Growth and Functional Maturation of -Cells in Implants of Endocrine Cells Purified From Prenatal Porcine Pancreas
http://www.100md.com
《糖尿病学杂志》
1 Diabetes Research Center, Brussels Free University and Juvenile Diabetes Research Foundation Center for Beta Cell Therapy in Diabetes, Brussels, Belgium
2 Research Laboratory for Stereology and Neuroscience, H:S Bispebjerg Hospital, University Hospital of Copenhagen, Denmark
BrdU, bromodeoxyuridine; CK-7, cytokeratin
ABSTRACT
The development of islet cell transplantation as a cure for diabetes is limited by the shortage of human donor organs. Moreover, currently used grafts exhibit a marginal -cell mass with an apparently low capacity for -cell renewal and growth. Although duct-associated nonendocrine cells have often been suggested as a potential source for -cell production, recent work in mice has demonstrated the role of -cells in postnatal growth of the pancreatic -cell mass. The present study investigated whether the -cell mass can grow in implants that are virtually devoid of nonendocrine cells. Endocrine islet cells were purified from prenatal porcine pancreases (gestation >110 days) and implanted under the kidney capsule of nude mice. -Cells initially presented with signs of immaturity: small size, low insulin content, undetectable C-peptide release, and an inability to correct hyperglycemia. They exhibited a proliferative activity that was highest during posttransplant week 1 (2.6 and 5% bromodeoxyuridine [BrdU]-positive -cells 4 and 72 h posttransplant) and then decreased over 20 weeks to rates measured in the pancreas (0.2% BrdU-positive cells). -Cell proliferation in implants first compensated for -cell loss during posttransplant week 1 and then increased the -cell number fourfold between posttransplant weeks 1 and 20. Rates of -cell proliferation were only shortly and moderately increased, which explained the shift in cellular composition of the implant (-cell 40 vs. 90% and -cell 40 vs. 7% at the start and posttransplant week 20, respectively). -Cells progressively matured during the 20 weeks after transplantation, with a twofold increase in cell volume, a sixfold increase in cellular insulin content, plasma C-peptide levels of 1eC2 ng/ml, and an ability to correct diabetes. They became structurally organized as homogenous clusters with their secretory vesicles polarized toward fenestrated capillaries. We concluded that the immature -cell phenotype provides grafts with a marked potential for -cell growth and differentiation and hence may have a potential role in curing diabetes. Cells with this phenotype can be isolated from prenatal organs; their presence in postnatal organs needs to be investigated.
Islet cell transplantation has long been considered as a potential cure for diabetes (1,2). A transplantation protocol has been reported that has achieved a consistent insulin independence that has been maintained beyond the first posttransplantation year (3). Clinical success has also accentuated obstacles to wide therapeutic implementation; a major drawback is the scarcity in islet grafts that can be isolated from human donor organs. Moreover, implanted -cell masses appear to be marginal and may therefore become inadequate with time. In rodent models, the size of adult islet implants decreases with time, probably as result of inadequate -cell renewal and/or neogenesis (4eC11). The capacity of fetal and neonatal pancreases to generate new -cells makes this tissue attractive for producing large numbers of -cells in vitro or for preparing grafts in which the -cell mass grows and matures (12eC16). Pancreatic cell implants from fetal or neonatal rodents or pigs have been found to increase their insulin content or -cell mass after transplantation and thus become able to normalize the diabetic state (12eC20). The mechanisms underlying this growth and functional maturation have not yet been clarified. They may involve -cell proliferation and/or -cell neoformation from duct cells or other cell types yet to be identified (13,15,19,21,22). The relative contribution of each pathway is still unknown and might vary with the cellular composition of the graft and the developmental stage of the donor organ. In all models where growth of the -cell mass has been demonstrated so far, the graft contained a majority of nonendocrine cells, some of which were seen as -cell precursors. In the present study, we implanted purified endocrine cells isolated from late-fetal pig pancreases and investigated the growth and functional maturation of -cells in normal and diabetic immunodeficient mice. Our data demonstrate that implants with immature -cells develop the ability to cure diabetes and relate this property to the proliferation and maturation of the -cells and their structural organization.
RESEARCH DESIGN AND METHODS
Prenatal porcine pancreatic endocrine cells were obtained from Beta Cell (Zellik, Belgium). They were prepared according to a procedure developed in our laboratory using fetuses of Belgian land race sows after 110eC115 days of gestation. Pancreases were dissociated by enzymatic digestion with collagenase P (Roche, Indianapolis, IN; 0.3 mg/ml) and trituration in calcium-free medium. Single endocrine cells were isolated by counterflow elutriation, purified by gradient centrifugation, and cultured in serum-free HAM’s F10 medium containing 6 mmol/l glucose, 5 mmol/l nicotinamide, 2 mmol/l glutamine, 50 e蘭ol/l isobutylmethylxanthine, 10eC6 mol/l hydrocortisone, and 0.5% BSA. At the start and after 1 week of culture, samples were taken to determine their cellular composition and hormone content (23,24). Grafts composed of >90% pure endocrine islet cells and containing 0.5eC1 million -cells were prepared from suspensions cultured for 1 week.
Transplantation of prenatal porcine pancreatic endocrine cells in nude mice.
Male immunodeficient nu/nu BALB/c ByJICo mice (age 8 weeks; Iffa Credo, France) were selected as recipients. Animals were maintained in acclimatized rooms with free access to water and pellet food. Transplants were carried out in nondiabetic and diabetic animals. Diabetes was induced by intravenous injection of alloxan (90 mg/kg; Sigma, Munich, Germany) the day before transplantation. Only animals with nonfasting glycemia >500 mg/dl were included in the study; because most of them died during the next 2 weeks irrespective of the presence of a fetal -cell implant, a "life-saving" rat islet graft was placed under the contralateral kidney capsule until the time that the fetal cell implants had grown and matured according to the studies in nondiabetic recipients; the islet-bearing kidney was then removed and the animals were followed metabolically. In the study of Korbutt et al. (15), such a "life-saving" rat islet graft was not necessary, possibly because a more differentiated neonatal pig islet cell preparation was used following pretreatment with high nicotinamide (10 mmol/l), which has been reported to favor -cell differentiation (25). Another reason could have been a lower degree of -cell depletion after alloxan in the Korbutt et al. study. The study protocols were approved by the Ethical Committee of Brussels Free University-VUB and carried out in accordance with Belgian and European Union regulations.
Mice were anesthetized with 2,2,2 tribromoethanol solution (Avertin; Sigma; 500 mg/kg body wt i.p.). Grafts were implanted within a blood clot that was placed under the kidney capsule (26); the use of a blood clot prevented a major cell loss after injection of the small and loosely formed cell aggregates. Control mice received a blood clot without cells. The function of the implant was monitored through plasma levels of glucose and porcine C-peptide. Tail vein blood was collected in EDTA/aprotinin (Trasylol) and plasma was stored at eC20°C until assayed for porcine C-peptide using a commercial radioimmunoassay kit (Linco, St. Charles, MO). Implants were removed at different time points after the transplantation and were processed for immunohistochemistry, electron microscopy, or the hormone assay. In one series of animals, bromodeoxyuridine (BrdU; 50 mg/kg) was injected intraperitoneally 1 h before the implant was removed and the percentage of proliferating cells was determined.
Analysis of cellular composition of grafts and implants.
Tissue samples were fixed in formalin and paraffin sections were incubated overnight with rabbit anti-synaptophysin (1:20; Dako, Glostrup, Denmark) (27), guinea pig anti-insulin (1:5,000) (27), rabbit anti-glucagon (1:5,000) (27), rabbit anti-somatostatin (1:5,000) (27), or rabbit anti-pancreatic polypeptide (1:50,000; gift of Dr. R.E. Chance, Lilly, Indianapolis, IN) (26). After the sections were washed in PBS (0.1 mmol, pH 7.4), they were incubated with peroxidase-labeled anti-mouse (1:20; Amersham, Amersham, U.K.), anti-rabbit (1:20; Amersham), or antieCguinea pig (1:100; Cappel, Aurora, OH) secondary antibody for 30 min at room temperature. The peroxidase reaction was developed by incubating sections in 0.3% H2O2 and 0.15% 3.3' diaminobenzidine tetrachloride (Sigma, St. Louis, MO). For the double-labeling immunohistochemistry, antieCguinea pig Ig labeled with fluorescein isothiocyanate and Cy3-labeled anti-mouse Ig secondary antibodies (Jackson Immunoresearch, West Groove, PA) were used. Proliferating cells were detected with a BrdU antibody (1:20; Cappel) and identified by an antibody against synaptophysin (28), insulin, glucagon, somatostatin, or pancreatic polypeptide. For each population positive for the endocrine marker, the percentage of BrdU-positive cells was determined and expressed as the proliferation index. For each implant, at least 2,000 cells were analyzed; no regional differences in structural organization were noticed. We also counted the percentage of insulin-positive cells with signs of apoptosis (i.e., condensed or fragmented nuclei after propidium iodide staining) (29).
Morphometry of implants.
The size of endocrine cell populations was determined according to Cavalieri’s principle. Total volumes of synaptophysin-, insulin-, and glucagon-positive cells were estimated by point counting (27). Grafts were cut in 5-e蘭-thick paraffin sections. Every 40th section was then immunostained and stereologically examined using a light microscope that projected the image onto a 600-point grid at a final magnification of 140x, with an area per point of 20,154 e蘭2. For each graft, 700eC2,000 points hitting the antibody-positive cells were counted. The total volume of antibody-positive cells was calculated by multiplying the total number of points overlaying the antibody-positive cells per section with the area per point and the distance between subsequent immune-stained sections (200 e蘭). To measure the mean volume of individual insulin cells, grafts were cut in 1-e蘭-thick sections, every 40th section was immunostained for insulin, and the total insulin cell volume was estimated as described above. The total insulin cell number was determined by the physical dissector principle (30,31) in randomly sampled and insulin-stained pairs of consecutive sections, including a reference (every 40th) and look-up (the consecutive) section. The mean volume of insulin cells was then calculated by dividing the total insulin cell volume by the total insulin cell number.
Statistical analysis.
Data are expressed as means ± SD of n independent experiments. The statistical significance of differences in plasma porcine C-peptide and glycemia levels between experimental and control groups was calculated by the Kruskal-Wallis test. Changes in cellular composition, total volume, mean insulin cell volume and number, graft insulin content, proliferation, and apoptosis index were tested by the Mann-Whitney U test. P <0.05 was considered significant.
RESULTS
Characteristics of fetal porcine -cell graft.
Fetal porcine islet cell grafts were composed of >90% endocrine cells, which mainly corresponded to insulin- and glucagon-positive cells in similar proportions (each representing 39 ± 5% of all cells); somatostatin-positive (7 ± 2%) and pancreatic polypeptideeCpositive (6 ± 3%) cells were much less abundant (Fig. 1). The remaining cells (<10%) corresponded to cytokeratin (CK-7)-positive cells (5.1 ± 2.3% of all cells) and damaged cells. We observed that >90% of the cells were positive for synaptophysin, indicating that this neuroendocrine marker can be used to identify fetal pancreatic endocrine cells. The CK-7 positive cells were copurified epithelial duct cells (32). Grafts contained 0.8 ± 0.1 x 106 -cells and 4.6 ± 1.6 e蘥 insulin; the insulin content of the -cells averaged 5.2 e蘥/106 cells. -Cells had an average cellular volume of 800 e蘭3, which is comparable with that of neonatal rat -cells (33) and smaller than that of adult rodent -cells (30,33). Electron microscopy illustrated endocrine cell aggregates with intact ultrastructure (<5% dead cells). Few apoptotic -cells were detected (0.5 ± 0.2%). Under basal serum-free culture conditions, 0.5 ± 0.3% of insulin cells were BrdU positive after a 1-h incubation with BrdU.
Change in cellular composition of fetal porcine -cell implants.
At posttransplant day 1, the composition of the implant (Fig. 2) was similar to that of the isolated graft, with equal percentages of insulin- (39 ± 5%) and glucagon- (39 ± 5%) positive cells, each contributing similar total cell volumes to the endocrine volume of the implant. During posttransplant week 1, no change occurred in the volume of the insulin cell population. However, that of glucagon cells decreased by 40% (Table 1) in association with a decrease in the percentage of glucagon cells and an increase in the percentage of insulin cells (Fig. 2). From posttransplant week 1 onward, the total volume of insulin cells increased with time (threefold between posttransplant weeks 1 and 10, twofold between posttransplant weeks 10 and 20) (Table 1); this was not the case for the total volume of glucagon cells, which remained stable (Table 1). The proportion of insulin cells further increased to 89% at posttransplant week 20, whereas that of glucagon cells decreased to 7% (Fig. 2).
The increase in the total volume of insulin cells was caused by an increase in the mean individual cell volume as well as an increase in the number of -cells. Between posttransplant weeks 1 and 20, the mean individual volume of insulin cells almost doubled (to an average volume of 1,500 e蘭3) whereas the number of insulin cells increased fourfold (Fig. 3). No significant difference was detected between the number of -cells on posttransplant day 1 and at posttransplant week 1 (Fig. 3). The increase in the total volume of insulin cells between posttransplant weeks 1 and 20 was accompanied by an increase in insulin content of the implant rising to 17- and 50-fold higher values, respectively, at posttransplant weeks 10 and 20 (Fig. 3). This increase in insulin content of the implant was much more pronounced than the increase in -cell number or the total volume of insulin cells, which indicated that the -cells increased their individual insulin store. At posttransplant week 20, the insulin content of the -cells averaged 31 e蘥/106 cells, which was sixfold higher than at the start.
Structural organization of fetal porcine -cell implants.
As well as changing their cellular composition, implants underwent a structural organization. At posttransplant day 1, they were composed of dispersed single cells, small cell aggregates, and blood cells without structural organization (Fig. 1). At posttransplant week 20, the implants appeared as a white compact mass on the kidney surface, with a diameter of 3eC8 mm and clearly visible interconnected capillaries (Fig. 4A). When viewed microscopically, they presented a homogenous assembly of endocrine cell clusters (Fig. 4B) that were predominantly formed by insulin-positive cells with a few other endocrine islet cells scattered at their border (Fig. 4C and D). The space in the center of some endocrine clusters (Fig. 4C) was filled with erythrocytes without endothelial cell lining (Fig. 4E). These lacunae do not exhibit the tubular or matrix pattern that is characteristic for vasculogenic mimicry; their formation and functional significance is so far unknown. Within the endocrine cell masses, capillaries with fenestrated endothelial cells were detected, toward which the endocrine cells were oriented with their pole of secretory vesicles (Fig. 4E and F, arrows).
Rates of -cell proliferation and apoptosis in fetal porcine -cell implants.
During posttransplant week 1, 1eC6% of insulin cells were positive for BrdU 1 h after this compound was injected (Fig. 4G). This proliferation index was already elevated 4 h after transplant (2.6 ± 0.7%) and reached peak values of 5 ± 0.7% after 72 h (Fig. 5). Between posttransplant week 1 and 10, the percentage of BrdU-positive insulin cells remained significantly higher than in the prenatal pancreas in situ (0.2%) but then slowly decreased with time to 0.3% at posttransplant week 20. The percentage of BrdU-positive glucagon cells was lower during the first week (2.3% after 72 h) and returned to basal levels from posttransplant week 1 and beyond (0.3% BrdU positive cells).
The first 72 h after the transplant were also characterized by a higher rate of insulin cell apoptosis (1eC4%) (Fig. 4H), which was markedly higher than in the graft before transplantation (0.5 ± 0.2%). This index decreased to <0.2% at posttransplant week 1 or later (Fig. 5).
Metabolic effects of fetal porcine -cell implants.
Porcine C-peptide was not detectable in plasma during the first 2 posttransplant weeks but was consistently measured from posttransplant week 5 and beyond, increasing to plateau levels of 1.0eC1.5 ng/ml at posttransplant weeks 10eC20 (Fig. 6). Blood glucose levels were not different from those in nontransplanted controls during the first 5 posttransplant weeks (91 ± 9 vs. 94 ± 8 mg/dl) but then progressively decreased to reach a level at and beyond posttransplant week 15 of 59 ± 27 mg/dl, which was significantly lower than the level in normal controls (110 ± 10 mg/dl; P < 0.05) (Fig. 6). This metabolic effect was associated with a 50% decrease in pancreatic insulin content (25 ± 6 vs. 50 ± 7 e蘥 in sham-operated normal controls; P < 0.05).
When fetal porcine islet cell preparations of similar size and composition were implanted in alloxan-induced diabetic nude mice, they did not immediately correct the hyperglycemic state and the mice died during the first 2 weeks after the transplant. This outcome indicated the metabolic incapacity of the fetal porcine -cell implant during the first posttransplant weeks and correlated with undetectable circulating and low tissue levels of porcine C-peptide or insulin during this period. We therefore examined the metabolic capacity of porcine -cell implants at a later time when they had undergone the above-described changes in composition and organization and had reached plateau levels of circulating C-peptide (i.e., beyond posttransplant week 10). To this end, the fetal pig -cell preparations were transplanted in diabetic mice that simultaneously received isolated adult rat islets under the contralateral kidney capsule. The rat islet implant established normoglycemia (Fig. 7) and thus prevented the death of these animals. The rise in circulating porcine C-peptide levels was similar to that in noneCalloxan-administered recipients that had received a fetal pig -cell graft (Fig. 7). Removal of the rat islet implant at posttransplant week 15 did not alter porcine C-peptide levels nor did it reverse the normoglycemic state, indicating that the fetal porcine -cell implant contained a metabolically responsive and adequate -cell mass. When the animals were killed at posttransplant week 17, their pancreatic insulin content was in the diabetic range (1.9 ± 1.2 e蘥 [n = 6] vs. 34 ± 11 e蘥 in noneCalloxan-administered recipients of fetal porcine -cell grafts) whereas the insulin content of their implant (60 ± 29 e蘥) was similar to that of the noneCalloxan-administered recipients (87 ± 29 e蘥 [n = 7]; P > 0.05).
DISCUSSION
The shortage in human donor organs has raised interests in pigs as a potential source of islet cell grafts for diabetic patients. Techniques have been developed for preparing insulin-producing tissue from fetal, neonatal, and adult pig pancreases (14,15,20,34). Grafts from all donor ages have been shown to normalize diabetes in immunodeficient mice (15,22,35eC37). The yield and purity of -cells is markedly lower in isolates from fetal and neonatal organs, but their implants develop a growing -cell mass, which is seen as an advantage. Islet cell grafts are indeed expected to achieve long-term metabolic control, which, in the normal endocrine pancreas, depends on the ability for -cell renewal and growth to compensate for a natural -cell loss and increasing metabolic demands (38eC41).
There is so far no evidence for adaptive growth or -cell renewal in implants from adult pig, rodent, or human organs; instead, a progressive decline in -cell mass has been observed with time and is considered to be responsible for the loss of metabolic control in long-term implants (9,10,42). On the other hand, implants of fetal pancreatic tissue grow in total volume, -cell mass, and insulin content, bringing their overall function with time to a state that corrects hyperglycemia (15,18,19,22). The rapidity of metabolic normalization may vary with the amount of tissue and its cellular composition and also with the stage of development and the species. For porcine islet cell clusters isolated at gestation days 50eC70 and implanted under the kidney capsule of nude mice, the correction of diabetes is achieved after only 16eC20 weeks; this period can be shortened by the administration of nicotinamide, which is considered to stimulate -cell differentiation (25,43). This was also the case for porcine neonatal islet cell clusters (15). The implants undergo a maturation process that seems to involve both neoformation and functional maturation of -cells (19,22,43). The relative contribution of both components is not yet clear.
It is also unknown to which extent new -cells are formed from nonendocrine cells or other -cells (22). Most authors underline the importance of a differentiation of nonendocrine cells to -cells (19,20,22,44). At the time of implantation, these grafts are indeed mainly composed of nonendocrine cells; the progressive disappearance of the nonendocrine cells is found to be compatible with a transition to -cells. The nonendocrine-to-endocrine differentiation route is also supported by the observation that several -cells share markers with adjacent nonendocrine cells (22). Moreover, little evidence has been provided for a major contribution of -cell proliferation. A study comparing the outcome of fetal islet cell clusters with a low percentage of -cells and fetal -cell aggregates even concluded that the absence of nonendocrine cells markedly reduced the normalizing capacity of the grafts (44).
To investigate more directly the growth potential of fetal -cells, we undertook a morphometric study on grafts of purified endocrine islet cells that were virtually devoid of nonendocrine cells. Purified endocrine islet cell grafts were prepared from late fetal (110eC115 days gestation) pig pancreases. They consisted of >90% endocrine cells, with similar percentages (40%) of insulin- and glucagon-positive cells. This composition markedly differed from that of fetal or neonatal porcine islet cell clusters prepared after 50eC70 days of gestation (14) or 1eC3 days after birth (15,20), which have 5 or 24% -cells and 1 or 8% -cells, respectively. The -cells in our preparation were immature, as determined by comparing them with adult -cells; their cellular volume was twofold smaller (30,33) and their insulin content (5 e蘥/million -cells) was sixfold lower. They were also unable to correct diabetes in nude mice when transplanted in numbers (20 million -cells/kg body wt) that were 5- to 10-fold higher than those of adult -cells that normalize rodent or human diabetes (4,5,7,45,46). The porcine -cells developed a functionally mature state over 10eC20 weeks after transplantation, as evidenced by their progressive increase in size and insulin content, the appearance of circulating porcine C-peptide, and their progressive gain of metabolic control.
In addition to presenting signs of maturing -cells, the endocrine cell implants also exhibited a continuous formation of new -cells. The number of -cells increased 4.3-fold between the end of posttransplant week 1 and posttransplant week 20. This increase could not be attributed to nonendocrine cells as these were virtually absent (<1% at the end of posttransplant week 1). Instead, it was due to the high proliferative activity of the -cells: at posttransplant week 1, the percentage of BrdU-positive -cells was 12-fold higher than in the prenatal pancreas in situ and remained elevated, although at a lower level, until posttransplant week 10. This index was also sixfold higher than in the same preparation tested in vitro in the absence of serum. Factors released during the inflammatory and tissue repair process probably stimulated the -cells that are prone to proliferation. Only 1 day after the implantation injury, the proliferating fibroblasts and endothelial cells were noticed and 2 days later, granulation tissue with small blood vessels was observed. Despite this increased rate of -cell proliferation during posttransplant week 1, neither the number nor total volume of -cells increased. This occurrence might have reflected a simultaneous equal loss of -cells that was compatible with the elevated percentages of apoptotic -cells during this period. Because apoptotic cells are rapidly cleared, it is not possible to calculate the cell loss from the percentages of apoptotic -cells. Apoptotic -cells were found in similar percentages, an observation that is consistent with nonspecific damage in the implant site. However, there were less BrdU-positive glucagon cells than insulin cells (2 vs. 5% at day 3), which explains the reduction in total volume of this cell population at posttransplant week 1. The rates of BrdU-positive glucagon cells remained low during the entire study period in contrast to our observations of -cells. The quantitative differences between both cell populations thus further increased with time, resulting at posttransplant week 20 in an implant with 89% -cells and only 7% -cells. An abundance of - over -cells has been previously described for grafts prepared from fetal and neonatal pancreatic tissue as well as for adult islet grafts (15,19,46,47).
The observed growth in -cell number and individual cell volume did occur under normoglycemic conditions, as maintained by the mouse endocrine pancreas or a rat islet graft. Under this condition, the implant induced, from posttransplant week 10 on, plasma porcine C-peptide levels of 1eC2 ng/ml. This level was maintained when the rat islet graft was removed; it corresponded to a -cell mass that had grown and matured to a higher insulin content than in the normal mouse pancreas. At this level of growth and differentiation, the porcine -cell implant prevented the development of diabetes, with no decline in its insulin reserves. Morphological examination revealed that the metabolically adequate porcine fetal -cell implant was structurally organized in contrast to the randomly distributed small cell clusters in the initial implant. It consisted of homogenous -cell clusters with peripherally scattered -cells surrounded by fenestrated capillaries, as described in normal islet tissue (48). The -cells presented intracellular signs of polarity, with their pole of secretory vesicles located at the pole that faced the capillaries. Between the clusters, blood-filled lacunae devoid of an endothelial lining were also noticed. The revascularization process proceeded throughout the 20-week study period, as evidenced by an ongoing capillary sprouting.
The present study showed that endocrine islet cells can be purified from porcine fetal pancreases, opening the potential for large scale isolations of -cells. Our preparations contained 40% -cells with signs of immaturity. When transplanted under the kidney capsule of normoglycemic mice, the -cells exhibited a prolonged proliferating activity, first compensating for -cell loss during engraftment and then increasing the -cell number fourfold over the subsequent 20 weeks. These data indicate that immature -cells can achieve a sustained renewal and growth of the -cell mass. It is conceivable that immature -cells were also responsible for the growth of the -cell mass that has been observed in human duct cell preparations isolated from young human donors (27). Our observations are consistent with recent work showing that the proliferation of -cells is the main mechanism for growth of the -cell mass in the postnatal mouse pancreas (49,50). They also illustrate that impure islet cell preparations are not needed to achieve -cell renewal or adaptive growth in implants of pancreatic tissue.
ACKNOWLEDGMENTS
This study was supported by grants from the Juvenile Diabetes Research Foundation (4eC2001-434), the Belgian Fund for Scientific Research (G.0375.00, G.0358.03), the Institute for the Promotion of Innovation by Science and Technology in Flanders (020442), and the Interuniversity Poles of Attraction Programme (IUAP P5/17) from the Belgian Science Policy.
We thank the team of Beta Cell for providing the fetal porcine endocrine cell preparations and Nicole Buelens and Violeta Ardelean for excellent technical assistance.
REFERENCES
Lacy PE, Scharp DW: Islet transplantation in treating diabetes. Annu Rev Med 37:33eC40, 1986
Robertson RP: Islet transplantation as a treatment for diabetes: a work in progress. N Engl J Med 350:694eC705, 2004
Shapiro AM, Lakey JR, Ryan EA, Korbutt GS, Toth E, Warnock GL, Kneteman NM, Rajotte RV: Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med 343:230eC238, 2000
Pipeleers-Marichal MA, Pipeleers DG, Cutler J, Lacy PE, Kipnis DM: Metabolic and morphologic studies in intraportal-isleteCtransplanted rats. Diabetes 25:1041eC1051, 1976
Orloff MJ, Macedo A, Greenleaf GE, Girard B: Comparison of the metabolic control of diabetes achieved by whole pancreas transplantation and pancreatic islet transplantation in rats. Transplantation 45:307eC312, 1988
Hiller WF, Klempnauer J, Luck R, Steiniger B: Progressive deterioration of endocrine function after intraportal but not kidney subcapsular rat islet transplantation. Diabetes 40:134eC140, 1991
Pipeleers DG, Pipeleers-Marichal M, Vanbrabandt B, Duys S: Transplantation of purified islet cells in diabetic rats. II. Immunogenicity of allografted islet beta-cells. Diabetes 40:920eC930, 1991
Keymeulen B, Teng H, Vetri M, Gorus F, In’t Veld P, Pipeleers DG: Effect of donor islet mass on metabolic normalization in streptozotocin-diabetic rats. Diabetologia 35:719eC724, 1992
Montanya E, Bonner-Weir S, Weir GC: Beta cell mass and growth after syngeneic islet cell transplantation in normal and streptozocin diabetic C57BL/6 mice. J Clin Invest 91:780eC787, 1993
Davalli AM, Ogawa Y, Ricordi C, Scharp DW, Bonner-Weir S, Weir GC: A selective decrease in the beta cell mass of human islets transplanted into diabetic nude mice. Transplantation 59:817eC820, 1995
Biarnes M, Montolio M, Nacher V, Raurell M, Soler J, Montanya E: -Cell death and mass in syngeneically transplanted islets exposed to short- and long-term hyperglycemia. Diabetes 51:66eC72, 2002
Lazarow A, Wells LJ, Carpenter AM, Hegre OD, Leonard RJ, McEvoy RC: Banting Lecture 1973: Islet differentiation, organ culture, and transplantation. Diabetes 22:877eC912, 1973
Sandler S, Andersson A, Schnell A, Mellgren A, Tollemar J, Borg H, Petersson B, Groth CG, Hellerstrom C: Tissue culture of human fetal pancreas: development and function of B-cells in vitro and transplantation of explants to nude mice. Diabetes 34:1113eC1119, 1985
Korsgren O, Sandler S, Landstrom AS, Jansson L, Andersson A: Large-scale production of fetal porcine pancreatic isletlike cell clusters: an experimental tool for studies of islet cell differentiation and xenotransplantation. Transplantation 45:509eC514, 1988
Korbutt GS, Elliott JF, Ao Z, Smith DK, Warnock GL, Rajotte RV: Large scale isolation, growth, and function of porcine neonatal islet cells. J Clin Invest 97:2119eC2129, 1996
Castaing M, Peault B, Basmaciogullari A, Casal I, Czernichow P, Scharfmann R: Blood glucose normalization upon transplantation of human embryonic pancreas into beta-cell-deficient SCID mice. Diabetologia 44:2066eC2076, 2001
Hegre OD, Leonard RJ, Rusin JD, Lazarow A: Transplantation of the fetal rat pancreas: quantitative morphological analysis of islet tissue growth. Anat Rec 185:209eC221, 1976
Brown J, Heininger D, Kuret J, Mullen Y: Islet cells grow after transplantation of fetal pancreas and control of diabetes. Diabetes 30:9eC13, 1981
Korsgren O, Jansson L, Eizirik D, Andersson A: Functional and morphological differentiation of fetal porcine islet-like cell clusters after transplantation into nude mice. Diabetologia 34:379eC386, 1991
Yoon KH, Quickel RR, Tatarkiewicz K, Ulrich TR, Hollister-Lock J, Trivedi N, Bonner-Weir S, Weir GC: Differentiation and expansion of beta cell mass in porcine neonatal pancreatic cell clusters transplanted into nude mice. Cell Transplant 8:673eC689, 1999
Leonard RJ, Lazarow A, Hegre OD: Pancreatic islet transplantation in the rat. Diabetes 22:413eC428, 1973
Trivedi N, Hollister-Lock J, Lopez-Avalos MD, O’Neil JJ, Keegan M, Bonner-Weir S, Weir GC: Increase in beta-cell mass in transplanted porcine neonatal pancreatic cell clusters is due to proliferation of beta-cells and differentiation of duct cells. Endocrinology 142:2115eC2122, 2001
Pipeleers DG, in’t Veld PA, Van de Winkel M, Maes E, Schuit FC, Gepts W: A new in vitro model for the study of pancreatic A and B cells. Endocrinology 117:806eC816, 1985
Ling Z, Pipeleers DG: Prolonged exposure of human beta cells to elevated glucose levels results in sustained cellular activation leading to a loss of glucose regulation. J Clin Invest 98:2805eC2812, 1996
Korsgren O, Sandler S, Jansson L, Groth CG, Hellerstrom C, Andersson A: Effects of culture conditions on formation and hormone content of fetal porcine isletlike cell clusters. Diabetes 38 (Suppl. 1):209eC212, 1989
Gray DW, McShane P, Morris PJ: The effect of hyperglycemia on isolated rodent islets transplanted to the kidney capsule site. Transplantation 41:699eC703, 1986
Bogdani M, Lefebvre V, Buelens N, Bock T, Pipeleers-Marichal M, In’t Veld P, Pipeleers D: Formation of insulin-positive cells in implants of human pancreatic duct cell preparations from young donors. Diabetologia 46:830eC838, 2003
Wiedenmann B, Franke WW, Kuhn C, Moll R, Gould VE: Synaptophysin: a marker protein for neuroendocrine cells and neoplasms. Proc Natl Acad Sci U S A 83:3500eC3504, 1986
Hoorens A, Van de Casteele M, Kloppel G, Pipeleers D: Glucose promotes survival of rat pancreatic beta cells by activating synthesis of proteins which suppress a constitutive apoptotic program. J Clin Invest 98:1568eC1574, 1996
Bock T, Svenstrup K, Pakkenberg B, Buschard K: Unbiased estimation of total beta-cell number and mean beta-cell volume in rodent pancreas. APMIS 107:791eC799, 1999
Gundersen HJ: Stereology of arbitrary particles: a review of unbiased number and size estimators and the presentation of some new ones, in memory of William R. Thompson. J Microsc 143:3eC45, 1986
Kasper M, von Dorsche H, Stosiek P: Changes in the distribution of intermediate filament proteins and collagen IV in fetal and adult human pancreas. I. Localization of cytokeratin polypeptides. Histochemistry 96:271eC277, 1991
Svenstrup K, Skau M, Pakkenberg B, Buschard K, Bock T: Postnatal development of beta-cells in rats: proposed explanatory model. APMIS 110:372eC378, 2002
Toso C, Brandhorst D, Oberholzer J, Triponez F, Buhler L, Morel P: Isolation of adult porcine islets of Langerhans. Cell Transplant 9:297eC305, 2000
Korsgren O, Jansson L: Porcine islet-like cell clusters cure diabetic nude rats when transplanted under the kidney capsule, but not when implanted into the liver or spleen. Cell Transplant 3:49eC54, 1994
Vo L, Tuch BE, Wright DC, Keogh GW, Roberts S, Simpson AM, Yao M, Tabiin MT, Valencia SK, Scott H: Lowering of blood glucose to nondiabetic levels in a hyperglycemic pig by allografting of fetal pig isletlike cell clusters. Transplantation 71:1671eC1677, 2001
Davalli AM, Ogawa Y, Scaglia L, Wu YJ, Hollister J, Bonner-Weir S, Weir GC: Function, mass, and replication of porcine and rat islets transplanted into diabetic nude mice. Diabetes 44:104eC111, 1995
Bonner-Weir S: Regulation of pancreatic beta-cell mass in vivo. Recent Prog Horm Res 49:91eC104, 1994
Sorenson RL, Brelje TC: Adaptation of islets of Langerhans to pregnancy: beta-cell growth, enhanced insulin secretion and the role of lactogenic hormones. Horm Metab Res 29:301eC307, 1997
Bonner-Weir S: Perspective: postnatal pancreatic beta cell growth. Endocrinology 141:1926eC1929, 2000
Withers DJ, White M: Perspective: the insulin signaling system: a common link in the pathogenesis of type 2 diabetes. Endocrinology 141:1917eC1921, 2000
Keymeulen B, Korbutt G, De Paepe M, Gorus F, Kloppel G, Pipeleers DG: Long-term metabolic control by rat islet grafts depends on the composition of the implant. Diabetes 45:1814eC1821, 1996
Otonkoski T, Ustinov J, Rasilainen S, Kallio E, Korsgren O, Hayry P: Differentiation and maturation of porcine fetal islet cells in vitro and after transplantation. Transplantation 68:1674eC1683, 1999
Tu J, Khoury P, Williams L, Tuch BE: Comparison of fetal porcine aggregates of purified beta-cells versus islet-like cell clusters as a treatment of diabetes. Cell Transplant 13:525eC534, 2004
Pipeleers DG, Pipeleers-Marichal M, Hannaert JC, Berghmans M, In’t Veld PA, Rozing J, Van de Winkel M, Gepts W: Transplantation of purified islet cells in diabetic rats. I. Standardization of islet cell grafts. Diabetes 40:908eC919, 1991
Keymeulen B, Ling Z, Gorus FK, Delvaux G, Bouwens L, Grupping A, Hendrieckx C, Pipeleers-Marichal M, Van Schravendijk C, Salmela K, Pipeleers DG: Implantation of standardized beta-cell grafts in a liver segment of IDDM patients: graft and recipient characteristics in two cases of insulin-independence under maintenance immunosuppression for prior kidney graft. Diabetologia 41:452eC459, 1998
Street CN, Lakey JR, Shapiro AM, Imes S, Rajotte RV, Ryan EA, Lyon JG, Kin T, Avila J, Tsujimura T, Korbutt GS: Islet graft assessment in the Edmonton Protocol: implications for predicting long-term clinical outcome. Diabetes 53:3107eC3114, 2004
Bearer EL, Orci L: Endothelial fenestral diaphragms: a quick-freeze, deep-etch study. J Cell Biol 100:418eC428, 1985
Bock T, Pakkenberg B, Buschard K: Increased islet volume but unchanged islet number in ob/ob mice. Diabetes 52:1716eC1722, 2003
Dor Y, Brown J, Martinez OI, Melton DA: Adult pancreatic beta-cells are formed by self-duplication rather than stem-cell differentiation. Nature 429:41eC46, 2004(Marika Bogdani, Krista Su)
2 Research Laboratory for Stereology and Neuroscience, H:S Bispebjerg Hospital, University Hospital of Copenhagen, Denmark
BrdU, bromodeoxyuridine; CK-7, cytokeratin
ABSTRACT
The development of islet cell transplantation as a cure for diabetes is limited by the shortage of human donor organs. Moreover, currently used grafts exhibit a marginal -cell mass with an apparently low capacity for -cell renewal and growth. Although duct-associated nonendocrine cells have often been suggested as a potential source for -cell production, recent work in mice has demonstrated the role of -cells in postnatal growth of the pancreatic -cell mass. The present study investigated whether the -cell mass can grow in implants that are virtually devoid of nonendocrine cells. Endocrine islet cells were purified from prenatal porcine pancreases (gestation >110 days) and implanted under the kidney capsule of nude mice. -Cells initially presented with signs of immaturity: small size, low insulin content, undetectable C-peptide release, and an inability to correct hyperglycemia. They exhibited a proliferative activity that was highest during posttransplant week 1 (2.6 and 5% bromodeoxyuridine [BrdU]-positive -cells 4 and 72 h posttransplant) and then decreased over 20 weeks to rates measured in the pancreas (0.2% BrdU-positive cells). -Cell proliferation in implants first compensated for -cell loss during posttransplant week 1 and then increased the -cell number fourfold between posttransplant weeks 1 and 20. Rates of -cell proliferation were only shortly and moderately increased, which explained the shift in cellular composition of the implant (-cell 40 vs. 90% and -cell 40 vs. 7% at the start and posttransplant week 20, respectively). -Cells progressively matured during the 20 weeks after transplantation, with a twofold increase in cell volume, a sixfold increase in cellular insulin content, plasma C-peptide levels of 1eC2 ng/ml, and an ability to correct diabetes. They became structurally organized as homogenous clusters with their secretory vesicles polarized toward fenestrated capillaries. We concluded that the immature -cell phenotype provides grafts with a marked potential for -cell growth and differentiation and hence may have a potential role in curing diabetes. Cells with this phenotype can be isolated from prenatal organs; their presence in postnatal organs needs to be investigated.
Islet cell transplantation has long been considered as a potential cure for diabetes (1,2). A transplantation protocol has been reported that has achieved a consistent insulin independence that has been maintained beyond the first posttransplantation year (3). Clinical success has also accentuated obstacles to wide therapeutic implementation; a major drawback is the scarcity in islet grafts that can be isolated from human donor organs. Moreover, implanted -cell masses appear to be marginal and may therefore become inadequate with time. In rodent models, the size of adult islet implants decreases with time, probably as result of inadequate -cell renewal and/or neogenesis (4eC11). The capacity of fetal and neonatal pancreases to generate new -cells makes this tissue attractive for producing large numbers of -cells in vitro or for preparing grafts in which the -cell mass grows and matures (12eC16). Pancreatic cell implants from fetal or neonatal rodents or pigs have been found to increase their insulin content or -cell mass after transplantation and thus become able to normalize the diabetic state (12eC20). The mechanisms underlying this growth and functional maturation have not yet been clarified. They may involve -cell proliferation and/or -cell neoformation from duct cells or other cell types yet to be identified (13,15,19,21,22). The relative contribution of each pathway is still unknown and might vary with the cellular composition of the graft and the developmental stage of the donor organ. In all models where growth of the -cell mass has been demonstrated so far, the graft contained a majority of nonendocrine cells, some of which were seen as -cell precursors. In the present study, we implanted purified endocrine cells isolated from late-fetal pig pancreases and investigated the growth and functional maturation of -cells in normal and diabetic immunodeficient mice. Our data demonstrate that implants with immature -cells develop the ability to cure diabetes and relate this property to the proliferation and maturation of the -cells and their structural organization.
RESEARCH DESIGN AND METHODS
Prenatal porcine pancreatic endocrine cells were obtained from Beta Cell (Zellik, Belgium). They were prepared according to a procedure developed in our laboratory using fetuses of Belgian land race sows after 110eC115 days of gestation. Pancreases were dissociated by enzymatic digestion with collagenase P (Roche, Indianapolis, IN; 0.3 mg/ml) and trituration in calcium-free medium. Single endocrine cells were isolated by counterflow elutriation, purified by gradient centrifugation, and cultured in serum-free HAM’s F10 medium containing 6 mmol/l glucose, 5 mmol/l nicotinamide, 2 mmol/l glutamine, 50 e蘭ol/l isobutylmethylxanthine, 10eC6 mol/l hydrocortisone, and 0.5% BSA. At the start and after 1 week of culture, samples were taken to determine their cellular composition and hormone content (23,24). Grafts composed of >90% pure endocrine islet cells and containing 0.5eC1 million -cells were prepared from suspensions cultured for 1 week.
Transplantation of prenatal porcine pancreatic endocrine cells in nude mice.
Male immunodeficient nu/nu BALB/c ByJICo mice (age 8 weeks; Iffa Credo, France) were selected as recipients. Animals were maintained in acclimatized rooms with free access to water and pellet food. Transplants were carried out in nondiabetic and diabetic animals. Diabetes was induced by intravenous injection of alloxan (90 mg/kg; Sigma, Munich, Germany) the day before transplantation. Only animals with nonfasting glycemia >500 mg/dl were included in the study; because most of them died during the next 2 weeks irrespective of the presence of a fetal -cell implant, a "life-saving" rat islet graft was placed under the contralateral kidney capsule until the time that the fetal cell implants had grown and matured according to the studies in nondiabetic recipients; the islet-bearing kidney was then removed and the animals were followed metabolically. In the study of Korbutt et al. (15), such a "life-saving" rat islet graft was not necessary, possibly because a more differentiated neonatal pig islet cell preparation was used following pretreatment with high nicotinamide (10 mmol/l), which has been reported to favor -cell differentiation (25). Another reason could have been a lower degree of -cell depletion after alloxan in the Korbutt et al. study. The study protocols were approved by the Ethical Committee of Brussels Free University-VUB and carried out in accordance with Belgian and European Union regulations.
Mice were anesthetized with 2,2,2 tribromoethanol solution (Avertin; Sigma; 500 mg/kg body wt i.p.). Grafts were implanted within a blood clot that was placed under the kidney capsule (26); the use of a blood clot prevented a major cell loss after injection of the small and loosely formed cell aggregates. Control mice received a blood clot without cells. The function of the implant was monitored through plasma levels of glucose and porcine C-peptide. Tail vein blood was collected in EDTA/aprotinin (Trasylol) and plasma was stored at eC20°C until assayed for porcine C-peptide using a commercial radioimmunoassay kit (Linco, St. Charles, MO). Implants were removed at different time points after the transplantation and were processed for immunohistochemistry, electron microscopy, or the hormone assay. In one series of animals, bromodeoxyuridine (BrdU; 50 mg/kg) was injected intraperitoneally 1 h before the implant was removed and the percentage of proliferating cells was determined.
Analysis of cellular composition of grafts and implants.
Tissue samples were fixed in formalin and paraffin sections were incubated overnight with rabbit anti-synaptophysin (1:20; Dako, Glostrup, Denmark) (27), guinea pig anti-insulin (1:5,000) (27), rabbit anti-glucagon (1:5,000) (27), rabbit anti-somatostatin (1:5,000) (27), or rabbit anti-pancreatic polypeptide (1:50,000; gift of Dr. R.E. Chance, Lilly, Indianapolis, IN) (26). After the sections were washed in PBS (0.1 mmol, pH 7.4), they were incubated with peroxidase-labeled anti-mouse (1:20; Amersham, Amersham, U.K.), anti-rabbit (1:20; Amersham), or antieCguinea pig (1:100; Cappel, Aurora, OH) secondary antibody for 30 min at room temperature. The peroxidase reaction was developed by incubating sections in 0.3% H2O2 and 0.15% 3.3' diaminobenzidine tetrachloride (Sigma, St. Louis, MO). For the double-labeling immunohistochemistry, antieCguinea pig Ig labeled with fluorescein isothiocyanate and Cy3-labeled anti-mouse Ig secondary antibodies (Jackson Immunoresearch, West Groove, PA) were used. Proliferating cells were detected with a BrdU antibody (1:20; Cappel) and identified by an antibody against synaptophysin (28), insulin, glucagon, somatostatin, or pancreatic polypeptide. For each population positive for the endocrine marker, the percentage of BrdU-positive cells was determined and expressed as the proliferation index. For each implant, at least 2,000 cells were analyzed; no regional differences in structural organization were noticed. We also counted the percentage of insulin-positive cells with signs of apoptosis (i.e., condensed or fragmented nuclei after propidium iodide staining) (29).
Morphometry of implants.
The size of endocrine cell populations was determined according to Cavalieri’s principle. Total volumes of synaptophysin-, insulin-, and glucagon-positive cells were estimated by point counting (27). Grafts were cut in 5-e蘭-thick paraffin sections. Every 40th section was then immunostained and stereologically examined using a light microscope that projected the image onto a 600-point grid at a final magnification of 140x, with an area per point of 20,154 e蘭2. For each graft, 700eC2,000 points hitting the antibody-positive cells were counted. The total volume of antibody-positive cells was calculated by multiplying the total number of points overlaying the antibody-positive cells per section with the area per point and the distance between subsequent immune-stained sections (200 e蘭). To measure the mean volume of individual insulin cells, grafts were cut in 1-e蘭-thick sections, every 40th section was immunostained for insulin, and the total insulin cell volume was estimated as described above. The total insulin cell number was determined by the physical dissector principle (30,31) in randomly sampled and insulin-stained pairs of consecutive sections, including a reference (every 40th) and look-up (the consecutive) section. The mean volume of insulin cells was then calculated by dividing the total insulin cell volume by the total insulin cell number.
Statistical analysis.
Data are expressed as means ± SD of n independent experiments. The statistical significance of differences in plasma porcine C-peptide and glycemia levels between experimental and control groups was calculated by the Kruskal-Wallis test. Changes in cellular composition, total volume, mean insulin cell volume and number, graft insulin content, proliferation, and apoptosis index were tested by the Mann-Whitney U test. P <0.05 was considered significant.
RESULTS
Characteristics of fetal porcine -cell graft.
Fetal porcine islet cell grafts were composed of >90% endocrine cells, which mainly corresponded to insulin- and glucagon-positive cells in similar proportions (each representing 39 ± 5% of all cells); somatostatin-positive (7 ± 2%) and pancreatic polypeptideeCpositive (6 ± 3%) cells were much less abundant (Fig. 1). The remaining cells (<10%) corresponded to cytokeratin (CK-7)-positive cells (5.1 ± 2.3% of all cells) and damaged cells. We observed that >90% of the cells were positive for synaptophysin, indicating that this neuroendocrine marker can be used to identify fetal pancreatic endocrine cells. The CK-7 positive cells were copurified epithelial duct cells (32). Grafts contained 0.8 ± 0.1 x 106 -cells and 4.6 ± 1.6 e蘥 insulin; the insulin content of the -cells averaged 5.2 e蘥/106 cells. -Cells had an average cellular volume of 800 e蘭3, which is comparable with that of neonatal rat -cells (33) and smaller than that of adult rodent -cells (30,33). Electron microscopy illustrated endocrine cell aggregates with intact ultrastructure (<5% dead cells). Few apoptotic -cells were detected (0.5 ± 0.2%). Under basal serum-free culture conditions, 0.5 ± 0.3% of insulin cells were BrdU positive after a 1-h incubation with BrdU.
Change in cellular composition of fetal porcine -cell implants.
At posttransplant day 1, the composition of the implant (Fig. 2) was similar to that of the isolated graft, with equal percentages of insulin- (39 ± 5%) and glucagon- (39 ± 5%) positive cells, each contributing similar total cell volumes to the endocrine volume of the implant. During posttransplant week 1, no change occurred in the volume of the insulin cell population. However, that of glucagon cells decreased by 40% (Table 1) in association with a decrease in the percentage of glucagon cells and an increase in the percentage of insulin cells (Fig. 2). From posttransplant week 1 onward, the total volume of insulin cells increased with time (threefold between posttransplant weeks 1 and 10, twofold between posttransplant weeks 10 and 20) (Table 1); this was not the case for the total volume of glucagon cells, which remained stable (Table 1). The proportion of insulin cells further increased to 89% at posttransplant week 20, whereas that of glucagon cells decreased to 7% (Fig. 2).
The increase in the total volume of insulin cells was caused by an increase in the mean individual cell volume as well as an increase in the number of -cells. Between posttransplant weeks 1 and 20, the mean individual volume of insulin cells almost doubled (to an average volume of 1,500 e蘭3) whereas the number of insulin cells increased fourfold (Fig. 3). No significant difference was detected between the number of -cells on posttransplant day 1 and at posttransplant week 1 (Fig. 3). The increase in the total volume of insulin cells between posttransplant weeks 1 and 20 was accompanied by an increase in insulin content of the implant rising to 17- and 50-fold higher values, respectively, at posttransplant weeks 10 and 20 (Fig. 3). This increase in insulin content of the implant was much more pronounced than the increase in -cell number or the total volume of insulin cells, which indicated that the -cells increased their individual insulin store. At posttransplant week 20, the insulin content of the -cells averaged 31 e蘥/106 cells, which was sixfold higher than at the start.
Structural organization of fetal porcine -cell implants.
As well as changing their cellular composition, implants underwent a structural organization. At posttransplant day 1, they were composed of dispersed single cells, small cell aggregates, and blood cells without structural organization (Fig. 1). At posttransplant week 20, the implants appeared as a white compact mass on the kidney surface, with a diameter of 3eC8 mm and clearly visible interconnected capillaries (Fig. 4A). When viewed microscopically, they presented a homogenous assembly of endocrine cell clusters (Fig. 4B) that were predominantly formed by insulin-positive cells with a few other endocrine islet cells scattered at their border (Fig. 4C and D). The space in the center of some endocrine clusters (Fig. 4C) was filled with erythrocytes without endothelial cell lining (Fig. 4E). These lacunae do not exhibit the tubular or matrix pattern that is characteristic for vasculogenic mimicry; their formation and functional significance is so far unknown. Within the endocrine cell masses, capillaries with fenestrated endothelial cells were detected, toward which the endocrine cells were oriented with their pole of secretory vesicles (Fig. 4E and F, arrows).
Rates of -cell proliferation and apoptosis in fetal porcine -cell implants.
During posttransplant week 1, 1eC6% of insulin cells were positive for BrdU 1 h after this compound was injected (Fig. 4G). This proliferation index was already elevated 4 h after transplant (2.6 ± 0.7%) and reached peak values of 5 ± 0.7% after 72 h (Fig. 5). Between posttransplant week 1 and 10, the percentage of BrdU-positive insulin cells remained significantly higher than in the prenatal pancreas in situ (0.2%) but then slowly decreased with time to 0.3% at posttransplant week 20. The percentage of BrdU-positive glucagon cells was lower during the first week (2.3% after 72 h) and returned to basal levels from posttransplant week 1 and beyond (0.3% BrdU positive cells).
The first 72 h after the transplant were also characterized by a higher rate of insulin cell apoptosis (1eC4%) (Fig. 4H), which was markedly higher than in the graft before transplantation (0.5 ± 0.2%). This index decreased to <0.2% at posttransplant week 1 or later (Fig. 5).
Metabolic effects of fetal porcine -cell implants.
Porcine C-peptide was not detectable in plasma during the first 2 posttransplant weeks but was consistently measured from posttransplant week 5 and beyond, increasing to plateau levels of 1.0eC1.5 ng/ml at posttransplant weeks 10eC20 (Fig. 6). Blood glucose levels were not different from those in nontransplanted controls during the first 5 posttransplant weeks (91 ± 9 vs. 94 ± 8 mg/dl) but then progressively decreased to reach a level at and beyond posttransplant week 15 of 59 ± 27 mg/dl, which was significantly lower than the level in normal controls (110 ± 10 mg/dl; P < 0.05) (Fig. 6). This metabolic effect was associated with a 50% decrease in pancreatic insulin content (25 ± 6 vs. 50 ± 7 e蘥 in sham-operated normal controls; P < 0.05).
When fetal porcine islet cell preparations of similar size and composition were implanted in alloxan-induced diabetic nude mice, they did not immediately correct the hyperglycemic state and the mice died during the first 2 weeks after the transplant. This outcome indicated the metabolic incapacity of the fetal porcine -cell implant during the first posttransplant weeks and correlated with undetectable circulating and low tissue levels of porcine C-peptide or insulin during this period. We therefore examined the metabolic capacity of porcine -cell implants at a later time when they had undergone the above-described changes in composition and organization and had reached plateau levels of circulating C-peptide (i.e., beyond posttransplant week 10). To this end, the fetal pig -cell preparations were transplanted in diabetic mice that simultaneously received isolated adult rat islets under the contralateral kidney capsule. The rat islet implant established normoglycemia (Fig. 7) and thus prevented the death of these animals. The rise in circulating porcine C-peptide levels was similar to that in noneCalloxan-administered recipients that had received a fetal pig -cell graft (Fig. 7). Removal of the rat islet implant at posttransplant week 15 did not alter porcine C-peptide levels nor did it reverse the normoglycemic state, indicating that the fetal porcine -cell implant contained a metabolically responsive and adequate -cell mass. When the animals were killed at posttransplant week 17, their pancreatic insulin content was in the diabetic range (1.9 ± 1.2 e蘥 [n = 6] vs. 34 ± 11 e蘥 in noneCalloxan-administered recipients of fetal porcine -cell grafts) whereas the insulin content of their implant (60 ± 29 e蘥) was similar to that of the noneCalloxan-administered recipients (87 ± 29 e蘥 [n = 7]; P > 0.05).
DISCUSSION
The shortage in human donor organs has raised interests in pigs as a potential source of islet cell grafts for diabetic patients. Techniques have been developed for preparing insulin-producing tissue from fetal, neonatal, and adult pig pancreases (14,15,20,34). Grafts from all donor ages have been shown to normalize diabetes in immunodeficient mice (15,22,35eC37). The yield and purity of -cells is markedly lower in isolates from fetal and neonatal organs, but their implants develop a growing -cell mass, which is seen as an advantage. Islet cell grafts are indeed expected to achieve long-term metabolic control, which, in the normal endocrine pancreas, depends on the ability for -cell renewal and growth to compensate for a natural -cell loss and increasing metabolic demands (38eC41).
There is so far no evidence for adaptive growth or -cell renewal in implants from adult pig, rodent, or human organs; instead, a progressive decline in -cell mass has been observed with time and is considered to be responsible for the loss of metabolic control in long-term implants (9,10,42). On the other hand, implants of fetal pancreatic tissue grow in total volume, -cell mass, and insulin content, bringing their overall function with time to a state that corrects hyperglycemia (15,18,19,22). The rapidity of metabolic normalization may vary with the amount of tissue and its cellular composition and also with the stage of development and the species. For porcine islet cell clusters isolated at gestation days 50eC70 and implanted under the kidney capsule of nude mice, the correction of diabetes is achieved after only 16eC20 weeks; this period can be shortened by the administration of nicotinamide, which is considered to stimulate -cell differentiation (25,43). This was also the case for porcine neonatal islet cell clusters (15). The implants undergo a maturation process that seems to involve both neoformation and functional maturation of -cells (19,22,43). The relative contribution of both components is not yet clear.
It is also unknown to which extent new -cells are formed from nonendocrine cells or other -cells (22). Most authors underline the importance of a differentiation of nonendocrine cells to -cells (19,20,22,44). At the time of implantation, these grafts are indeed mainly composed of nonendocrine cells; the progressive disappearance of the nonendocrine cells is found to be compatible with a transition to -cells. The nonendocrine-to-endocrine differentiation route is also supported by the observation that several -cells share markers with adjacent nonendocrine cells (22). Moreover, little evidence has been provided for a major contribution of -cell proliferation. A study comparing the outcome of fetal islet cell clusters with a low percentage of -cells and fetal -cell aggregates even concluded that the absence of nonendocrine cells markedly reduced the normalizing capacity of the grafts (44).
To investigate more directly the growth potential of fetal -cells, we undertook a morphometric study on grafts of purified endocrine islet cells that were virtually devoid of nonendocrine cells. Purified endocrine islet cell grafts were prepared from late fetal (110eC115 days gestation) pig pancreases. They consisted of >90% endocrine cells, with similar percentages (40%) of insulin- and glucagon-positive cells. This composition markedly differed from that of fetal or neonatal porcine islet cell clusters prepared after 50eC70 days of gestation (14) or 1eC3 days after birth (15,20), which have 5 or 24% -cells and 1 or 8% -cells, respectively. The -cells in our preparation were immature, as determined by comparing them with adult -cells; their cellular volume was twofold smaller (30,33) and their insulin content (5 e蘥/million -cells) was sixfold lower. They were also unable to correct diabetes in nude mice when transplanted in numbers (20 million -cells/kg body wt) that were 5- to 10-fold higher than those of adult -cells that normalize rodent or human diabetes (4,5,7,45,46). The porcine -cells developed a functionally mature state over 10eC20 weeks after transplantation, as evidenced by their progressive increase in size and insulin content, the appearance of circulating porcine C-peptide, and their progressive gain of metabolic control.
In addition to presenting signs of maturing -cells, the endocrine cell implants also exhibited a continuous formation of new -cells. The number of -cells increased 4.3-fold between the end of posttransplant week 1 and posttransplant week 20. This increase could not be attributed to nonendocrine cells as these were virtually absent (<1% at the end of posttransplant week 1). Instead, it was due to the high proliferative activity of the -cells: at posttransplant week 1, the percentage of BrdU-positive -cells was 12-fold higher than in the prenatal pancreas in situ and remained elevated, although at a lower level, until posttransplant week 10. This index was also sixfold higher than in the same preparation tested in vitro in the absence of serum. Factors released during the inflammatory and tissue repair process probably stimulated the -cells that are prone to proliferation. Only 1 day after the implantation injury, the proliferating fibroblasts and endothelial cells were noticed and 2 days later, granulation tissue with small blood vessels was observed. Despite this increased rate of -cell proliferation during posttransplant week 1, neither the number nor total volume of -cells increased. This occurrence might have reflected a simultaneous equal loss of -cells that was compatible with the elevated percentages of apoptotic -cells during this period. Because apoptotic cells are rapidly cleared, it is not possible to calculate the cell loss from the percentages of apoptotic -cells. Apoptotic -cells were found in similar percentages, an observation that is consistent with nonspecific damage in the implant site. However, there were less BrdU-positive glucagon cells than insulin cells (2 vs. 5% at day 3), which explains the reduction in total volume of this cell population at posttransplant week 1. The rates of BrdU-positive glucagon cells remained low during the entire study period in contrast to our observations of -cells. The quantitative differences between both cell populations thus further increased with time, resulting at posttransplant week 20 in an implant with 89% -cells and only 7% -cells. An abundance of - over -cells has been previously described for grafts prepared from fetal and neonatal pancreatic tissue as well as for adult islet grafts (15,19,46,47).
The observed growth in -cell number and individual cell volume did occur under normoglycemic conditions, as maintained by the mouse endocrine pancreas or a rat islet graft. Under this condition, the implant induced, from posttransplant week 10 on, plasma porcine C-peptide levels of 1eC2 ng/ml. This level was maintained when the rat islet graft was removed; it corresponded to a -cell mass that had grown and matured to a higher insulin content than in the normal mouse pancreas. At this level of growth and differentiation, the porcine -cell implant prevented the development of diabetes, with no decline in its insulin reserves. Morphological examination revealed that the metabolically adequate porcine fetal -cell implant was structurally organized in contrast to the randomly distributed small cell clusters in the initial implant. It consisted of homogenous -cell clusters with peripherally scattered -cells surrounded by fenestrated capillaries, as described in normal islet tissue (48). The -cells presented intracellular signs of polarity, with their pole of secretory vesicles located at the pole that faced the capillaries. Between the clusters, blood-filled lacunae devoid of an endothelial lining were also noticed. The revascularization process proceeded throughout the 20-week study period, as evidenced by an ongoing capillary sprouting.
The present study showed that endocrine islet cells can be purified from porcine fetal pancreases, opening the potential for large scale isolations of -cells. Our preparations contained 40% -cells with signs of immaturity. When transplanted under the kidney capsule of normoglycemic mice, the -cells exhibited a prolonged proliferating activity, first compensating for -cell loss during engraftment and then increasing the -cell number fourfold over the subsequent 20 weeks. These data indicate that immature -cells can achieve a sustained renewal and growth of the -cell mass. It is conceivable that immature -cells were also responsible for the growth of the -cell mass that has been observed in human duct cell preparations isolated from young human donors (27). Our observations are consistent with recent work showing that the proliferation of -cells is the main mechanism for growth of the -cell mass in the postnatal mouse pancreas (49,50). They also illustrate that impure islet cell preparations are not needed to achieve -cell renewal or adaptive growth in implants of pancreatic tissue.
ACKNOWLEDGMENTS
This study was supported by grants from the Juvenile Diabetes Research Foundation (4eC2001-434), the Belgian Fund for Scientific Research (G.0375.00, G.0358.03), the Institute for the Promotion of Innovation by Science and Technology in Flanders (020442), and the Interuniversity Poles of Attraction Programme (IUAP P5/17) from the Belgian Science Policy.
We thank the team of Beta Cell for providing the fetal porcine endocrine cell preparations and Nicole Buelens and Violeta Ardelean for excellent technical assistance.
REFERENCES
Lacy PE, Scharp DW: Islet transplantation in treating diabetes. Annu Rev Med 37:33eC40, 1986
Robertson RP: Islet transplantation as a treatment for diabetes: a work in progress. N Engl J Med 350:694eC705, 2004
Shapiro AM, Lakey JR, Ryan EA, Korbutt GS, Toth E, Warnock GL, Kneteman NM, Rajotte RV: Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med 343:230eC238, 2000
Pipeleers-Marichal MA, Pipeleers DG, Cutler J, Lacy PE, Kipnis DM: Metabolic and morphologic studies in intraportal-isleteCtransplanted rats. Diabetes 25:1041eC1051, 1976
Orloff MJ, Macedo A, Greenleaf GE, Girard B: Comparison of the metabolic control of diabetes achieved by whole pancreas transplantation and pancreatic islet transplantation in rats. Transplantation 45:307eC312, 1988
Hiller WF, Klempnauer J, Luck R, Steiniger B: Progressive deterioration of endocrine function after intraportal but not kidney subcapsular rat islet transplantation. Diabetes 40:134eC140, 1991
Pipeleers DG, Pipeleers-Marichal M, Vanbrabandt B, Duys S: Transplantation of purified islet cells in diabetic rats. II. Immunogenicity of allografted islet beta-cells. Diabetes 40:920eC930, 1991
Keymeulen B, Teng H, Vetri M, Gorus F, In’t Veld P, Pipeleers DG: Effect of donor islet mass on metabolic normalization in streptozotocin-diabetic rats. Diabetologia 35:719eC724, 1992
Montanya E, Bonner-Weir S, Weir GC: Beta cell mass and growth after syngeneic islet cell transplantation in normal and streptozocin diabetic C57BL/6 mice. J Clin Invest 91:780eC787, 1993
Davalli AM, Ogawa Y, Ricordi C, Scharp DW, Bonner-Weir S, Weir GC: A selective decrease in the beta cell mass of human islets transplanted into diabetic nude mice. Transplantation 59:817eC820, 1995
Biarnes M, Montolio M, Nacher V, Raurell M, Soler J, Montanya E: -Cell death and mass in syngeneically transplanted islets exposed to short- and long-term hyperglycemia. Diabetes 51:66eC72, 2002
Lazarow A, Wells LJ, Carpenter AM, Hegre OD, Leonard RJ, McEvoy RC: Banting Lecture 1973: Islet differentiation, organ culture, and transplantation. Diabetes 22:877eC912, 1973
Sandler S, Andersson A, Schnell A, Mellgren A, Tollemar J, Borg H, Petersson B, Groth CG, Hellerstrom C: Tissue culture of human fetal pancreas: development and function of B-cells in vitro and transplantation of explants to nude mice. Diabetes 34:1113eC1119, 1985
Korsgren O, Sandler S, Landstrom AS, Jansson L, Andersson A: Large-scale production of fetal porcine pancreatic isletlike cell clusters: an experimental tool for studies of islet cell differentiation and xenotransplantation. Transplantation 45:509eC514, 1988
Korbutt GS, Elliott JF, Ao Z, Smith DK, Warnock GL, Rajotte RV: Large scale isolation, growth, and function of porcine neonatal islet cells. J Clin Invest 97:2119eC2129, 1996
Castaing M, Peault B, Basmaciogullari A, Casal I, Czernichow P, Scharfmann R: Blood glucose normalization upon transplantation of human embryonic pancreas into beta-cell-deficient SCID mice. Diabetologia 44:2066eC2076, 2001
Hegre OD, Leonard RJ, Rusin JD, Lazarow A: Transplantation of the fetal rat pancreas: quantitative morphological analysis of islet tissue growth. Anat Rec 185:209eC221, 1976
Brown J, Heininger D, Kuret J, Mullen Y: Islet cells grow after transplantation of fetal pancreas and control of diabetes. Diabetes 30:9eC13, 1981
Korsgren O, Jansson L, Eizirik D, Andersson A: Functional and morphological differentiation of fetal porcine islet-like cell clusters after transplantation into nude mice. Diabetologia 34:379eC386, 1991
Yoon KH, Quickel RR, Tatarkiewicz K, Ulrich TR, Hollister-Lock J, Trivedi N, Bonner-Weir S, Weir GC: Differentiation and expansion of beta cell mass in porcine neonatal pancreatic cell clusters transplanted into nude mice. Cell Transplant 8:673eC689, 1999
Leonard RJ, Lazarow A, Hegre OD: Pancreatic islet transplantation in the rat. Diabetes 22:413eC428, 1973
Trivedi N, Hollister-Lock J, Lopez-Avalos MD, O’Neil JJ, Keegan M, Bonner-Weir S, Weir GC: Increase in beta-cell mass in transplanted porcine neonatal pancreatic cell clusters is due to proliferation of beta-cells and differentiation of duct cells. Endocrinology 142:2115eC2122, 2001
Pipeleers DG, in’t Veld PA, Van de Winkel M, Maes E, Schuit FC, Gepts W: A new in vitro model for the study of pancreatic A and B cells. Endocrinology 117:806eC816, 1985
Ling Z, Pipeleers DG: Prolonged exposure of human beta cells to elevated glucose levels results in sustained cellular activation leading to a loss of glucose regulation. J Clin Invest 98:2805eC2812, 1996
Korsgren O, Sandler S, Jansson L, Groth CG, Hellerstrom C, Andersson A: Effects of culture conditions on formation and hormone content of fetal porcine isletlike cell clusters. Diabetes 38 (Suppl. 1):209eC212, 1989
Gray DW, McShane P, Morris PJ: The effect of hyperglycemia on isolated rodent islets transplanted to the kidney capsule site. Transplantation 41:699eC703, 1986
Bogdani M, Lefebvre V, Buelens N, Bock T, Pipeleers-Marichal M, In’t Veld P, Pipeleers D: Formation of insulin-positive cells in implants of human pancreatic duct cell preparations from young donors. Diabetologia 46:830eC838, 2003
Wiedenmann B, Franke WW, Kuhn C, Moll R, Gould VE: Synaptophysin: a marker protein for neuroendocrine cells and neoplasms. Proc Natl Acad Sci U S A 83:3500eC3504, 1986
Hoorens A, Van de Casteele M, Kloppel G, Pipeleers D: Glucose promotes survival of rat pancreatic beta cells by activating synthesis of proteins which suppress a constitutive apoptotic program. J Clin Invest 98:1568eC1574, 1996
Bock T, Svenstrup K, Pakkenberg B, Buschard K: Unbiased estimation of total beta-cell number and mean beta-cell volume in rodent pancreas. APMIS 107:791eC799, 1999
Gundersen HJ: Stereology of arbitrary particles: a review of unbiased number and size estimators and the presentation of some new ones, in memory of William R. Thompson. J Microsc 143:3eC45, 1986
Kasper M, von Dorsche H, Stosiek P: Changes in the distribution of intermediate filament proteins and collagen IV in fetal and adult human pancreas. I. Localization of cytokeratin polypeptides. Histochemistry 96:271eC277, 1991
Svenstrup K, Skau M, Pakkenberg B, Buschard K, Bock T: Postnatal development of beta-cells in rats: proposed explanatory model. APMIS 110:372eC378, 2002
Toso C, Brandhorst D, Oberholzer J, Triponez F, Buhler L, Morel P: Isolation of adult porcine islets of Langerhans. Cell Transplant 9:297eC305, 2000
Korsgren O, Jansson L: Porcine islet-like cell clusters cure diabetic nude rats when transplanted under the kidney capsule, but not when implanted into the liver or spleen. Cell Transplant 3:49eC54, 1994
Vo L, Tuch BE, Wright DC, Keogh GW, Roberts S, Simpson AM, Yao M, Tabiin MT, Valencia SK, Scott H: Lowering of blood glucose to nondiabetic levels in a hyperglycemic pig by allografting of fetal pig isletlike cell clusters. Transplantation 71:1671eC1677, 2001
Davalli AM, Ogawa Y, Scaglia L, Wu YJ, Hollister J, Bonner-Weir S, Weir GC: Function, mass, and replication of porcine and rat islets transplanted into diabetic nude mice. Diabetes 44:104eC111, 1995
Bonner-Weir S: Regulation of pancreatic beta-cell mass in vivo. Recent Prog Horm Res 49:91eC104, 1994
Sorenson RL, Brelje TC: Adaptation of islets of Langerhans to pregnancy: beta-cell growth, enhanced insulin secretion and the role of lactogenic hormones. Horm Metab Res 29:301eC307, 1997
Bonner-Weir S: Perspective: postnatal pancreatic beta cell growth. Endocrinology 141:1926eC1929, 2000
Withers DJ, White M: Perspective: the insulin signaling system: a common link in the pathogenesis of type 2 diabetes. Endocrinology 141:1917eC1921, 2000
Keymeulen B, Korbutt G, De Paepe M, Gorus F, Kloppel G, Pipeleers DG: Long-term metabolic control by rat islet grafts depends on the composition of the implant. Diabetes 45:1814eC1821, 1996
Otonkoski T, Ustinov J, Rasilainen S, Kallio E, Korsgren O, Hayry P: Differentiation and maturation of porcine fetal islet cells in vitro and after transplantation. Transplantation 68:1674eC1683, 1999
Tu J, Khoury P, Williams L, Tuch BE: Comparison of fetal porcine aggregates of purified beta-cells versus islet-like cell clusters as a treatment of diabetes. Cell Transplant 13:525eC534, 2004
Pipeleers DG, Pipeleers-Marichal M, Hannaert JC, Berghmans M, In’t Veld PA, Rozing J, Van de Winkel M, Gepts W: Transplantation of purified islet cells in diabetic rats. I. Standardization of islet cell grafts. Diabetes 40:908eC919, 1991
Keymeulen B, Ling Z, Gorus FK, Delvaux G, Bouwens L, Grupping A, Hendrieckx C, Pipeleers-Marichal M, Van Schravendijk C, Salmela K, Pipeleers DG: Implantation of standardized beta-cell grafts in a liver segment of IDDM patients: graft and recipient characteristics in two cases of insulin-independence under maintenance immunosuppression for prior kidney graft. Diabetologia 41:452eC459, 1998
Street CN, Lakey JR, Shapiro AM, Imes S, Rajotte RV, Ryan EA, Lyon JG, Kin T, Avila J, Tsujimura T, Korbutt GS: Islet graft assessment in the Edmonton Protocol: implications for predicting long-term clinical outcome. Diabetes 53:3107eC3114, 2004
Bearer EL, Orci L: Endothelial fenestral diaphragms: a quick-freeze, deep-etch study. J Cell Biol 100:418eC428, 1985
Bock T, Pakkenberg B, Buschard K: Increased islet volume but unchanged islet number in ob/ob mice. Diabetes 52:1716eC1722, 2003
Dor Y, Brown J, Martinez OI, Melton DA: Adult pancreatic beta-cells are formed by self-duplication rather than stem-cell differentiation. Nature 429:41eC46, 2004(Marika Bogdani, Krista Su)