Kinetics of insulin secretion from MIN6 pseudoislets after encapsulation in a prototype device of a bioartificial pancreas.

Xenotransplantation of insulin-secreting cells from nonhuman sources is an alternative therapeutic approach to bypass the shortage of human pancreatic islet tissue for transplantation in order to treat insulin deficiency in type 1 diabetes mellitus. Therefore, we studied the suitability of pseudoislets generated from insulin-secreting MIN6 tissue culture cells to serve as a surrogate for replacement of pancreatic islets after encapsulation in a minicell, representing a prototype of a new bioartificial pancreas device. MIN6 pseudoislets showed an excellent insulin secretory responsiveness with a typical biphasic secretory pattern to glucose stimulation. When encapsulated in the minicell, insulin release from the pseudoislets in response to glucose stimulation was reduced. The initial first phase insulin secretory response was greatly attenuated. In contrast, the first phase insulin secretory response of the encapsulated pseudoislets was restored on stimulation with the sulfonylurea drug tolbutamide. Our results indicate that the reason for the attenuated first phase of release is the restricted permeability of the pores in the separating membrane in the minicell for the hydrophilic glucose molecule rather than a limited permeability for the secretion product insulin. The reduced release of insulin from the encapsulated pseudoislets could be compensated by overexpression of glucokinase in MIN6 cells, which resulted in an increased glucose responsiveness of the pseudoislets for stimulation with glucose. Thus, this minicell is a well-suited miniature test system for the evaluation of the feasibility of encapsulation of insulin-secreting cells and allows the testing of permeability properties of separating membranes in bioartificial pancreas devices.

Perfluorocarbons: new tool for islets preservation in vitro.

Pancreatic islet transplantations to treat type 1 diabetes often fail to function because of hypoxia. Perfluorocarbons (PFCs) exhibit a high oxygen solubility coefficient and maintain high oxygen partial pressure for extended times. They also serve as oxygen "reservoirs" for harvested organs in pancreas organ transplantation. Previous studies have shown the PFCs display antiadhesive effects on beta cells. The aim of this study was to evaluate the effects of PFC on islet viability and functionality and on extracellular matrix (ECM) disruption of islets via inhibition of adhesion. Primary cultures of rat islets were incubated for 24 hours in the presence or absence of 3.5% (weight/volume) PFCs in culture media. We studied viability (FDA/PI), stimulation index linked to insulin secretion (ELISA), and expression of insulin and laminin messenger RNAs (mRNAs). Immunostaining was performed on insulin and laminin. Islet viability was similar in the presence or absence of PFCs (about 80%). Stimulation index showed preservation of islet functionality in the presence of PFC (4.9 +/- 0.7) as compared with controls (2.8 +/- 0.5). Moreover, laminin mRNA expression was lower compared with controls (55% of PFC incubated vs control islets). Immunohistochemistry studies showed preservation of ECM inside the islets in the presence of PFCs versus controls at 24 hours after islet isolation. In conclusion, PFCs preserved islet viability and functionality and prevented ECM disruption. PFCs may represent a new tool for islet preservation in vitro.

Overexpression of vascular endothelial growth factor in vitro using deferoxamine: a new drug to increase islet vascularization during transplantation.

During pancreatic islet transplantation, delayed and insufficient revascularization can deprive islets of oxygen and nutrients, resulting in cell death and early graft failure. Deferoxamine (DFO), an iron chelator, increases vascular endothelial growth factor (VEGF) expression in cells. The aim of this work was to study the effect of DFO on beta-cell and pancreatic islet viability as well as VEGF expression. beta-cell lines from rat insulinoma (Rin m5f) and primary cultures of pancreatic islets from Wistar rats were incubated with DFO (10, 100, and 1000 micromol/L). The viability was evaluated using fluorescein diacetate/propidium iodide for dying pancreatic islets and using cell titers for Rin m5f. Expression of VEGF messenger RNA (mRNA) was quantified using reverse transcriptase polymerase chain reaction (RT-PCR). Finally, VEGF secretion was determined using enzyme-linked immunosorbent assays at 1 to 3 days after treatment. The addition of 10 micromol/L of DFO preserved Rin m5F viability at 24 hours after treatment (10 micromol/L; 101.33% +/- 5.66%; n = 7). However, 100 and 1000 micromol/L of DFO induced cell death (68.92% +/- 5.83% and 65.89% +/- 5.83%, respectively; n = 4). In the same way, viability of pancreatic islets in the presence of DFO was preserved. RT-PCR analysis showed stimulation of VEGF mRNA in the presence of 10 micromol/L of DFO in islets at 3 days after culture. Finally, 10 micromol/L of DFO stimulated secretion of VEGF 7.95 +/- 0.84 versus 1.80 +/- 1.10 pg/microg total protein with 10 micromol/L of DFO in rat islets at 3 days after culture, n = 3; P < .001). The use of DFO to stimulate VEGF expression and increase islet vascularization may be a realistic approach to improve islet viability during transplantation.

Role of chemokine signaling pathways in pancreatic islet rejection during allo- and xenotransplantation.

During transplantation, pancreatic islets release chemokines promoting macrophage attraction, hampering engraftment of islets. The aim of this work was to examine the mechanism of macrophage-pancreatic islets interaction that mediates islet rejection during transplantation. Human macrophages exposed to supernates of human and porcine pancreatic islets for the allogeneic and xenogeneic models, respectively, were evaluated for chemotaxis and expression of chemokine receptors (CCR-5). To modulate migration and identify the signaling pathway of macrophages, we tested pertussis toxin (PTX) to block Gi protein, and staurosporin and wortmannin to inhibit the protein kinase, and phosphoinositol-3 kinase, respectively. The addition of these agents significantly reduced macrophage migration induced by human islet supernates from 3.2 +/- 0.5 to 1.5 +/- 0.2, 0.9 +/- 0.1, and 1 +/- 0.1, respectively (P < .001, n = 3). In a xenotransplantation model, the reduction was less decreased, from 4.1 +/- 0.4 to 2.7 +/- 0.3 (P < .01), to 2.5 +/- 0.3 (P < .01), or to 1 +/- 0.1 (P < .001). Western blot analysis of chemokine receptor expression showed increased CCR-5 expression with human pancreatic islet supernates. Moreover, decreased islet purity increased CCR-5 expression. Pharmacologic study showed that PTX induced an increase in CCR-5 expression in allogeneic transplantation, whereas only staurosporin induced an increased receptor expression in the xenogeneic model, suggesting that chemokines participate in islet rejection even though the chemokine signaling pathways differ between allo- and xenotransplantation. Understanding the molecular mechanisms of islet rejection may improve graft survival.

Vitamin D-binding protein gene polymorphism association with IA-2 autoantibodies in type 1 diabetes.

BACKGROUND: Vitamin D-binding protein (DBP) is the main systemic transporter of 1.25(OH)2D3 and is essential for its cellular endocytosis. There are two known polymorphisms in exon 11 of the DBP gene resulting in amino acid variants: GAT-->GAG substitution replaces aspartic acid by glutamic acid in codon 416; and ACG-->AAG substitution in codon 420 leads to an exchange of threonine for lysine. These DBP variants lead to differences in the affinity for 1.25(OH)2D3. Correlations between DBP alleles and type 1 diabetes have been described in different populations. Therefore, we investigated the polymorphism in codon 416 of the DBP gene for an association with autoimmune markers of type 1 diabetes. DESIGN AND METHODS: The present analysis was a case control study. 110 patients, 68 controls, and 115 first-degree relatives were genotyped for the DBP polymorphism in codon 416. DNA typing of DBP locus was performed by the PCR-restriction fragment length polymorphism method (RFLP). RESULTS: The frequencies of the Asp/Glu and Glu/Glu were significantly increased in diabetic subjects with detectable IA-2 antibodies (P < 0.01). On the contrary, the DBP Glu-containing genotype was not accompanied by differences in the prevalence of GAD65 antibodies. These finding supports a role of the vitamin D endocrine system in the autoimmune process of type 1 diabetes.

Entrapment of dispersed pancreatic islet cells in CultiSpher-S macroporous gelatin microcarriers: Preparation, in vitro characterization, and microencapsulation.

Immunoprotection of pancreatic islets for successful allo- or xenotransplantation without chronic immunosuppression is an attractive, but still elusive, approach for curing type 1 diabetes. It was recently shown that, even in the absence of fibrotic overgrowth, other factors, mainly insufficient nutrition to the core of the islets, represent a major barrier for long-term survival of intraperitoneal microencapsulated islet grafts. The use of dispersed cells might contribute to solve this problem due to the conceivably easier nutritional support to the cells. In the present study, purified bovine islets, prepared by collagenase digestion and density gradient purification, and dispersed bovine islet cells, obtained by trypsin and DNAsi (viability > 90%), were entrapped into either 2% (w/v) sodium alginate (commonly used for encapsulation purposes) or (dispersed islet cells only) macroporous gelatin microcarriers (CulthiSpher-S, commonly used for the production of biologicals by animal cells). Insulin release studies in response to glucose were performed within 1 week and after 1 month from preparation of the varying systems and showed no capability of dispersed bovine islet cells within sodium alginate microcapsules to sense glucose concentration changes. On the contrary, bovine islet cells entrapped in CulthiSpher-S microcarriers showed maintained capacity of increasing insulin secretion upon enhanced glucose concentration challenge. In this case, insulin release was approximately 60% of that from intact bovine islets within sodium alginate microcapsules. MTT and hematoxylineosin staining of islet cell-containing microcarriers showed the presence of viable and metabolically active cells throughout the study period. This encouraging functional data prompted us to test whether the microcarriers could be immunoisolated for potential use in transplantation. The microcarriers were embedded within 3% sodium alginate, which was then covered with a poly-L-lysine layer and a final outer alginate layer. Maintained insulin secretion function of this system was observed, which raises the possibility of using microencapsulated CulthiSpher-S microcarriers, containing dispersed pancreatic islet cells, in experimental transplantation studies.

The HLA-DQB alleles and amino acid variants of the vitamin D-binding protein in diabetic patients in Alsace.

BACKGROUND: The HLA-DQB1 chain, in particular the amino acid in position 57, and genetic variants of the vitamin D-binding protein (DBP) have been reported to be associated with type 1 diabetes. There are two known polymorphisms in exon 11 of the DBP gene resulting in amino acid variants: codons 416 GAT --> GAG (Asp --> Glu) and 420 ACG --> AAG (Thr --> Lys). We compared distribution of DQB1 alleles and amino acid variants of DBP in type 1 diabetic patients (n = 44) in the Alsacian population and in healthy controls (n = 58). METHODS: The second exon of the DQB1 gene and exon 11 of DBP were analyzed by restriction mapping after polymerase chain reaction. RESULTS: A significant enrichment in DQB1 alleles encoding for an amino acid different from Asp in position 57 (NA) was observed in diabetic subjects as compared to controls (94.3 vs. 32.8%; p < 0.001). Combinations other than Ala/Ala carried the highest relative risk (OR = 52; p < 0.001). The analysis of the polymorphism in exon 11 of DBP showed a significant difference in the allele frequency of the HaeIII site, but not of the StyI site between patients and controls. Allele frequencies of HaeIII in diabetic subjects were 36% and 64% for Asp and Glu respectively (p < 0.001; chi(2) = 29.5). The frequency of Asp/Asp and Glu/Glu genotypes was increased in controls and diabetic subjects respectively. DBP alleles in individuals carrying the DQB1 NA combination revealed that 46.6% of diabetics were DBP Asp/Glu, but this was not statistically significant using the Fisher exact test (16/31 vs. 0/3; p = 0.23). CONCLUSIONS: The study of the DQB1 chain confirmed the value of alleles encoding for an amino acid different from Asp in position 57 (NA) in the susceptibility to type 1 diabetes. The allele frequency of the HaeIII site, but not of the StyI site, differed between patients and controls (HaeIII p < 0.001; StyI p > 0.05).

Macrophage activation in type 1 diabetic patients with catheter obstruction during peritoneal insulin delivery with an implantable pump.

OBJECTIVE: The purpose of this study was to evaluate the activation of macrophages in type 1 diabetic patients during peritoneal insulin delivery with an implantable pump against two types of insulin: that which was collected from the pump reservoir and that which came straight fromn the bottle (i.e., vial insltlin). Macrophage activation was studied in patients with and without cathcter obstruction and compared with activation in healthy subjects. RESEARCH DESIGN AND METHODS: Human insulin (21 PH, 400 U/ml; Hoescht) was collected from the pump reservoir (Minimed) of diabetic patients with (n = 3) or without (n = 7) catheter obstruction, as assessed by histological examination of the catheter tip. Monocytes were obtained from venous blood samples from both kinds of diabetic patients and from healthy subjects (n = 5) and were differentiated into monocyte-derived macrophages in culture. Their chemotaxis and tumor necrosis factor-alpha (TNF-alpha) release were studied with respect to both types of insulin, as previously stated. Formyl-methionyl-leucyl-phenylalanine (fMLP) and lipopolysaccharide (LPS) were used as controls. RESULTS: Neither insulin recovered from the pump reservoir nor vial insulin proved chemotactic to macrophages from either healthy subjects or those diabetic patients with and without catheter obstruction. The migration toward fMLP of macrophages from patients presenting a catheter obstruction was significantly higher than that observed with macrophages from either diabetic patients without obstruction or healthy subjects, the chemotactic index (mean +/- SD) was 3.81 +/- 0.36 vs. 2.30 +/- 0.89 and 2.60 +/- 0.80, respectively (P < 0.05). LPS significantly stimulated the TNF-alpha secretion of macrophages from diabetic subjects with a catheter obstruction, whereas both native and reservoir-recovered insulin had no effect on this release (144.83 +/- 67.25 vs. 5.15 +/- 2.93 and 5.27 +/- 2.43 pg/ml, P < 0.001). CONCLUSIONS: The human insulin used in implantable pumps, regardless of how long it had remained in the pump reservoir, did not induce macrophage activation in diabetic patients treated through intraperitoneal insulin delivery. In some of these diabetic patients, catheter obstruction could be explained by their high capacity of macrophage chemotaxis.

Chemotaxis activation of peritoneal murine macrophages induced by the transplantation of free and encapsulated pancreatic rat islets.

The present study concerns the influence of the transplantation of free and encapsulated (AN69 membrane, Hospal) islets on the chemotaxis of peritoneal macrophages. Fifty free or encapsulated rat islets, cultured for 24 h, were transplanted in the peritoneal cavity of mice (n = 12). Three days after transplantation, the chemotaxis of peritoneal murine macrophages was tested towards formyl-methionyl-leucyl-phenylalanine (fMLP) and a culture medium conditioned for 3 days by free rat islets isolated from the same rat donor. In response to fMLP, the chemotactic indexes of macrophages from mice transplanted with free or encapsulated islets were 8.09 +/- 2.10 and 9.45 +/- 2.76, respectively. These values were significantly higher than those obtained when macrophages from untreated mice were tested (2.42 +/- 0.23; p < 0.01). In response to culture medium conditioned by free islets, the transplanted encapsulated islets failed to enhance macrophage chemotaxis (2.41 +/- 0.53) compared to transplanted free islets (7.00 +/- 2.63; p < 0.01). Thus, encapsulation decreased the specific chemotactic activity of peritoneal macrophages induced by free islet transplantation, probably by prohibiting the diffusion of chemoattractants.

Immobilization of pancreatic islet cells with preserved secretory potential.

Dispersed pancreatic islet cells from rats were cultured overnight in the presence of macroporous gelatin microcarriers. The cells attached to the microcarriers were then incubated for 90 min in the absence or presence of 15.0 mM D-glucose and/or 1.25 mM theophylline. The release of insulin during incubation was about three times higher in the simultaneous presence of these two secretagogues than in their absence. This procedure could thus be used for the immobilization of pancreatic islet cells with preserved secretory potential.

Chemotactic activity of culture supernatants of free and encapsulated pancreatic rat islets towards peritoneal macrophages.

Longterm efficiency of encapsulated pancreatic islet transplantation is limited by macrophagic reaction at the surface of biocompatible membrane. The aim of this work was to investigate the influence of soluble factors released by free and encapsulated islets on macrophage chemotaxis. The culture mediums conditioned for 6 days by free and encapsulated rat islets were incubated with peritoneal murine, rat allo and syngenic macrophages to study their migration. Culture supernatants of rat fibroblasts and acinar cells, glucose-stimulated free rat islets and supernatants of free rat islets treated by heat and proteinase K were also tested for their chemotactic activity. Islets encapsulation decreased the chemotactic activity of culture medium conditioned for 6 days by free rat islets on murine (1.66 +/- 0.20 vs. 3.10 +/- 0.23; p < 0.001, n = 5) and rat allogenic macrophages (1.63 +/- 0.21 vs. 4.70 +/- 0.36; p < 0.001, n = 9). There was no migration of rat macrophages towards syngenic islets. Fibroblasts exhibited a very strong chemotactic effect as compared to acinar cells. Insulin was not involved in macrophage migration. Proteinase K treatment of culture supernatant of free rat islets totally inhibited the chemotactic activity. After heating at 56 degrees C and 100 degrees C, this activity was reduced to 41 +/- 7% and 32 +/- 5% of the initial activity, respectively. In conclusion, pancreatic islet stimulated macrophage migration by release of immunological specific proteins partly retained by macroencapsulation.

Influence of acinar tissue contamination on encapsulated pancreatic islets: morphological and functional studies.

BACKGROUND: The present study concerns the influence of acinar contamination on pancreatic islets encapsulated in an artificial membrane (AN 69). METHODS: Pure, handpicked pancreatic islets were contaminated by the addition of acinar tissue (ratio, 1:1). The morphological aspect and insulin release of both pure (n=12) and contaminated (n=12) encapsulated islets were assessed after 10 days of culture or implantation in the peritoneal cavity of rats. RESULTS: After implantation, the encapsulated islets, irrespective of their purity, were totally altered, whereas the morphological aspect of the cultivated islets remained intact only in the absence of acinar tissue contamination. This contamination induced a significant decrease in the stimulation index of insulin release. The stimulation index decreased by 42% for fresh islets and by 52% and 34% for cultivated and implanted islets, respectively, without modification in their basal release of insulin. CONCLUSIONS: The acinar tissue proved detrimental to the encapsulated implanted and cultivated islets.

Physicochemical and biological studies of corona-treated artificial membranes used for pancreatic islets encapsulation: mechanism of diffusion and interface modification.

The artificial AN69 membrane (Hospal), a synthetic copolymer composed of acrylonitrile and sodium methallyl sulphonate suitable for pancreatic islet encapsulation, was submitted to physicochemical treatment (Corona discharge) to improve its insulin permeability. X-ray photoelectron spectroscopy (XPS) analysis of the AN69 membrane indicated the presence of up to two molecular layers of glycerol at its surface while the surface energies revealed the presence of hydrophilic sites (-SO3Na/glycerol) located at the membrane surface and acrylonitrile hydrophobic groups inside the material. The Corona discharges decreased the number of glycerol molecules at the membrane surface and from a biological point of view, produced a threefold increase in insulin diffusion. Furthermore, the biocompatibility of the treated membrane was preserved after 1 year of intraperitoneal implantation. The increase in insulin permeability should result from a decrease of the membrane polarity and of a steric hindrance in pores. Thus, Corona discharge treatment may serve to optimize the properties of artificial membranes used for pancreatic islets encapsulation.

Cytotoxicity of peritoneal murine macrophages against encapsulated pancreatic rat islets: in vivo and in vitro studies.

The aim of this work was to study in vivo and in vitro the involvement of macrophages and interleukin-1beta (IL-1beta) in the necrosis of encapsulated islets during xenograft and to evaluate the immunoprotective efficiency of the AN69 membrane. In vivo, 6 days after implantation, 65% of the membrane surface of the devices containing the islets was colonized with macrophages compared with only 5% of the surface of the empty control devices. The morphological aspect of implanted islets was altered and their insulin release decreased significantly compared with freshly isolated ones (265 +/- 50 vs. 507 +/- 81 microU/mL). In vitro, the insulin release of encapsulated islets cultured for 2 days decreased to 32 and 28%, respectively, in the presence of IL-1beta and macrophages. The addition of anti-IL-1beta antibody to the co-culture of macrophages and islets did not modify this loss of functional activity. Furthermore, IL-1beta passed through the AN69 membrane. In conclusion, macrophages are involved in damaging encapsulated pancreatic islets and are probably partly responsible for islet transplantation failure.

Pancreatic islet macroencapsulation: a new device for the evaluation of artificial membrane.

We propose a new device especially designed to test membranes for islet macroencapsulation. It is composed of three independent parts: a support made of three polytetrafluoroethylene rings, the membrane that forms the walls of the encapsulation chamber, and a collagen gel that immobilizes the islets. Studies of this device were performed with the AN69 membrane. After 2 months of implantation in the peritoneal cavity of rats, the empty device was found to be biocompatible, referred to as weak cellular adhesion. In vitro the encapsulation preserved the peak of insulin release in response to high glucose during a perifusion test (0.36 +/- 0.02 microU/ml/islet for free or encapsulated islets). As a result of the collagen gel, the morphological aspect and functional activity were still preserved after 7 days of culture. In vivo xenotransplantation into diabetic mice normalized the fasting glycemia up to 30 days. Numerous macrophages adhered to the outer surface of the membrane, and a layer of cells emerging from the destruction of islets covered the inner surface. In addition, the morphological aspect of many islets was altered. By showing that the AN69 membrane was only partially efficient for islet xenotransplantation, this new device proved to be of interest for testing a variety of membranes.

Role of imidazoline receptors in cardiovascular regulation.

The involvement of nonadrenergic imidazoline specific receptors in the central control of the vasomotor tone and in the mechanism of action of drugs bearing an imidazoline structure, or analogs, is now well documented. Imidazoline-specific binding sites were found in many tissues and species. Moreover, until now, it is only in the brainstem that such binding sites are associated with a function: the hypotensive effect of imidazoline-like drugs. Rilmenidine, which is an oxazoline structurally related to the reference imidazolines, exerts a central hypotensive effect of central origin involving imidazoline receptors. The selectivity of rilmenidine for the imidazoline receptors compared to alpha 2-adrenergic receptors could explain the low incidence of sedative side effects observed with this antihypertensive drug. A specific anti-imidazoline radioimmunoassay allowed us to detect the presence of an immunoreactive imidazoline-like substance in human sera. High levels of this immunoreactive substance are associated with high blood pressure in 20-30% of the hypertensive patients. This observation indicates that high levels of this immunoreactive substance in the serum can be associated with some kinds of primary hypertension. The cause-and-effect relation between these 2 phenomena has not yet been determined. This substance is in process of purification; it could be a candidate to be an endogenous ligand of the imidazoline receptors.