matrix was digested at 37 ° C for 30 min with 0.075% type I collagenase (Sigma, St. Louis, Mo., USA). Enzyme activity was neutralized with DMEM containing 10% FBS, and the tissue was centrifuged at 1,200 rpm for 10 min to obtain the SVF pellet. The pellet was resuspended in 0.16 M NH4Cl for 5 min to lyse red blood cells and then centrifuged at 1,200 rpm for 10 min. The collected SVF pellet was filtered through 100-μm nylon mesh to remove cellular debris and incubated overnight at 37 ° C in 5% CO2 in control medium (DMEM containing 10% FBS, 1% antibiotic/antimycotic solution).
Table 1. List of primers used in this study Gene
Primers
Amplicon Annealing length, bp temperature, ° C
Insulin
GCTGGTAGAGGGAGCAGATG AGCCTTTGTGAACCAACACC
243
58
Glucagon
ACAAGGCAGCTGGCAACGTTCCCT 343 CCTTCCTCCGCCTTTCACCAGCCA
58
Oct3/4
CTT GCT GCA GAA GTG GGT GGA GGA A CTG CAG TGT GGG TTT CGG GCA
169
59
PDX-1
CCTTTCCCATGGATGAAGTC TTCA- 199 ACATGACAGCCAGCTC
54
Fluorescence-Activated Cell Sorting Analysis The fourth passage of hASCs cultured in stem-cell medium were detached, fixed and labeled with anti-CD73-phycoerythrin (PE), CD90-PE (Becton-Dickinson, San Diego, Calif., USA), CD105-PE (DiNonA Inc., Seoul, Korea) and CD45-FITC (BD Pharmingen) antibodies. Labeled cells (1 × 106) in 0.5 ml of 1% paraformaldehyde/PBS were analyzed using a FACSCalibur (Becton-Dickinson). Mouse anti-pancreatic and duodenal homeobox-1 (PDX-1) and mouse anti-cytokeratin 19 (CK19; Chemicon) were used in the intracellular staining protocol. PE-conjugated rabbit anti-mouse IgG (Jackson ImmunoResearch) was used as secondary antibody. Mouse isotype antibodies were used as controls. Western Blot Analysis Western blot analysis was performed according to standard protocols. Lysate samples of hASCs/ESCs containing 10 μg of protein were subjected to Western blot analysis. The following primary antibodies were used: rabbit polyclonal anti-Oct3/4 (1:200), anti-REX-1 (1:100), anti-Nanog (1:200) and goat polyclonal antiactin (1:100). All antibodies were from Santa Cruz Biotechnology (Santa Cruz, Calif., USA). Blots were incubated with horseradish peroxidase-conjugated secondary antibody (1: 2,000; Cell Signaling Technology Inc., Danvers, Mass., USA), and bands were detected using an Amersham ECLTM Western Blotting Analysis System (GE Healthcare, Buckinghamshire, UK). Quantitative Real-Time PCR Total RNA was extracted using a TRI reagent kit (Qiagen) according to the user manual. cDNA was synthesized from 1 μg of RNA using random primers and a RevertAidTM First Strand Synthesis Kit (Fermentas). Quantitative real-time PCR was performed with the QuantiTect SYBR Green RT-PCR Kit (Qiagen) and the ABI PRISM 6700. All gene expression values were normalized to those of β-actin calculated using the 2–ΔΔ method [Moriscot et al., 2005]. Briefly, the first Δ presents the difference between the β-actin threshold cycle and the gene of interest threshold cycle. A sample of human islet was used as a reference. The difference in Δ values
Differentiation of hASCs
Nkx2.2 CGG CGA GTG CTT TTC TCC AA GCG CTT CAT CTT GTA GCG G
165
56
FoxA2
GCG ACC CCA AGA CCT ACA G GGT TCT GCC GGT AGA AGG G
162
56
Pax-6
TGC GAC ATT TCC CGA ATT CT GAT GGA GCC AGT CTC GTA ATA CCT
81
59
REX-1
GCG TCA TAA GGG GTG AGT TTT AGA ACA TTC AAG GGA GCT TGC
134
53
Nanog
TGA TTT GTG GGC CTG AAG AAA 60 A GAG GCA TCT CAG CAG AAG ACA
51
CK19
CAGCAGCGAGCAGGTGTTGA CCAAGGTAGATCTGTGCTTAGC
241
58
Nestin
CAGGAGAAACAGGGCCTACA GTGTCTCAAGGGTAGCAGGC
294
58
β-Actin AGAGCTACGAGCTGCCTGAC AGCACTGTGTTGGCGTACAG
181
58
Nkx2.2 = NK homeobox gene-2.2.
between the ILCs and the islet represents the second Δ. The 2–ΔΔ formula is applied to calculate actual changes in expression levels of genes of interest. The sequences of the primers, the annealing temperatures and the expected amplicon sizes are listed in table 1. INGAP-PP Synthesis Synthesis of the INGAP-PP (NH-Ile-Gly-Leu-His-Asp-ProSer-His-Gly-Thr-Leu-Pro-Asn-Gly-Ser-COOH) was performed on a 431A Applied Biosystems peptide synthesizer using Fmoc solid-phase methodology on p-hydroxymethylphenoxymethyl polystyrene resin at GL Biochem Ltd. (Shanghai, China; ≥95% purity and a molecular weight of 1,501.63). To evaluate the specificity of the INGAP-PP effect, a pentadecapeptide with the same 15 amino acids arranged in a completely different order (Ser-SerThr-Gly-Gly-Gly-Asp-Ile-Pro-Pro-His-Leu-Leu-His-Asn), called Scrambled-P, was used as control. This ‘scrambled peptide’ had been previously reported to be inactive by Rosenberg et al. [2004]. All peptides were reconstituted in isotonic saline to 250 μg/ml.
Cells Tissues Organs DOI: 10.1159/000362500
3
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Materials and Methods
3 days
Stage 1 High-glucose DMEM/F12 medium, 55 nM TSA, 10 mg/l insulin 5 days
Stage 2 induced group (250 ng/ml INGAP-PP)
control group (250 ng/ml Scrambled-P) 5 days
Stage 3 Dexamethasone (10 mM) Nicotinamide (10 mM) 3 days
Stage 4 TGF-DŽ1 (10 ng/ml) GLP-1 (10 nM) Exendin-4 (10 mM)
Fig. 1. Experimental schema of the differentiation of hASCs. The timeline and the induction treatments of the INGAP-PP protocol are illustrated in the figure. The number of days to the right of each arrow indicates the length of the treatment in each stage. At stage 2, the cells were randomly divided into two groups, i.e. the INGAPPP and Scrambled-P groups, the latter serving as control. TGF-β1 = Transforming growth factor β1.
Isolation of Human Islets Human islets were isolated essentially as described elsewhere [Dominici et al., 2006]. Briefly, islets were isolated from pancreata obtained from organ donors after a cold ischemia time of 4.2 ± 1.6 h. The pancreatic duct was perfused with a cold enzyme mixture containing Liberase HI (Roche, Indianapolis, Ind., USA). Tissue was then transferred to a Ricordi chamber and separated by gentle mechanical agitation and enzymatic digestion at 37 ° C. Islets were purified with the use of discontinuous gradients of Ficolldiatrizoic acid in an aphaeresis system (model 2991, Cobe Laboratories, Lakewood, Colo., USA). The discontinuous Ficoll gradient used solution densities of 1.108, 1.096 and 1.037 g/ml layered upon each other before the separation step. During centrifugation, islets migrated to the interfaces between 1.037 and 1.096 g/ml, and between 1.096 and 1.108 g/ml. Islets collected by Ficoll were handpicked under the microscope and maintained in CMRL 1066 medium at 37 ° C in 5% CO2 for a week before use.
Insulin and C-Peptide Quantitation by ELISA To measure secreted insulin and C-peptide levels, 50 handpicked ILCs from each of the INGAP-PP and Scrambled-P groups and fresh islets were preincubated for 90 min at 37 ° C in Hanks’ buffered salt solution supplemented with 2.5 mM glucose. For induced insulin release, cells were further incubated in Hanks’ buffered salt solution supplemented with 5.6 mM glucose or 25 mM glucose for 15 min at 37 ° C. An ELISA kit (Mercodia) was used to determine the amount of secreted insulin and C-peptide.
Differentiation of hASCsA 4-step, 16-day protocol was performed to induce differentiation of hASCs into ILCs. The schema of the experiment is illustrated in figure 1. Briefly, sixth-passage hASCs were expanded to 90% confluency after being cultured for 3 days in high-glucose DMEM (Gibco) supplemented with 55 nM TSA, 10 mg/l insulin and 1% FBS. Afterwards, the cells were randomly divided into two groups and cultured in DMEM with 250 ng/ml INGAP-PP (INGAP-PP group) or 250 ng/ml Scrambled-P (Scrambled-P group) for 5 days. Subsequently, the cells were cultured in low-glucose DMEM supplemented with 10 mM nicotinamide and 10 mM dexamethasone for 5 days. Finally, cells were cultured in low-glucose DMEM supplemented with 10 ng/ml transforming growth factor β1, 10 nM GLP-1 and 10 nM exendin-4 for another 3 days. All cultures were maintained in a 5% CO2 incubator at 37 ° C. Culture medium was changed every 3–4 days until ILCs were assessed.
Immunocytochemistry and Semiquantitative Immunoassay The hASCs and ILCs were fixed with 4% paraformaldehyde for 30 min followed by incubation with mouse anti-insulin (Sigma),
4
Cells Tissues Organs DOI: 10.1159/000362500
Cell Transplantation Diabetes was induced in 8-week-old male Sprague-Dawley rats by streptozotocin (Sigma; 55 mg/kg body weight in citrate buffer, pH 4.4, i.p.). Then, 48 h after injection, animals with blood glucose levels ≥16.65 mM for 2 consecutive days were transplanted with ILCs induced by INGAP-PP (1,000 IEQ/100 μl high-glucose DMEM) into the left subrenal capsule (n = 10). As controls, freshly isolated human islets (1,000 IEQ/100 μl high-glucose DMEM) or saline without cells were transplanted. Two transplantation groups were administered tacrolimus (50 ng/kg/day, p.o.) to avoid immune rejection. To evaluate the effects of INGAP-PP protein on remission of diabetes, we also injected purified INGAP-PP (250 μg/0.5 ml, i.p.) into diabetic rats twice a day for 20 days. Blood glucose levels of all rats were measured every 3 days. Glucose tolerance tests were performed after injection of glucose (2 g/kg body weight,
Ren /Chen /Qi /Li /Gong
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Sixth passage of hASCs
rabbit anti-C-peptide (AbD Serotec), rabbit anti-PDX-1 (Chemicon), rabbit anti-CK19 (Chemicon), rabbit anti-glucagon (Chemicon), rabbit anti-somatostatin (Chemicon) and mouse anti-nestin (Chemicon) antibodies. FITC-labeled rabbit anti-mouse IgG and PE-labeled goat anti-rabbit IgG (Jackson ImmunoResearch) were used as secondary antibodies. Nonimmune serum was used as negative control. Nonspecific staining was excluded using isotypematched irrelevant primary antibodies. These samples were analyzed using a fluorescence microscope (Eclipse TE300, Nikon) and a confocal laser-scanning microscope (FV500, Olympus). A semiquantitative immunofluorescence imaging analysis was performed to determine the amount of insulin-positive cells using the HPIAS-100 color image analyzing system (Qianping, Wuhan, China). Randomly selected (n = 25) but representative areas of each sample were analyzed for the labeling index, defined as the ratio of the positive signal area to the measured area.
Statistical Analysis The results are presented as the mean ± SEM, and statistical comparisons between the INGAP-PP group and the Scrambled-P group were performed using Student’s unpaired t test. p < 0.05 indicated a significant difference.
Results
Characterization of hASCs Twenty-four hours after initial plating of the washed SVF pellet, spindle-shaped primary hASCs grew slowly and reached 90% confluency within 14 days. After the second passage, cells proliferated much more quickly and were subcultured at 1:3 ratios every 3 days. At the end of the fourth passage, hASCs developed a uniform fibroblast-like morphology (fig. 2a). Fluorescence-activated cell sorting analysis of the fourth-passage cells showed that purified hASCs expressed high levels of the cell surface markers CD73 (99.1%), CD90 (97.9%) and CD105 (97.2%). These markers are highly expressed in MSCs [Oh et al., 2004]. hASCs had extremely low expression levels of CK19 and PDX-1, a distinctive feature of pancreatic stem cells. We confirmed that these cells are not hematopoietic stem cells because they are CD45 negative (fig. 2b). To further characterize the hASCs, we examined the expression of markers of undifferentiated ESCs including Oct3/4, Nanog and REX-1. Real-time PCR analysis and Western blot showed that undifferentiated hASCs expressed Oct3/4, Nanog and REX-1, both at the mRNA and protein level (fig. 2c, d). Taken together, these results indicate that hASCs not only have MSC characteristics but also some ESC characteristics and might be an alternative stem cell source, free of the ethical issues impeding the use of ESCs. Differentiation of hASCs with the INGAP-PP Protocol The sixth-passage hASCs were induced according to a 4-step protocol described in detail in Materials and Methods. The cells expanded rapidly in expansion medium and reached 100% confluency in about 3 days. At the end of stage 2, the spindle-shaped cells shrank to a cobblestone-like morphology (fig. 3a). Cell aggregations were observed at about 4 days after initial treatment (fig. 3b). Before step 2 of differentiation, the reconstitutDifferentiation of hASCs
ing cells were equated at 4 ×105 cells/ml in the INGAPPP group and the Scrambled-P group. When the last differentiation step was finished, the production of ILCs in the INGAP-PP group (fig. 3d) comprised 87 ± 12 ILCs/106 cells, significantly higher than the 45 ± 7 ILCs/106 cells produced in the Scrambled-P group (fig. 3c; p < 0.01). Moreover, most of the ILCs in the INGAP-PP group (fig. 3d) were more regular in morphology and more compact in structure than the ILCs in the Scrambled-P group (fig. 3c). Also, there was a significant difference in the size of the ILCs between the two groups (163 ± 26.2 μm for INGAP-PP, 77.4 ± 11.9 μm for Scrambled-P; p < 0.01). Furthermore, figures 3e and f show DTZ-stained ILCs from the INGAP-PP group and freshly isolated islets, which revealed that ILCs from the INGAP-PP group did express insulin, although less than native islets. Expression of Cell Markers in hASCs and ILCs during the Differentiation The immunocytochemistry results (fig. 4) and quantitative real-time PCR results (fig. 5) showed that ILCs from the INGAP-PP group and fresh islets did not express Nanog, Oct3/4 and REX-1 anymore, which are highly expressed by undifferentiated hASCs. At the early stage, the morphology of hASCs changes from spindle-like to epithelial-like cells (fig. 4a, b), and cells start to express the nestin gene (fig. 4a), which is the marker to isolate pancreatic precursor cells and duct epithelium cells, suggesting that there is differentiation from hASCs to ILCs. High-level expression of CK19 (fig. 4b) suggested a possible transition from mesenchymal cells to epithelial cells. Both the quantitative real-time PCR and immunocytochemistry showed pancreas hormone expressions (insulin, glucagon, somatostatin) of the ILCs from the INGAP-PP group, with fresh human islets as control. Quantitative real-time PCR results showed that by adding the INGAP-PP, the differentiation of hASCs to ILCs was significantly promoted in the INGAP-PP group compared to the Scrambled-P group. The expression levels (percentage islet expression) of the endocrine transcription factors (PDX-1, NK homeobox gene-2.2, FoxA2 and Pax-6) in ILCs from the INGAP-PP group were 76.2 ± 2.1, 87.6 ± 3.7, 65.5 ± 2.9 and 63.7 ± 3.2%, respectively, while they were 39.7 ± 2.1, 30 ± 4.7, 18.6 ± 0.5 and 15.9 ± 0.37%, respectively, in the Scrambled-P group. The pancreas hormone expression levels (insulin, glucagon, somatostatin) of the ILCs from the INGAP-PP group were significantly higher than those in the ScramCells Tissues Organs DOI: 10.1159/000362500
5
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
i.p.). Normal rats as well as diabetic rats transplanted with fresh islets were used as positive controls. Diabetic rats were used as negative control. Every group contained 10 rats. Statistical analyses were performed using the SPSS 11.5 software.
Relative mRNA level (%)
120
c
a
Undifferentiated ESCs Undifferentiated hASCs
100 80 60 Oct3/4 40
Undifferentiated ESCs Undifferentiated hASCs
20 0
REX-1 Nanog
Oct3/4
REX-1
Nanog
DŽ-Actin
d
99.1%
97.1%
b
CD105-PE
1.1%
CK19-PE
CD45-FITC
97.5% 0.4%
CD90-PE
PDX-1-PE
Fig. 2. Characterization of the hASCs. a Image of the sixth-passage hASCs. Undifferentiated hASCs turned into a uniform spindleshaped morphology. Scale bar = 50 μm. b Fluorescence-activated cell sorting analysis of different cell markers in hASCs. Sixth-passage hASCs were stained with PE- or FITC-labeled antibodies and analyzed by fluorescence-activated cell sorting. Undifferentiated hASCs were positive for CD73, CD90 and CD105 and negative for
CD45, PDX-1 and CK19. c Quantitative real-time PCR analysis presents the relative expression of undifferentiated hASCs compared to undifferentiated ESCs for Oct3/4, Nanog and REX-1. d Western blot analysis for Oct3/4, Nanog and REX-1. Undifferentiated hASCs and ESCs expressed Oct3/4, Nanog and REX-1 at the protein level.
bled-P group (15.7 ± 3.9, 4.1 ± 1.2 and 5.8 ± 1.9%, respectively, compared to 3.1 ± 0.4, 0.47 ± 0.11 and 1.1 ± 0.21%, respectively). Insulin expression was enhanced from 5–10% of the islets in previous research to about 10–20% in the present
study. Semiquantitative images for insulin labeling results showed 18.9 ± 3.2% (percentage islet expression) in the INGAP-PP group and 5.1 ± 1.1% (percentage islet expression) in the Scrambled-P group, which coincides with the results of quantitative real-time PCR.
6
Cells Tissues Organs DOI: 10.1159/000362500
Ren /Chen /Qi /Li /Gong
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
CD73-PE
2.4%
a
b
c
d
e
f
spindle-shaped hASCs changed into cobblestone-like shapes at the end of stage 2. b Cells started to aggregate after stage 1 and stage 2 inductions. c ILCs formed in the Scrambled-P group. d ILCs formed in the INGAP-PP group. e DTZ-stained ILCs formed in the INGAP-PP group. f DTZstained fresh islet. ×100. Scale bars = 50 μm.
ILCs Produced Using the INGAP-PP Protocol Released Higher Amounts of Insulin and C-Peptide in Response to Glucose To determine the glucose response of ILCs, insulin and C-peptide secretion in the presence of low (5.6 mM)
and high (25 mM) glucose concentrations were measured using ELISA. Islets and ILCs from both the INGAP-PP and Scrambled-P groups could all secrete insulin and C-peptide in response to glucose, but the levels of secretion varied. ILCs from the INGAP-PP group se-
Differentiation of hASCs
Cells Tissues Organs DOI: 10.1159/000362500
7
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Fig. 3. Images of hASCs or ILCs at different stages of the INGAP-PP protocol. a The
Nestin
50.0 μm
Insulin
50.0 μm
Insulin
e
c
50.0 μm
h
50.0 μm
C-peptide
50.0 μm
f
Merge
g
50.0 μm
Insulin/glucagon
50.0 μm
i
50.0 μm
Insulin
Insulin/somatostatin
8
20.0 μm
Merge
d
j
b
50.0 μm
Cells Tissues Organs DOI: 10.1159/000362500
k
50.0 μm
(For figure 4l and legend see next page.)
Ren /Chen /Qi /Li /Gong
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
a
4
PDX-1
CK19
***#
1.0
tive image for insulin labeling. Semiquantitative immunofluorescence analysis was performed to estimate the labeling index of insulin-producing cells derived from hASCs by different treatment. Twenty-five areas were chosen each time. Each value (n = 3 experiments) represents the mean ± SEM. ** p < 0.001, *** p < 0.01 versus Scrambled-P. # p < 0.05 versus hASCs.
0.7 0.6 0.5 0.4 0.3
90
** #
0.2 0.1
l
80
#
0
INGAP-PP group Scrambled-P group
hASCs
INGAP-PP
Scrambled-P
Islets
***
60
**
**
-6
**
70
Fo xA 2
** **
50 40 30
4
1 XRE
3/ ct O
og an N
Pa x
2 2.
ratios between rats transplanted with ILCs induced by INGAP-PP and normal islets. Because of the poor induction results in the Scrambled-P group, the transplantation step excluded the ILCs from the Scrambled-P group. Streptozotocin-induced diabetic rats often have a blood glucose level higher than 16.7 mM. Blood glucose levels of rats transplanted with INGAP-induced ILCs fell to 7.8 ± 0.8 mM and stayed at this level for the entire course of the experiment. Diabetic rats transplanted with normal islets, as a control, showed normal blood glucose levels after 1–2 weeks. Nontransplanted diabetic rats remained hyperglycemic and subsequently died. Interestingly, rats injected with INGAP-PP also showed Cells Tissues Organs DOI: 10.1159/000362500
9
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Remission of Diabetes by Transplantation of INGAP-Induced Cells To evaluate whether ILCs could reverse diabetes in rats, we compared the blood glucose levels and survival
CK 19 N es tin
** kx
**
1
0
**
N
10
PD X-
20
creted a significantly higher amount of insulin and Cpeptide when compared to ILCs from the Scrambled-P group. The levels of insulin and C-peptide release were about 10% of that of normal islets. Moreover, the ILCs from the INGAP-PP group secreted 6–7 times more insulin and C-peptide under high-glucose than low-glucose conditions, similar to the change in the islets group (fig. 6a, b).
Differentiation of hASCs
0.8
In su lin Gl uc ag So on m at os ta tin
vious immunocytochemical results, endocrine hormone (insulin, glucagon, somatostatin) expression levels in ILCs of the INGAP-PP group were compared to those in islets. ESC markers (REX-1, Oct3/4 and Nanog) were undetectable in both ILCs and islets. Human islet precursor cell markers (nestin and CK19) were detected, and while most endocrine transcription factors assessed [NK homeobox gene-2.2 (Nkx2.2), Pax-6, FoxA2] were found in islets and ILCs, the expression of one such factor that has also been proposed as a progenitor cell marker (PDX-1) was somewhat maintained in both ILCs and islets at a similar level. Data are presented as means ± SEMs of 3 independent experiments. ** p < 0.01, *** p < 0.001.
100 Expression/DŽ-actin (% islet expression)
Fig. 5. Real-time PCR. Confirming the pre-
0.9
Labeling index
Fig. 4. Immunocytochemical analysis of genes related to pancreatic development. a–j Gene expression levels in the INGAP-PP group. a Nestin (green). b CK19 (red). c PDX-1 (red). d Insulin (green). e Merge of c and d. f C-peptide (red). g Insulin (green). h Merge of f and g. i Double staining of insulin (green) and glucagon (red). j Double staining of insulin (green) and somatostatin (red). a, b Cells harvested on day 7. ×400. c–h Cells harvested on day 16. ×100. k Insulin expression of islet (green). l Semiquantita-
Insulin (mIU/l)
40
45.6
5,000
30 20
** 7.8
10 0
a –10
0.74
* 0.18
6,000
**
INGAP-PP group Scrambled-P group Islets
C-peptide (ng/l)
50
4.8
5.6 mM
*
1.6 25 mM
INGAP-PP group Scrambled-P group Islets
**
4,860.5
4,000 3,000 2,000 1,000 0
b –1,000
**
812.1
654.7
114.3 5.6 mM
*
25 mM
Fig. 6. Secretion of insulin (a) and C-peptide (b) in response to 5.6 or 25 mM glucose in INGAP-PP and Scram-
bled-P groups and normal human islets (n = 3). Each value represents the mean ± SEM. * p < 0.05, ** p < 0.01 (Student’s t test).
Discussion
Insulin-producing cell transplantation has been shown to be the most promising therapy to cure insulin-dependent diabetes. Though ESCs, pancreatic epithelium stem cells, BMSCs and hASCs are all capable of differentiating into insulin-secreting cells or even ILCs [Tang et al., 2004; Choi et al., 2005; Moriscot et al., 2005], the efficiency of differentiation was too low to produce adequate functional cells for clinical applications.
10
Cells Tissues Organs DOI: 10.1159/000362500
To overcome this obstacle, we developed a 4-step strategy to induce ILC formation based on a modified protocol used by previous studies [Vinik et al., 1997; Rosenberg et al., 2004; Timper et al., 2006]. Our strategy consisted of 4 stages, each involving different kinds of reagents to induce hASC differentiation. TSA, a histone deacetylase inhibitor, was added to the culture medium to provide better access to the chromatin. Histone acetylation decreases the ability of histones to bind DNA, causing a more ‘open’ chromatin state. Inhibition of histone deacetylase prevents histone deacetylation and therefore can alter gene expression patterns and help trans-acting factors to bind their target DNA sequences. INGAP-PP and transcription factors exendin-4 and GLP-1 all play important roles in pancreas development. Sequentially stimulating hASCs with these reagents helped efficient differentiation into functional ILCs. By examining the molecular marker of differentiated cells, we found that hASCs might differentiate into functional ILCs through a mesenchymal-to-epithelial transition. This transition has been reported to be present during the differentiation of human islet precursor cells into insulin-producing cells [Gershengorn et al., 2004]. Our results also demonstrated that the expression levels of early developmental-stage pancreatic genes, such as CK19, Nestin and PDX-1, were similar in the INGAP-PP and Scrambled-P groups. In contrast, genes related to pancreatic function, such as insulin and C-peptide, were expressed at significantly higher levels in the INGAP-PP Ren /Chen /Qi /Li /Gong
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
a decrease in blood glucose level after 5 days of injections. However, the glucose level increased again 2 weeks after INGAP-PP injections were discontinued (fig. 7a). Diabetic rats transplanted with INGAP-PP-induced ILCs or normal islets survived the whole 69-day process, while diabetic rats without treatment suffered from hyperglycemia and died within 51 days. All of the diabetic rats who received INGAP-PP injections also survived until the end of the experiment (fig. 7b) Intraperitoneal glucose tolerance tests showed that Sprague-Dawley rats transplanted with ILCs induced by INGAP-PP had similar kinetics of glucose clearance to the group transplanted with normal islets, although the blood glucose levels were higher at all time points. In contrast, untreated diabetic rats revealed no reaction to the high glucose concentration (fig. 7c).
40
Blood glucose (mM)
35 30
**
25
** **
**
** ** ** ** **
** **
**
20 15 10 5 0 0
a
3 6
9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 Days after transplantation
1.0 0.9 0.8
Diabetic rats Diabetic rats + INGAP-PP-induced cells Diabetic rats + INGAP-PP injection Diabetic rats + normal islets Normal rats
Survival ratio
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
b
Fig. 7. Remission of diabetes by transplantation of ILCs into diabetic rats. a Blood
10
20 30 40 50 Days after transplantation
40 35 Blood glucose (mM)
**
30
**
70
60
**
**
25
**
20 15 10 5 0
c
0
20
40 60 80 Time after glucose loading (min)
100
120
group than in the Scrambled-P group. These results strongly suggest an essential role of INGAP-PP in stimulating the maturation and insulin secretion ability of β-cells rather than initiating pancreatic lineage differentiation.
The result that ILCs from the INGAP-PP group are more similar to fresh islets is consistent with the previous conclusion that INGAP-PP promotes maturation of pancreas and insulin secretion [Borelli et al., 2005]. Although levels of insulin and C-peptide secreted by them
Differentiation of hASCs
Cells Tissues Organs DOI: 10.1159/000362500
11
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
glucose levels of transplanted rats. ILCs and normal islets were transplanted under the kidney capsule of streptozotocin-induced diabetic rats (blood glucose levels ≥16.7 mM), and blood glucose levels were measured every 3 days. The arrow shows that the kidney containing the graft was removed 60 days after transplantation. b Survival rates of transplanted rats. c Intraperitoneal glucose tolerance test. The tests were performed 4 weeks after transplantation. Each value represents the mean ± SEM. ** p < 0.01 (Student’s t test). n = 10/group.
0
were revealed to be only 10–20% of those of the normal islets by quantitative PCR and immunocytochemistry and about 10% by the glucose-stimulating test, it was a promising improvement compared with the approximately 2–5% secretion level reported previously [Lumelsky et al., 2001]. ILC transplantation into diabetic rats did indeed decrease the blood glucose levels and extend the life span of the diabetic rats. However, disease remission was not as complete as with fresh islet transplantation. This was probably due to the fact that the amount of ILCs was not sufficient to compensate the function of destroyed pancreas. Consistent with in vitro glucose stimulation test results, intraperitoneal glucose tolerance tests showed that ILCs upregulated insulin secretion and reversed hyperglycemia. In addition, we have to say that the effects we got in vivo experiments are much more satisfactory than those in vitro, the reason may lie in the continuous differentiation of ILCs. Interestingly, INGAP-PP injection alone led to a decrease in blood glucose level in diabetic rats 5–6 days after the injections. However, the blood glucose level increased
2 weeks after termination of the injections. These results may imply that although INGAP-PP could promote islet neogenesis, the regeneration islets were subsequently destroyed in diabetic rats. In summary, induced production of ILCs or insulinproducing cells represents an area of increasing interest in regenerative medicine. With our 4-stage INGAP-PP protocol, we enhanced the differentiation efficiency of hASCs into insulin/C-peptide-positive ILCs, the transplantation of which partially reversed diabetes in an animal model. Thus, our studies suggest hASCs as an alternative, autologous source for cell replacement therapy in diabetes.
Acknowledgements This research was supported by the 973 special plan of China (No. 2007CB516811) and the key Science and Technology plan of Shenzhen (No. 200901001). We thank colleagues in our laboratory for technical assistance and advice. We appreciate ChunYan Deng for providing technical support for flow cytometry analysis.
References
12
Cells Tissues Organs DOI: 10.1159/000362500
Fujikawa, T., S.H. Oh, L. Pi, H.M. Hatch, T. Shupe, B.E. Petersen (2005) Teratoma formation leads to failure of treatment for type I diabetes using embryonic stem cell-derived insulin-producing cells. Am J Pathol 166: 1781– 1791. Gershengorn, M.C., A.A. Hardikar, C. Wei, E. Geras-Raaka, B. Marcus-Samuels, B.M. Raaka (2004) Epithelial-to-mesenchymal transition generates proliferative human islet precursor cells. Science 306: 2261–2264. Lumelsky, N., O. Blondel, P. Laeng, I. Velasco, R. Ravin, R. McKay (2001) Differentiation of embryonic stem cells to insulin-secreting structures similar to pancreatic islets. Science 292: 1389–1394. Moriscot, C., F. Fraipont, M.J. Richard, M. Marchand, P. Savatier, D. Bosco, M. Favrot, P.Y. Benhamou (2005) Human bone marrow mesenchymal stem cells can express insulin and key transcription factors of the endocrine pancreas developmental pathway upon genetic and/or microenvironmental manipulation in vitro. Stem Cells 23: 594–603. Oh, S.H., T.M. Muzzonigro, S.H. Bae, J.M. LaPlante, H.M. Hatch, B.E. Petersen (2004) Adult bone marrow-derived cells trans-differentiating into insulin-producing cells for the treatment of type I diabetes. Lab Invest 84: 607–617.
Okura, H., H. Komoda, Y. Fumimoto, C.M. Lee, T. Nishida, Y. Sawa, A. Matsuyama (2009) Transdifferentiation of human adipose tissue-derived stromal cells into insulin-producing clusters. J Artif Organs 12: 123–130. Pittenger, G.L., D.A. Taylor-Fishwick, R.H. Johns, N. Burcus, S. Kosuri, A.I. Vinik (2007) Intramuscular injection of islet neogenesisassociated protein peptide stimulates pancreatic islet neogenesis in healthy dogs. Pancreas 34: 103–111. Planat-Benard, V.P., J.S. Silvestre, B. Cousin, M. Andre, M. Nibbelink, R. Tamarat, M. Clergue, C. Manneville, C. Saillan-Barreau, M. Duriez, A. Tedgui, B. Levy, L. Penicaud, L. Casteilla (2004) Plasticity of human adipose lineage cells toward endothelial cells: physiological and therapeutic perspectives. Circulation 109: 656–663. Rafaeloff, R, G.L. Pittenger, S.W. Barlow, X.F. Qin, B. Yan, L. Rosenberg (1997) Cloning and sequencing of the pancreatic islet neogenesis associated protein (INGAP) gene and its expression in islet neogenesis in hamsters. J Clin Invest 99: 2100–2109. Ramiya, V.K., M. Maraist, K.E. Arfors, D.A. Schatz, A.B. Peck, J.G. Cornelius (2000) Reversal of insulin-dependent diabetes using islets generated in vitro from pancreatic stem cells. Nat Med 6: 278–282.
Ren /Chen /Qi /Li /Gong
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Atkinson, M.A., G.S. Eisenbarth (2001) Type 1 diabetes: new perspectives on disease pathogenesis and treatment. Lancet 358: 221–229. Barbosa, H., S. Bordin, L. Stoppiglia, K. Silva, M. Borelli, H. Del Zotto, et al. (2006) Islet Neogenesis Associated Protein (INGAP) modulates gene expression in cultured neonatal rat islets. Regul Pept 136: 78–84. Borelli, M.I., L.F. Stoppiglia, L.F. Rezende, L.E. Flores, H. Del Zotto, A.C. Boschero, et al. (2005) INGAP-related pentadecapeptide: its modulatory effect upon insulin secretion. Regul Pept 131: 97–102. Butler, A.E., J. Janson, S. Bonner-Weir, R. Ritzel, R.A. Rizza, P.C. Butler (2003) Beta-cell deficit and increased beta-cell apoptosis in humans with type 2 diabetes. Diabetes 52: 102–110. Choi, K.S., J.S. Shin, J.J. Lee, Y.S. Kim, S.B. Kim, C.W. Kim (2005) In vitro trans-differentiation of rat mesenchymal cells into insulinproducing cells by rat pancreatic extract. Biochem Biophys Res Commun 330: 1299–1305. Dominici, M., K. LeBlanc, I. Mueller, I. SlaperCortenbach, F. Marini, D. Krause, R. Deans, A. Keating, D. Prockop, E. Horwitz (2006) Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy 8: 315–317.
Differentiation of hASCs
Song, C. K., J.H. Duck, J.Y. Lee (2010) Adipose tissue derived stem cells for regeneration and differentiation into insulin-producing cells. Curr Stem Cell Res Ther 5: 190–194. Tang, D.Q., L.Z. Cao, B.R. Burkhardt, C.Q. Xia, S.A. Litherland, M.A. Atkinson, L.J. Yang (2004) In vivo and in vitro characterization of insulin-producing cells obtained from murine bone marrow. Diabetes 53: 1721–1732. Taylor-Fishwick, D.A., S. Rittman, H. Kendall, L. Roy, W. Shi, Y. Cao (2003) Cloning genomic INGAP: a Reg-related family member with distinct transcriptional regulation sites. Biochim Biophys Acta 1638: 83–89.
Cells Tissues Organs DOI: 10.1159/000362500
Timper, K., D. Seboek, M. Eberhardt, P. Linscheid, M. Christ-Crain, U. Keller, B. Müller, H. Zulewski (2006) Human adipose tissuederived mesenchymal stem cells differentiate into insulin, somatostatin, and glucagons expressing cells. Biochem Biophys Res Commun 341: 1135–1140. Vinik, A, R. Rafaeloff, G. Pittenger, L. Rosenberg, W. Duguid (1997) Inducement of pancreatic islet neogenesis. Horm Metab Res 29: 278– 293. Zuk, P.A., N Zhu, H. Mizuno, J. Huang, J.W. Futrell, A.J. Katz, P. Benhaim, H.P. Lorenz, M.H. Hedrick (2001) Multilineage cells from human adipose tissue: implications for cellbased therapies. Tissue Eng 7: 211–228. Zuk, P.A., M. Zhu, P. Ashjian, J.I. De Ugarte, D.A. Huang, H. Mizuno, Z.C. Alfonso, J.K. Fraser, P. Benhaim, M.H. Hedrick (2002) Human adipose tissue is a source of multipotent stem cells. Mol Biol Cell 13: 4279–4295.
13
Downloaded by: UCSF Library & CKM 169.230.243.252 - 12/15/2014 7:45:31 AM
Rosenberg, L., M. Lipsett, J.W. Yoon, M. Prentki, R. Wang, H.S. Jun, et al. (2004) A pentadecapeptide fragment of islet neogenesis-associated protein increases beta-cell mass and reverses diabetes in C57BL/6J mice. Ann Surg 240: 875–884. Seeberger, K.L., J.M. Dufour, A.M. Shapiro, J.R. Lakey, R.V. Rajotte, G.S. Korbutt (2006) Expansion of mesenchymal stem cells from human pancreatic ductal epithelium. Lab Invest 86: 141–153. Seo, M.J., S.Y. Suh, Y.C. Bae, J.S. Jung (2005) Differentiation of human adipose stromal cells into hepatic lineage in vitro and in vivo. Biochem Biophys Res Commun 328: 258–264. Shapiro, A.M., J.R. Lakey, E.A. Ryan, G.S. Korbutt, E. Toth, G.L. Warnock, N.M. Kneteman, R.V. Rajotte (2000) Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Eng J Med 343: 230–238.
