3-D imaging of islets in obesity: formation of the islet-duct complex and neurovascular remodeling in young hyperphagic mice.

BACKGROUND: Obesity and insulin resistance lead to islet hyperplasia. However, how the islet remodeling influences the pancreatic environment and the associated neurovascular networks is largely unknown. The lack of information is primarily due to the difficulty of global visualization of the hyperplasic islet (>200 µm) and the neurovascular environment with high definition. METHODS: We modulated the pancreatic optical property to achieve 3-dimensional (3-D) whole-islet histology and to integrate transmitted light microscopy (which provides the ground-truth tissue information) with confocal fluorescence imaging. The new optical and imaging conditions were used to globally examine the hyperplastic islets of the young (2 months) obese db/db and ob/ob mice, which otherwise cannot be easily portrayed by the standard microtome-based histology. The voxel-based islet micrographs were digitally processed for stereo projection and qualitative and quantitative analyses of the islet tissue networks. RESULTS: Paired staining and imaging of the pancreatic islets, ducts and neurovascular networks reveal the unexpected formation of the 'neuro-insular-ductal complex' in the young obese mice. The complex consists of the peri- and/or intra-islet ducts and prominent peri-ductal sympathetic nerves; the latter contributes to a marked increase in islet sympathetic innervation. In vascular characterization, we identify a decreased perivascular density of the ob/ob islet pericytes, which adapt to ensheathing the dilated microvessels with hypertrophic processes. CONCLUSIONS: Modulation of pancreatic optical property enables 3-D panoramic examination of islets in the young hyperphagic mice to reveal the formation of the islet-duct complex and neurovascular remodeling. On the basis of the morphological proximity of the remodeled tissue networks, we propose a reactive islet microenvironment consisting of the endocrine cells, ductal epithelium and neurovascular tissues in response to the metabolic challenge that is experienced early in life.

Effects of all-trans retinoid acid and exendin-4 on islet transplantation in NOD mice.

Type 1 diabetes usually develops due to autoimmune destruction of ß-cells in the pancreas. It has been shown that all-trans retinoid acid (ATRA), a potent derivative of vitamin A, hinders the development of autoimmune diabetes by inducing immune tolerance status. In addition, exendin-4, a glucagon-like peptide-1 receptor agonist, stimulates growth and differentiation of ß-cells and exerts anti-apoptotic effect on ß-cells. Thus, we hypothesized that the ATRA and exendin-4 therapy may improve the outcome of islet transplantation in non-obese diabetic (NOD) mice. After the onset of diabetes, each NOD mouse was transplanted with 300 or 600 islets isolated from NOD/severe combined immunodeficient (SCID) mice with or without treatment of ATRA (0.5 mg intraperitoneally every other day) and/or exendin-4 (3 µg/kg subcutaneously twice daily) for 6 weeks. After 300 or 600 NOD/SCID islet transplantation without any other treatment, all NOD recipients remained diabetic. However, the lowest blood glucose level in mice transplanted with 600 but not 300 islets was significantly lower than those without islet transplantation (P < .05), although their survival time was comparable. Among recipients treated with ATRA, exendin-4, ATRA and exendin-4, and without treatment, their lowest blood glucose levels and survival time were not different. However, one recipient treated with ATRA survived for 223 days with intermittent hyperglycemia and the other who was treated with ATRA and exendin-4 achieved normoglycemia. In conclusion, islet transplantation lowered blood glucose levels in diabetic NOD mice. With a few exceptions, treatment with ATRA and exendin-4 alone or in combination in islet recipients could not reverse diabetes or prolong survival.

Coculture of hepatocytes with islets.

Bioartificial liver support (BAL) systems are potential new therapeutic approaches for use as liver support to prevent nutrient deficiencies, hypoxia, or ischemia before the acquisition of donated organs. To investigate whether islets are beneficial for hepatocyte function and survival, we cocultured BALB/c mouse islets with C57BL/6J hepatocytes to assess hepatocyte viability, function, and apoptosis. We observe cell viability to decrease progressively by 50% from day 0 to day 3 among isolated hepatocytes (group A) and hepatocytes cocultured with islets (group B). However, group A was prone to necrosis and reduced albumin secretion during culture. In contrast, at day 7 group B maintained albumin secretion (0.3351 ± 0.0581 vs 0.1451 ± 0.0329 µg/h/mL; P < .05). Early apoptosis was observed at day 3 among group A but at day 7 in group B. In addition, quantitative analysis of the apoptotic cells revealed group B to show a delayed phenotype of both early and late apoptosis compared with group A. Our results indicated that islets could retain hepatocyte function and delay apoptosis, suggesting that the coculture system is potentially applicable to develop a high-performance BAL.

Modest effects of Fas-ligand and heme oxygenase-1 double transgenic mouse islets on transplantation outcomes.

Interactions of Fas with its ligand (FasL) play an important role in the maintenance of immunologic homeostasis and peripheral tolerance. Heme oxygenase-1 (HO-1) is a protein capable of cytoprotection via radical scavenging and apoptosis prevention. The aim of this study was to test whether overexpression of FasL and HO-1 in murine islets resulted in cell protection and improved functional performance after transplantation. We first generated FasL and HO-1 double transgenic mice to investigate the protective effect of transgenic islets on transplantation. Islets were isolated from FasL and HO-1 double transgenic and nontransgenic Balb/c mice, for transplantation of 300 islets under the left kidney capsule of each streptozotocin-diabetic Balb/c mouse. During 6 weeks after transplantation, the blood glucose gradually decreased in recipients of double transgenic and nontransgenic islets. However, the decrease in blood glucose was more pronounced in the former (450 ± 16 mg/dL at day 0 to 302 ± 55 mg/dL at day 42; P = .01) than the latter (468 ± 17 mg/dL at day 0 to 379 ± 71 mg/dL at day 42; P = .24). The areas under the curve of intraperitoneal glucose tolerance tests at 2, 4, and 6 weeks were not significantly different between recipients of double transgenic and nontransgenic islets. The body weight increased in recipients of double transgenic islets (21.1 ± 1.4 g at day 0 to 26.2 ± 0.8 g at day 42; P = .0002) and nontransgenic islets (21.0 ± 1.4 g at day 0 to 25.1 ± 0.4 g at day 42; P = .0448). Our data suggested modest beneficial effects of transgenic islets with FasL and HO-1 overexpression for transplantation.

Magnetic resonance imaging study of mouse islet allotransplantation.

Although only 10% of islet transplant recipients maintain insulin independence, 80% of them are C-peptide positive at 5 years. To better understand the fate of transplanted islets, a magnetic resonance imaging (MRI) technique has been used to detect superparamagnetic iron oxide (SPIO)-labeled transplanted islets. Recently, we successfully used a novel MRI contrast agent, chitosan-coated SPIO (CSPIO) nanoparticles, to monitor mouse islet isografts for 18 weeks after transplantation. In the present study, we tested whether CSPIO could be applied to monitor islet allografts, which are supposedly rejected without immune interventions. Male C57BL/6 and Balb/c mice were used as donors and recipients of islet transplantation, respectively. After overnight incubation with or without CSPIO (10 µg/mL), 300 C57BL/6 islets were transplanted under the left kidney capsule of each Balb/c mouse. Starting from day 10 after transplantation, 3.0-Tesla MRI of the recipients was performed weekly. Four mice were followed for ≥38 days. At 38 and 45 days, 1 islet graft was removed for insulin and Prussian blue staining, respectively. From days 10 to 45 after transplantation, CSPIO-labeled islet grafts were visualized on MRI scans as sustained distinct hypointense spots homogeneously located at the upper pole of left kidney, the site of transplantation. At days 38 and 45, the histology of CSPIO-labeled islet grafts revealed insulin and iron staining colocalized in the same areas. Our results in a mouse allotransplantation model indicated that CSPIO-labeled islets survived as long as 45 days with positive MRI.

Effects of cyclooxygenase-2 inhibitor and adenosine triphosphate-sensitive potassium channel opener in syngeneic mouse islet transplantation.

In the initial days after transplantation, islet grafts may be attacked by cytokines via cyclooxygenase-2 (COX-2), producing primary nonfunction. In addition, chronic overstimulation of ß-cells may impair insulin secretion. To enhance the function of transplanted islets, the present study investigated the effects of rofecoxib, a COX-2 inhibitor, and NN414 (6-chloro-3-[1-methylcyclopropyl]amino-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-dioxide), an adenosine triphosphate-sensitive potassium channel opener, on islet transplantation. Male inbred C57BL/6 mice were used as donors and recipients. One hundred fifty islets were isolated via collagenase digestion and density gradient, and syngeneically transplanted under the kidney capsule in mice with streptozotocin-induced diabetes. Recipients were treated with or without rofecoxib, 10 mg/kg/d orally, or with or without NN414, 3 mg/kg/d orally, for 4 weeks. After transplantation, recipient body weight, blood glucose concentration, and intraperitoneal glucose tolerance were measured. The grafted kidney was extracted for determination of insulin content at 4 weeks. In the rofecoxib-treated and NN414-treated groups and both control groups, body weight remained stable, and the blood glucose concentration decreased progressively. However, at 4 weeks after transplantation in the groups treated or not treated with rofecoxib or NN414, no significant difference was observed in recipient body weight, blood glucose concentration, and glucose tolerance or in insulin content of the graft. These data indicate that posttransplantation treatment with rofecoxib or NN414 has no beneficial effect on transplantation outcome in diabetic mouse recipients engrafted with a marginal islet mass.

Magnetic resonance imaging of transplanted mouse islets labeled with chitosan-coated superparamagnetic iron oxide nanoparticles.

Although only 10% of islet recipients maintain insulin independence, 80% of them are C-peptide positive at 5 years after transplantation. To better understand the fate of transplanted islets, a magnetic resonance imaging (MRI) technique has been used to detect Feridex-labeled islet grafts in rodents. In this study, we used a novel MRI contrast agent, chitosan-coated superparamagnetic iron oxide (CSPIO) nanoparticles, to monitor mouse islet grafts. Male inbred C57BL/6 mice were used as donors and recipients of islet transplantation. The islet cytotoxicity was evaluated by fluorescein diacetate and propidium iodide staining for RAW cells incubated with CSPIO. After being incubated overnight with and without CSPIO (10 mg/mL), 300 islets were transplanted under the left kidney capsule of each mouse. After transplantation, 3.0-Tesla MRI of the recipients was performed biweekly until 19 weeks. At the end of study, the islet graft was removed for insulin and Prussian blue staining. The cell death rates in RAW cells did not increase with increasing CSPIO concentrations or incubation time. The grafts of CSPIO-labeled islets were visualized on MRI scans as distinct hypointense spots homogeneously located at the upper pole of left kidney. Their MRI signal was 30%-50% that of control islets and was maintained throughout the follow-up period. At 18 weeks, the histology of CSPIO-labeled islet graft revealed the insulin- and iron-stained areas to be almost identical. Our results indicate that isolated mouse islets labeled with CSPIO nanoparticles can be effectively and safely imaged by using MRI as long as 18 weeks after transplantation.

Effects of insulin-like growth factor-1 and donor age on transplantation of porcine neonatal pancreatic cell clusters.

Porcine neonatal pancreatic cell clusters (NPCCs) isolated from 1- to 3-day-old pigs cured diabetic nude mice more than 14 weeks after transplantation. To shorten the latent period between transplantation and reversal of hyperglycemia, we investigated the effects of insulin-like growth factor-1 (IGF-1) and NPCCs isolated from 1-month-old pigs after transplantation. Pig pancreata were cut into fragments, collagenase digested, and then cultured. Three hundred and 2000 NPCCs were transplanted under the kidney capsule of nondiabetic and diabetic nude mice, respectively. After transplantation, the graft-bearing kidneys were removed to measure insulin content. NPCCs isolated from 1- to 3-day-old pigs were cultured with or without IGF-1 for 6 days. The stimulation index was not significantly different between the 2 groups at 1, 2, or 4 weeks. Moreover, at 4 weeks after transplantation of 300 NPCCs to nondiabetic nude mice yielded comparable graft insulin content as the recipients of NPCCs precultured with or without IGF-1. Two thousand cultured NPCCs isolated from 1-to 3-day-old pigs or 1-month-old pigs were transplanted into diabetic nude mice. The blood glucose levels of diabetic recipients in both groups decreased at the same rate after transplantation, achieving normoglycemia at 8 weeks. The graft insulin content at 12 weeks was not different between the 2 groups. Our data indicated that isolated NPCCs cultured with IGF-1 showed no beneficial effects on insulin secretion and transplantation; NPCCs isolated from 1-to 3-day-old and 1-month-old pigs displayed similar effects on transplantation.

Effects of heme oxygenase-1 transgenic islets on transplantation.

Heme oxygenase-1 (HO-1) has been described as a protein capable of cytoprotection via radical scavenging and apoptosis prevention. The aim of this study was to analyze whether HO-1 overexpression in freshly isolated murine transgenic islets resulted in cell protection and improved in vivo functional performance after transplantation. We produced transgenic mice in which the human HO-1 transgene driven by chicken beta-actin promoter was expressed in the heart, liver, spleen, lung, kidney, muscle, intestine, and pancreas in Balb/c mice. One hundred fifty islets isolated from HO-1 transgenic and control Balb/c mice were syngeneically transplanted under the left kidney capsule of the streptozotocin-diabetic Balb/c mice. The recipients who underwent transplantation with HO-1 transgenic islets showed higher blood glucose than those with control islets at 4 weeks (320 +/- 25 vs 189 +/- 43 mg/dL; P < .05). Body weight was not significantly different between the 2 groups. Our data indicate transgenic islets with high HO-1 expression did not improve transplantation outcome.

Effects of insulin sensitizers on islet transplantation.

To solve the problems of islet engraftment, we investigated the effects of insulin sensitizers, metformin and rosiglitazone, on the in vitro and in vivo function of mouse islets. The in vitro study was done by culturing 30 isolated C57BL/6 mouse islets with glucose (100 or 300 mg/dL) or rosiglitazone (4.5 mumol/L) for 2, 4, 8, or 12 hours. The in vivo study was performed by syngeneically transplanting 150 C57BL/6 mouse islets under the kidney capsule of streptozotocin-diabetic mice. The metformin group was treated with 200 mg/kg/d in water and the control group was pair-fed the same volume of liquid diet. In the in vitro study, insulin release stimulated by 300 mg/dL glucose (n = 6) was the highest at all time points. That stimulated by rosiglitazone (n = 6) was greater than by 100 mg/dL glucose (n = 6) only at 8 hours. In the recipients treated with metformin (n = 17) and controls (n = 13), the blood glucose decreased and body weight increased gradually after transplantation. However, there was no significant difference between the two groups. Their tolerance to intraperitoneal glucose challenge at 2 and 4 weeks was also comparable. At 4 weeks, 12/17 (71%) in the metformin group and 8/13 (62%) in the control group achieved normoglycemia (P = .60). At 4 weeks, the insulin content of the graft was 8.35 +/- 3.42 mg in the metformin group and 5.28 +/- 4.28 mg in the control group (P = .59). Our data indicate that (1) rosiglitazone stimulated isolated islets to release insulin but was less effective than high levels of glucose; and (2) metformin treatment had no beneficial effect on islet recipients.

Islet transplantation at subcutaneous and intramuscular sites.

The subcutaneous site is ideal for clinical islet transplantation because it has the advantage of being accessible and can be biopsied when needed. Sadly, the results at subcutaneous sites were disappointing. The reason for this failure is not known, but poor vascularization may play a role. We tested the hypothesis that islet grafts would do better if more vasculature or oxygen could be supplied. Six hundred isolated C57BL/6 mouse islets were syngeneically transplanted into inbred streptozotocin-diabetic recipients at a subcutaneous site on the back with (Group A, n = 6) or without (Group B, n = 8) postoperative hyperbaric oxygen (2.4 ATA, 100% O(2)) therapy, or at a calf muscle (Group C, n = 9). During 13-week posttransplantation follow-up, recipients' blood glucose decreased and body weight increased significantly in all 3 groups (P < .05). However, there was no significant difference among the 3 groups. At 13 weeks, the insulin contents of the graft was also comparable among the 3 groups. Our data indicate the following: (1) postoperative hyperbaric oxygen therapy did not improve the outcome of islet transplantation at a subcutaneous site; and (2) a muscular site was not superior to a subcutaneous site for islet transplantation.

Effects of intrahepatic canine islet autotransplantation on the portal vein pressure and liver function.

Currently, the most common method used for human islet transplantation is intrahepatic implantation via the portal vein, which may affect portal vein pressure and liver function. The aim of this study was to investigate the effects of intrahepatic canine islet autotransplantation on portal vein pressure and liver function. After total pancreatectomy was performed in 30 mongrel dogs, islets were isolated and transplanted back into the portal vein of the same dog. In our series, 12 dogs achieved normoglycemia (fasting glucose <200 mg/dL) without exogenous insulin after transplantation. The portal vein pressure increased from 4.6 +/- 1.5 to 7.7 +/- 2.9 cm H(2)O after islet infusion (P < .05). Alanine transferase amino transferase (ALT) levels gradually increased after pancreatectomy with the peak at 4 weeks after islet infusion. But the changes of portal vein pressure and ALT were not significantly different between successful and failed islet transplantation. In summary, elevation of portal vein pressure and liver enzymes were noted after intrahepatic canine islet autotransplantation. However, they did not influence the transplant outcome.

Enhancing islet engraftment with rosiglitazone.

To study the role of a peroxisome proliferator-activated receptor agonist, rosiglitazone, on islet engraftment, streptozotocin-induced diabetic C57BL/6 mice were fed daily rosiglitazone (2.4 mg/kg) for 9 and 31 days starting 2 days before transplantation with 75 and 150 syngeneic islets, respectively. After receiving 75 islets and 9 days of rosiglitazone, half of the treated diabetic mice became normoglycemic at 4 weeks, while none were normoglycemic among those mice that did not receive treatment. After transplanting 150 islets and receiving 31 days of rosiglitazone, 80% of the treated diabetic mice became normoglycemic while the incidence was only 25% for the controls. The insulin content of the islet grafts in the rosiglitazone groups was 0.8 times (75-islet group) and 1.3 times (150-islet group) higher than that of control mice. The insulin content of pancreatic remnants did not differ significantly among all groups. An in vitro study revealed that the glucose-stimulated insulin secretion and insulin content of cultured islets was not different in the presence versus absence of 4.5 or 22.5 micromol/L rosiglitazone. In vitro study revealed that rosiglitazone inhibited the lipopolysaccharide-induced secretion of interleukin-1 beta and interferon-gamma from peritoneal exudate cells. In conclusion, our data suggest that short-term administration of rosiglitazone enhances islet engraftment.

Cobalt-protoporphyrin treatment enhances murine isoislets engraftment.

To study the effect of treatment with cobalt-protoporphyrin (CoPP) for the induction of the heme oxygenase-1 (HO-1) enzyme on islet engraftment donor mice received either a single intraperitoneal injection of CoPP (20 mg/kg body weight) 1 day prior to islet isolation or this injection plus a 9 day posttransplantation course of Copp. After a single injection of CoPP, the CoPP-induced islets contained higher HO-1 proteins than did the normal islets both at 12 (5.3 +/- 1.5 vs 0.1 +/- 0.1 ng/mg protein, P < .01) and at 30 hours (6.8 +/- 2.1 vs 0.4 +/- 0.3 ng/mg protein, P < .05), but not at 56 hours (1.9 +/- 0.8 vs 1.6 +/- 0.8 ng/mg protein, P > .05). In contrast, diabetic mice that received 75 CoPP-induced islets and a 9-day CoPP injection course posttransplantation showed better improvement in blood glucose levels and body weights than did the mice that only received CoPP-induced islets. Mice of both CoPP-treated groups displayed better improvement in glycemic control than mice that received control islets. At 8 weeks after transplantation, the insulin content of grafts from both CoPP groups was significantly higher than that in the control group. In conclusion, CoPP treatment for the induction of HO-1 enhances engraftment of islets in a syngeneic murine transplantation model.

Cobalt-Protoporphyrin treatment renders islets tolerant to interleukin-1 beta suppression.

This study examined whether treating donor mice with a single dose of cobalt protoporphyrin (CoPP) induced heme oxygenase-1 (HO-1) and protected islet cells from interleukin-1 beta (IL-1 beta) suppression. Islets were isolated from mice receiving a single dose of either CoPP (20 mg/kg of body weight, CoPP islets) or isotonic NaCl solution vehicle (control islets), 24 hours before isolation. Glucose-stimulated insulin secretion (GSIS) and insulin content (IC) of the islets were determined following incubation in the presence versus absence of murine IL-1 beta for 21 or 65 hours. The HO-1 protein level of CoPP-induced islets, as determined by an enzyme immunoassay, was significantly higher than that of control islets at 12 hours (P <.01) and 30 hours (P <.05), and returned to basal levels at 56 hours (P = NS). Following a 21-hour incubation with IL-1 beta, CoPP islets secreted significantly more insulin upon glucose stimulation and preserved significantly more IC than control islets. After 65-hour incubation with IL-1 beta, CoPP islets secreted significantly less insulin upon glucose stimulation than control islets and preserved significantly less IC compared to islets incubated without IL-1 beta. In conclusion, treatment with cobalt-protoporphyrin to induce heme oxygenase-1 protects islets against the suppressive effects of IL-1 beta.

Characteristics and transplantation of the porcine neonatal pancreatic cell clusters isolated from 1- to 3-day-old versus 1-month-old pigs.

Porcine neonatal pancreatic cell clusters (NPCCs) isolated from 1- to 3-day-old pigs (I-A) cured diabetic nude mice within 8 weeks after transplantation. To shorten the latent period between transplantation and reversal of hyperglycemia, we studied NPCCs isolated from 1-month-old pigs (I-B). One- to 3-day-old or 1-month-old pig pancreata were cut into fragments, digested by collagenase, and then studied for islet characteristics. In addition, 300 cultured NPCCs were transplanted under kidney capsule of nondiabetic nude mice. At 1 and 3 months after transplantation, the grafts were removed to measure the insulin content and beta-cell mass. Immediately after isolation, I-B was larger than I-A (0.211 +/- 0.006 vs 0.189 +/- 0.003 mm(2), P =.0003) and after a 6-day culture period, I-B contained more insulin than I-A (6.8 +/- 1.4 vs 2.3 +/- 0.2 microg/150 NPCCs, P =.02). However, the stimulation indices of I-A and I-B during static incubation with 500 mg/dL glucose (26.5 +/- 3.2 vs 23.9 +/- 1.7) or 500 mg/dL glucose plus 50 mol/L IBMX (41.9 +/- 4.4 vs 62.2 +/- 14.0) were not significantly different. Furthermore, neither I-A nor I-B showed first or second phase insulin secretion during sequential perifusion with 100 or 300 mg/dL glucose. In nondiabetic recipients, the insulin content of the graft at 1 month after transplantation was 0.3 +/- 0.0 and 0.3 +/- 0.1 microg, and the beta-cell mass of the graft at 3 months was 0.069 +/- 0.022 and 0.067 +/- 0.023 mg in mice receiving I-A or I-B, respectively (P >.05). Our data indicate NPCCs isolated from 1- to 3-day-old and 1-month-old pigs have different characteristics but similar transplantation effects.