| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
Diabetes Transplant Unit, Prince of Wales Hospital and University of New South Wales, Sydney, New South Wales 2031, Australia
Address all correspondence and requests for reprints to: Bernard E. Tuch, M.D., Ph.D., Diabetes Transplant Unit, Prince of Wales Hospital, High Street, Randwick, New South Wales 2031, Australia. E-mail: b.tuch{at}unsw.edu.au.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The fetal ß-cell is immature, being unable to secrete insulin when exposed to glucose (10, 11). Maturation of these ß-cells occurs mostly after birth (12, 13, 14) at a time when oral ingestion of food commences. We (15) and others (16) have hypothesized that incretins released at the time of oral feeding might be responsible for ß-cell maturation. In support of this we have shown that human fetal pancreatic explants, which contain ß-cells, exposed to GLP-1 or CCK for 4 d become glucose responsive (15). Otonkoski and Hayek (16) have shown that acute exposure of human fetal islet-like cell clusters (ICCs) to a combination of GLP-1 and glucose results in synergism in insulin secretion; either agent alone has minimal effect. In mice, GLP-1 agonist has been shown to enhance ß-cell neogenesis and stimulate the expression of PDX-1 (17), a transcription factor critical for pancreas development. Whether porcine fetal pancreatic tissue behaves in a similar manner remains to be tested.
Transplantation of pancreatic cells from fetal pigs has the potential to normalize blood glucose levels in diabetic immunoincompetent (18) and immunosuppressed (19) recipients, with beneficial effects observed after several months. During this time ß-cells proliferate, differentiate from ductal precursors, and mature in their ability to secrete insulin when exposed to glucose (18). It would be of interest to know whether exposing fetal pig pancreatic cells to incretins in vitro before transplantation favored their differentiation and maturation, thereby shortening the time required for reversal of diabetes. Here, we exposed fetal porcine ICCs to GLP-1, CCK, or NIC in vitro for 6 h to 4 d and assessed the effects of these agents on replication, differentiation, and maturation of ß-cells both before and after transplantation into immunodeficient mice. Nicotinamide (NIC) was used as a positive control; its beneficial effect on the differentiation of fetal porcine ß-cells has been previously demonstrated (20).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Source of mice
Male, inbred, severe combined immunodeficient (SCID) mice were obtained from the Animal Resources Center (Perth, Australia). They were housed in sterilized polycarbonate cages covered with autoclaved filter tops and had unlimited supply of sterilized drinking water and 
-irradiated mouse food. The temperature of the room in which the colony was housed was kept at 20 ± 3 C; the relative humidity was 53–70%. All procedures carried out on these mice were approved by the animal care and ethics committee of the institution and were in accordance with accepted standards of humane animal care as outlined in the Endocrine Society Code of Ethics.
Preparation and culture of fetal porcine ICCs
Fetal porcine ICCs were isolated following a procedure described previously (21). Briefly, pancreata were minced into 1- to 2-mm3 fragments and digested with 20 ml 3 mg/ml collagenase P (Roche Molecular Biochemicals, Mannheim, Germany) for 15–20 min at 37 C. Digests were washed twice with Dulbecco’s PBS (Trace Biosciences, Sydney, Australia) and then plated in RPMI 1640 on to 90-mm nonadherent petri dishes (Johns, Auckland, New Zealand).
For experiments requiring exposure to the growth factors, digests of two pancreases were initially mixed and then divided among four dishes. Culture medium in three of these dishes was supplemented either with 100 nM GLP-1-(7–36) (Sigma, St. Louis, MO), 5 µM sulfated cholecystokinin-8 (Sigma), or 10 mM NIC (Sigma), and the fourth dish served as a control. Petri dishes were maintained at 37 C in humidified air and 5% CO2 for 3–4 d, after which the ICCs were hand-picked for further experiments. The concentrations of GLP-1 and CCK used were based on a dose-response curve for their effects on human fetal pancreatic tissue (15). The concentration of NIC selected was based on its reported beneficial effect on fetal pig pancreatic tissue (20). To examine how GLP-1 was exerting its effect, ICCs in a parallel set of experiments were cultured in the presence of exendin-(9–39) (1 µM), an agent that inhibits GLP-1 binding to its receptor.
In experiments in which proliferative activity of ICCs was determined, 0.1 mM 5-bromo-2'-deoxyuridine (BrdU; Berhring Diagnostics, La Jolla, CA) was added to the culture medium 16 h before the ICCs were hand-picked.
Insulin secretion
ICCs cultured for 3 d after isolation were counted, and 200 of these were hand-picked into each sterile Eppendorf tube (Trace Plastics). These were then incubated in RPMI 1640 medium without (control) or with the addition of either incretins (100 nM GLP-1 and 5 µM CCK) or NIC (10 mM) for a period of 6 h (in some experiments ICCs were exposed to these agents for 3 d). After incubation of ICCs for 6 h, their ability to secrete insulin in response to glucose was examined by in vitro static stimulation (23). Briefly, ICCs were washed with PBS (1 mM CaCl2, 0.5 mM MgCl2, 26.7 mM NaHCO3, and 20 mM HEPES) containing 0.2% BSA (Sigma) and exposed in quadruplets to this basal (PBS and 2.8 mM glucose) or stimulated (PBS and 20 mM glucose) buffer for 1 h at 37 C. At the end of the hour, cells were pelleted down by centrifugation, and the supernatant was assayed for the insulin released.
Insulin content
After static stimulation, as described above, the insulin content of the ICCs was estimated after sonicating the ICC pellet in 200 µl acid/ethanol (18 ml 10 M HCl/liter 70% ethanol). Insulin was extracted by incubating the cells overnight in acid ethanol at 4 C. Insulin concentrations were measured by RIA of the lysates.
Semiquantitative fluorescent immunocytochemistry
For measurement of insulin and PDX-1 immunofluorescence, ICCs plated onto coverslips were probed with rabbit antiserum against PDX-1 (gift from Prof. Marc Montiminy, The Salk Institute, La Jolla, CA) and a guinea pig antiserum to insulin (DAKO Corp., Carpinteria, CA). Fluorescent secondary antisera coupled to tetramethyl rhodamine isothiocyanate or fluorescein isothiocyanate were obtained from Amrad Biotec (Melbourne, Australia). To minimize variability between different sections, the staining procedures for all culture conditions (control and experimental) were performed simultaneously and in parallel with the same batches of solutions and antisera. In addition, identical times for fixation, permeabilization, blocking, and incubation with antisera were employed for all ICCs. All og these ICC-containing coverslips were mounted in mounting medium (DAKO Corp.), with DAPI (Sigma) used as a nuclear stain. After fluorescent staining, ICCs were scanned using a TCS MP confocal microscope (Leica Corp., Heidelberg, Germany) equipped with a two-photon as well as 488-nm argon laser and 543-/633-nm helium/neon laser. Images were captured in confocal mode, and semiquantitative assessment of fluorescent intensity was performed for images of all sections collected in a single series of experiments to avoid alterations in signal intensities over time. Brightness and contrast parameters were kept constant for all photomultiplier tubes. For fluorescent intensity measurements, images were directly analyzed using KS400 image quantitation software (Kontron Instruments Ltd., Munich, Germany). In each image the islet used for fluorescent quantitation was defined as the region of interest (ROI) using the freehand tool. The average fluorescent intensity within this ROI was then determined by analysis of pixel intensity and is independent of the area of the measured ROI. The average fluorescent intensity of 100 islet equivalents per pancreas and three pancreatic preparations was determined. Fluorescent values, obtained from ICCs processed in the absence of primary antisera, were considered to reflect the nonspecific fluorescent background values and were subtracted from the ROI values. All values obtained on a scale of 0–256 are represented as a percentage of the average fluorescent intensity per 100 islet equivalents counted. Values are then expressed as the mean ± SEM.
To assess any differences in the number of islet cells that were PDX-1 positive, we also counted the number of PDX-1+ cells in these ICCs. At least 450 cells from 3 different preparations were counted for ICCs exposed to either of the incretins, NIC, or the untreated controls.
To quantify any differences in size of ICCs after exposure to incretins, we measured the diameters of ICCs that were incubated in RPMI 1640 medium without (control) or with the addition of either incretins (100 nM GLP-1 and 5 µM CCK) or NIC (10 mM) for a 3-d period. ICCs from each group were stained with dithizone (diphenylthiocarbazone, Sigma) at a final concentration of 33 mg/ml and taken for quantification of islet diameter on a Axiovert imaging system (Carl Zeiss, Jena, Germany). Images were captured using the Axiocam Imaging CCD camera (Carl Zeiss), with identical parameters maintained (using the same macro) for the control and experimental samples. Images were then processed to binary images, and morphometric estimates were made using KS 400 software (Kontron Instruments Ltd., Zurich, Switzerland).
Transplantation of ICCs
ICCs cultured in different media for 4 d were picked in multiples of 500 and transplanted in a plasma clot beneath the left renal capsule of nondiabetic SCID mice (21). Two months after transplantation, the function of some of these grafts was assessed in terminal experiments by perfusion of the graft-bearing kidney. The remaining grafts were removed at 1 or 2 months for morphological analysis, 4 h after the mice had received an ip injection of 100 mg/kg BrdU.
Perfusion of graft-bearing kidneys
This procedure was carried out by a modification of the method previously described (24, 25). Transplanted mice were anesthetized with an ip injection of pentobarbitone (75 mg/kg), and laparotomy was performed. The abdominal aorta and inferior vena cava were exposed, and all branches from these vessels, except the left renal artery and vein, were ligated. Ligatures were then tied separately around the aorta and the vena cava above the origin of the left renal vessels. The abdominal aorta was cannulated with a catheter of 0.47-mm internal diameter (Terumo Corp., Tokyo, Japan) connected to a peristaltic pump (Gilson, Villiers de Bel, France). The renal vein was then cannulated with a similar size catheter to collect effluent samples, and the left ureter was cut to avoid stasis of urine within the kidney. The medium pumped through the kidney was Krebs-Ringer bicarbonate buffer supplemented with 10 mM HEPES, 2% (wt/vol) dextran T70 (Pharmacia Biotech, Uppsala, Sweden), 2.8 mM glucose, and 2% BSA. This was kept at 37 C and continuously gassed with carbogen (5% CO2/95% O2); the flow rate was 0.7 ml/min. Once perfusion of the kidney was established, the mouse was killed, and the perfused kidney was covered with cotton gauze moistened in saline, and kept at 37 C with a warming lamp. Stimuli applied to this graft-bearing kidney were 20 mM glucose, followed by 20 mM glucose supplemented with 10 mM theophyline; the latter was used to test viability of ß-cells, as fetal ß-cells are responsive to theophyline (11, 18).
The perfusion started with 15 min of perfusate containing 2.8 mM glucose, followed by 30 min with 20 mM glucose, 10 min with 2.8 mM glucose, 20 min with 20 mM glucose supplemented with 10 mM theophyline, and finally 10 min with 2.8 mM glucose. A sample of the effluent medium was collected every fifth minute, except for the first 10 min of perfusion with high glucose concentration when samples were taken every minute for the first 5 min and at 7 and 10 min. The insulin content of all samples was measured by RIA. The amount of insulin released during each phase of the perfusion was measured by calculating the area under the insulin curve. A stimulation index was calculated during each phase of stimulation by dividing the area under the curve by that obtained during the preceding basal period, corrected for an equivalent period of time. Stimulation indexes were calculated for the first 10 min (first phase) and the last 20 min (second phase) of exposure to 20 mM glucose and the 20 min of exposure to 20 mM glucose and 10 mM theophyline.
RT-PCR analysis
ICCs that were cultured in presence or absence (controls) of GLP-1 were taken for RT-PCR analysis after 3 d of culture. RNA was isolated following standard protocol using TRIzol reagent (Invitrogen, Carlsbad, CA), and RNA obtained was quantified by spectrophotometry. To confirm that the signal obtained was not saturated, 0.5, 1.0, or 2.0 µg RNA was subjected to RT reaction followed by PCR. Previously described primers (26) were used as follows: PDX-1 (product, 262 bp), 5'-CCCATGGATGAAGTCTACC-3' and 5'-GTCCTCCTCCTTTTTCCAC-3'; and glyceraldehyde-3-phosphate dehydrogenase (GAPDH; product, 350 bp) 5'-AATCCCATCACCATCTTCCA-3' and 5'-GGCAGTGATGGCATGGACTG-3'. PCR was carried out in a thermal cycler (Biometra, Gottingen, Germany) for 30 cycles. PCR conditions were the same for PDX-1 and G3PDH: denaturing for 30 sec at 92 C, annealing for 30 sec at 52 C, and extension for 40 sec at 72 C. The resulting products were loaded on 1.5% agarose gel and analyzed over a gel documentation system (Fluo-S, Bio-Rad Laboratories, Inc., Hercules, CA). Band intensities were measured for each of the control as well as the GLP-1-treated clusters. The intensities of the control PDX-1 were normalized to that of the GAPDH intensities, and the fold difference was calculated.
Morphological analysis
ICCs that were exposed to incretins or exendin-(9–39) or controls as well as the 2-month harvested grafts were double stained for insulin and the thymidine analog BrdU (27) using the immunoalkaline phosphatase and immunoperoxidase techniques, respectively. The 1-month harvested grafts and the exendin-(9–39)/GLP-1-exposed ICCs were stained solely for insulin. ICCs were fixed in 10% buffered formalin for 1 h and centrifuged. The pellet was placed in 2% (wt/vol) agarose (FMC Bioproducts, Rockland, ME) at 37 C and snap-cooled to 4 C, and the part of agarose block containing the ICCs was embedded in paraffin for routine histological analysis. Graft-bearing kidneys were fixed overnight in 4% paraformaldehyde (Probing and Structure, Thuringowa, Australia) before being embedded in paraffin. Four-micron sections were cut and mounted on poly-L-lysine (Sigma)-coated glass slides. Incubating the tissue with goat serum and 3% hydrogen peroxide, respectively, before addition of the primary antibodies blocked nonspecific binding and endogenous peroxidase activity.
Insulin-producing cells were detected using polyclonal guinea pig antiinsulin antibody as the primary antibody, followed by sequential incubations with rabbit antiguinea pig antibody, biotinylated goat antirabbit antibody, and alkaline phosphatase-labeled streptavidin. The chromogen new fuchsin was then added to stain the cytoplasm of insulin-positive cells red, and the sections were counterstained with hematoxylin. BrdU was detected using mouse monoclonal anti-BrdU as the primary antibody, followed by biotinylated goat antimouse antibody and peroxidase-labeled streptavidin. All reagents were obtained from DAKO Corp. Before staining, DNA was denatured by incubating sections with 95% formamide in 0.15 M trisodium citrate for 45 min at 70 C. BrdU-positive cells were visualized with the chromogen 3,3-diaminobenzidine tetrahydrochloride (Sigma), staining the nuclei brown; the counterstain was hematoxylin. Negative controls were conducted by omitting the primary antibodies. To estimate the rate of ß-cell proliferation in tissues, both of the above techniques were applied to the same tissue section.
For each group 2–6 samples (3–5 sections/sample, 40 µm apart) were analyzed, with 500-1000 cells being counted in each section. The following indexes were calculated: insulin labeling index (%) = (no. of insulin-positive cells x 100)/(total no. of epithelial cells); BrdU labeling index (%) = (no. of BrdU-positive cells x 100)/(total no. of epithelial cells); and double-labeling index (%) = (no. of cells positive for both insulin and BrdU x 100)/(no. of cells positive for BrdU).
Statistical analysis
Data were analyzed with the computer statistical package NCSS (28). Statistical significance of differences between groups was tested by paired t test or, if there were more than two groups, by one-way ANOVA after log transformation of the data. Where the P value for the latter test was less than 0.05, the differences between individual groups were evaluated with Duncan’s multiple range test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). The magnitude of stimulation was 3- and 1.7-fold, respectively. The NIC-treated ICCs also showed a significant increase in glucose-induced insulin secretion. No such enhancement was observed in ICCs exposed to culture medium alone.
|
).
Insulin content.
The insulin content of ICCs cultured in the presence of GLP-1, CCK, and NIC for 3 d, but not 6 h, was enhanced 1.6-, 2.5-, and 2.2-fold, respectively (P < 0.01; Table 1). It is probable that the enhancement of insulin content was due to an increase in the number of ß-cells in the ICCs (see Morphological studies).
|
) in ICCs cultured for 4 d, during which they were exposed to GLP, CCK, or NIC, increased 2.3-, 2.6-, and 2.0-fold, respectively (P < 0.001; Table 2). Most cells in an ICC did not contain insulin; the percentage of ß-cells varied from 8% in the controls to 20% in the ICCs exposed to CCK (Table 2). The effects of GLP-1, CCK, and NIC on enhancing the formation of ß-cells were time dependent. Thus, no effect was observed at 16 h (Fig. 2), but an effect was found at 3–4 d.
|
|
). NIC caused a significant 1.4-fold increase in the BrdU index at 4 d (P < 0.05), but had no effect on the double-labeling index. That GLP-1 was exerting its effect through its own receptor was determined by analyzing the percentage of ß-cells in ICCs exposed to GLP-1 with or without exendin-(9–39) for 3 d. The percentage of cells positive for insulin declined from 13.8 ± 1.9% in the presence of GLP-1 to 6.4 ± 0.5% when exendin 9–39 was added with the GLP-1 (exendin alone, 6.8 ± 0.6%; no agent, 7.1 ± 0.5%).
Morphometric studies.
ICCs that were treated with incretins or NIC for 3 d were quantified for their longest diameter. Control (untreated) ICCs had a diameter of 119 ± 17 µm, which was not different from ICCs that were exposed to GLP-1 (96 ± 6 µm), CCK (107 ± 10 µm), or NIC (126 ± 17 µm). Therefore, exposure of ICCs to GLP-1, CCK, or NIC for 3 d did not result in any increase in size of the ICCs.
Immunofluorescent studies.
ICCs exposed to GLP-1, CCK, and NIC, but not controls, showed an increase in the intensity of insulin fluorescence after 3 d of exposure. The increase in average fluorescent intensity was between 3- and 4-fold (Fig. 3) for insulin and was in accord with our immunohistochemical analysis, where we found a similar increase in the number of ß-cells (Fig. 2) and an increase in insulin content (Table 1). Similar semiquantitative studies that we carried out with reference to GLP-1 revealed a 1.9-fold increase in PDX-1 immunofluorescence. Actual measurements on the number of PDX-1-positive cells in these ICCs revealed that 52% (258 of 500 cells) in the GLP-1-treated ICCs (from 3 different preparations) were measured as PDX-1 positive compared with 27% (128 of 472 cells) that were PDX-1 positive in the control (untreated) islet cell population. We did not see any significant differences in the PDX-1 signal for ICCs that were treated with CCK or NIC for 3 d. In all groups PDX-1 was located mostly in the nucleus, but small amounts also were observed in the cytoplasm.
|
). Band intensity analysis after normalizing the intensities for G6PDH showed 1.7-fold up-regulation of PDX-1 in clusters exposed to GLP-1.
Transplanted ICCs
Insulin release from graft.
After 2 months in vivo fetal porcine ICCs that had been exposed to GLP-1, CCK, or NIC in vitro for 4 d before transplantation were perfused with 20 mM glucose. GLP-1-treated ICCs responded to glucose with a 4-fold increase in dynamic insulin secretion (P < 0.05; Fig. 4). This enhancement of insulin secretion occurred throughout the 30 min of exposure to glucose. In the first 10 min (the first phase) the increase was 3.3-fold, and in the last 20 min (second phase) the increase was 4.4-fold (P < 0.05; Fig. 4). CCK-treated ICCs also had enhancement of insulin secretion during the first phase with a 2.4-fold increase (P < 0.05), but not thereafter (Fig. 4A). NIC-treated ICCs had a pattern similar to that of the GLP-1-treated group, namely a significant increase throughout the entire period of stimulation, the increase being 3.9-fold in the first phase and 3-fold in the second phase (P < 0.05; Fig. 4). The grafts that had not been exposed to GLP-1, CCK, and NIC before transplantation were unresponsive to glucose alone. These (control) grafts did, however, respond positively to the combination of 10 mM theophyline and 20 mM glucose, enhancing dynamic insulin secretion 3.7-fold indicating that ß-cells were present (Fig. 4).
|
). The percentage of ß-cells in the control grafts had increased to 42 ± 6%. This was not significantly different from the grafts treated with CCK, it which the percentage of ß-cells was 45 ± 4%. The percentage of ß-cells in GLP-1 and NIC-treated grafts had increased to 79 ± 5% and 62 ± 2%, respectively, both significantly greater than the control value (P < 0.05).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Normally, fetal ß-cells secrete little or no insulin when exposed acutely to glucose (10, 11, 16). The reason for this is unclear, although an inadequate increase in the ATP to ADP ratio as glucose is metabolized seems the most likely reason (32). In an adult ß-cell glucose is transported into the cell, where it is metabolized initially in the anaerobic glycolytic pathway and subsequently in the aerobic citric acid cycle. ATP is produced during this metabolism, with an increase in the ATP to ADP ratio, thereby closing the ATP-dependent K+ channels, which depolarizes the membrane and results in opening of the voltage-activated Ca2+ channels. Extracellular Ca2+ enters the cell, and this causes the secretory granules to move toward the cell membrane and discharge their content of insulin. Protein kinases A and C prime these granules to the effect of the increase in intracellular Ca2+. GLP-1 enhances the activity of protein kinase A by stimulating production of cAMP (8), whereas CCK increases protein kinase C activity by stimulating levels of inositol phosphate (9). It is probable, therefore, that these two incretins, after binding to their specific receptors, allow a previously unresponsive ß-cell to respond to glucose by reducing the stimulus required from the normal signal transduction pathway. Indeed, our studies with the GLP-1 antagonist exendin-(9–39) indicate that GLP-1 was exerting its effect through its own receptor. It should be noted that the experiment we conducted showing a beneficial effect of GLP-1 on biochemical maturation of the fetal pig ß-cell was different from that described by Otonkoski on fetal human ß-cells (16), although the effector mechanism is likely to be the same. In our experiments ICCs were exposed to GLP-1 for 6 h before stimulation with glucose, but not at the same time; Otonkoski added GLP-1 to glucose during the stimulation, but did not expose the ICCs to incretin beforehand. The beneficial effect of NIC on glucose-induced insulin secretion from fetal pig ICCs has been noted previously (20, 33).
Fetal ß-cells normally develop mostly from progenitor duct cells, with few forming by proliferation of existing ß-cells (34). We observed an increase in the number of insulin-expressing cells after incretin exposure. However, there were no significant differences in the proliferation index of insulin-positive cells in either group. This indicates that the increase in the number of insulin-positive cells was not due to the proliferation of insulin-expressing cells, but was probably caused by differentiation of progenitor cells. NIC has previously been shown to enhance this process of differentiation in fetal pig (20, 33, 35) and fetal human ICCs (36). Our present data confirm the effect of NIC and show that the incretins GLP-1 and CCK are capable of achieving the same result. That CCK was able to exert this effect was not surprising, as a homolog, gastrin, has previously been reported to enhance ß-cell formation in transgenic mice primed to enhance the formation of pancreatic ducts (37). Differentiation is probably stimulated by enhanced expression of transcription factors, such as PDX-1, NeuroD, or NKX6.1, which are involved in the formation of ß-cells in the developing pancreas (38). Indeed, we observed GLP-1 to cause an increase in the PDX-1 signal in ICCs exposed to the incretin for 3 d. Others have shown that GLP-1 induces PDX-1+ cells in a cell line to differentiate into ß-cells (39) and increases PDX-1 transcription (40, 41) and translation in ß-cells (17, 31, 40, 41, 42).
The beneficial effects of GLP-1 and CCK on ß-cell formation were observed only several days after commencement of exposure to these agents. The same was true for NIC, suggesting that it takes time for the effect to be observed regardless of the metabolic pathway stimulated. This is understandable, as the development of a cell from its precursor does not occur instantaneously. That the incretins caused glucose to enhance insulin secretion in a much shorter time is best explained by the fact that the secretion of insulin is a much more rapid event. Indeed, glucose causes insulin secretion from adult ß-cells within a minute of exposure to it. Why the incretins failed to have a significant effect on glucose-induced insulin secretion after 3-d exposure to them is best explained by their degradation in culture medium. We measured the concentration of GLP-1 on d 0 and 3 of tissue culture using an ELISA that measures intact GLP-1 (Linco Research, Inc., St. Charles, MO). The level of GLP-1 was less than 1% of that present at the start of culture, presumably because of cleavage to the inactive form by the ubiquitous enzyme dipeptidyl peptidase IV (43). That ß-cell differentiation occurred at 3–4 d despite degradation of the incretins is presumably because these agents are required only to initiate the process of differentiation and need not be present to maintain it.
With the evidence that maturation and differentiation of porcine ß-cells can be induced by CCK and GLP-1 in vitro, it became relevant to address the issue of whether pretreating ICCs with these gut hormones before transplantation would accelerate the development of fetal ß-cells after grafting. We have demonstrated that these pretreated fetal pig ICCs transplanted under the kidney capsule of SCID mice matured functionally after 2 months, being able to secrete insulin acutely in response to a glucose challenge. There was a difference in effect between GLP-1 and CCK, the former causing an adult-type response with stimulation throughout the entire period of exposure to glucose and the latter a response only during the first 10 min of exposure. NIC had an effect similar to that of GLP-1. Others have reported this stimulation of insulin secretion by NIC previously in fetal pig ICCs 13–18 wk after they were transplanted (20, 33). It supplements our data, which were obtained 8 wk after transplantation. Control grafts did not respond to glucose alone, as has been reported by Korsgren et al. (18) in 75% of the mice studied at a similar time after transplantation. That all grafts secreted insulin when exposed to the combined stimulus of theophyline and glucose is consistent with the stimulatory effect on fetal ß-cells of agents that enhance levels of cAMP (10, 11, 18, 32).
The experiments we conducted demonstrate that GLP-1 and NIC continue to increase the rate of differentiation into ß-cells after the graft has been transplanted. The effect of GLP-1 continues for at least 1 month posttransplantation, after which time the insulin labeling index for GLP-1 and the controls was the same. NIC continued to have an effect on the differentiation of ß-cells, with the percentage of insulin-positive cells being significantly greater than the controls at 2 months (P < 0.05). CCK had no beneficial effect on continuing differentiation after transplantation, with the insulin labeling index being the same as the control value after both 1 and 2 months.
In summary, we have shown that GLP-1 and CCK can induce differentiation and biochemical maturation of fetal porcine ß-cells during culture of ICCs, the former of which seems to be acting by increasing the number and/or levels of PDX-1+ cells. Two months after these cells are transplanted, they secrete insulin when exposed to glucose, whereas untreated cells do not. These beneficial effects of the incretins suggest that exposure of fetal pig ICCs to the incretins in vitro would be advantageous in enhancing their effect to normalize blood glucose levels in diabetic recipients.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Present address: National Institute of Diabetes and Digestive and Kidney Disease, National Institutes of Health, Bethesda, Maryland 20892. E-mail: anand_hardikar{at}nih.gov. 
2 Recipient of an Overseas Postgraduate Award from the Department of Employment, Education, Training, and Youth Affairs, Commonwealth Government of Australia.
Abbreviations: BrdU, 5-Bromo-2'-deoxyuridine; CCK, cholecystokinin; G3PDH, glyceraldehyde-3-phosphate dehydrogenase; GLP-1, glucagon-like peptide 1; ICC, islet-like cell cluster; NIC, nicotinamide; ROI, region of interest; SCID, severe combined immunodeficient.
Received November 30, 2001.
Accepted for publication May 20, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-induced ductular precursor cells. J Clin Invest 92:1349–1356
This article has been cited by other articles:
 |
|
 |
|
  K. M. Picha, M. R. Cunningham, D. J. Drucker, A. Mathur, T. Ort, M. Scully, A. Soderman, T. Spinka-Doms, V. Stojanovic-Susulic, B. A. Thomas, et al. Protein Engineering Strategies for Sustained Glucagon-Like Peptide-1 Receptor-Dependent Control of Glucose Homeostasis Diabetes, July 1, 2008; 57(7): 1926 - 1934. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Liu and J. F. Habener Glucagon-like Peptide-1 Activation of TCF7L2-dependent Wnt Signaling Enhances Pancreatic Beta Cell Proliferation J. Biol. Chem., March 28, 2008; 283(13): 8723 - 8735. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L Bai, G Meredith, and B E Tuch Glucagon-like peptide-1 enhances production of insulin in insulin-producing cells derived from mouse embryonic stem cells J. Endocrinol., August 1, 2005; 186(2): 343 - 352. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Ogawa, J. F. List, J. F. Habener, and T. Maki Cure of Overt Diabetes in NOD Mice by Transient Treatment With Anti-Lymphocyte Serum and Exendin-4 Diabetes, July 1, 2004; 53(7): 1700 - 1705. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. D'Alessio and T. P. Vahl Glucagon-like peptide 1: evolution of an incretin into a treatment for diabetes Am J Physiol Endocrinol Metab, June 1, 2004; 286(6): E882 - E890. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. L. Brubaker and D. J. Drucker Minireview: Glucagon-Like Peptides Regulate Cell Proliferation and Apoptosis in the Pancreas, Gut, and Central Nervous System Endocrinology, June 1, 2004; 145(6): 2653 - 2659. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Farilla, A. Bulotta, B. Hirshberg, S. Li Calzi, N. Khoury, H. Noushmehr, C. Bertolotto, U. Di Mario, D. M. Harlan, and R. Perfetti Glucagon-Like Peptide 1 Inhibits Cell Apoptosis and Improves Glucose Responsiveness of Freshly Isolated Human Islets Endocrinology, December 1, 2003; 144(12): 5149 - 5158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Drucker Enhancing Incretin Action for the Treatment of Type 2 Diabetes Diabetes Care, October 1, 2003; 26(10): 2929 - 2940. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Drucker Glucagon-Like Peptides: Regulators of Cell Proliferation, Differentiation, and Apoptosis Mol. Endocrinol., February 1, 2003; 17(2): 161 - 171. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
