Variables in organ donors that affect the recovery of human islets of Langerhans

In an attempt to reduce the variability in the yields of human islets isolations and to identify donor factors that were potentially deleterious, we retrospectively reviewed 153 human islets isolations in our center over a 3-year period. Isolations were performed using controlled collagenase perfusion via the duct, automated dissociation, and Ficoll purification. Factors leading to successful isolations (recovery of >100,000 islet equivalents at a purity >50%) were analyzed retrospectively using univariate and multivariate analysis. Critical factors in the multiorgan cadaveric donors that were identified using univariate analysis included donor age (P<0.01), body mass index (BMI)(P<0.01), cause of death (P<0.01), and prolonged hypotensive episodes (systolic blood pressure <90 mmHg or mean arterial pressure <60 mmHg for > 15 min) requiring high vasopressors (>15 microgram/kg/min dopamine or >5 microgram/kg/min Levophed) (P>0.01). Independent analysis of 19 donor variables using multivariate logistic stepwise regression showed six factors were statistically significant. Odds ratio (OR) showed that donor age (OR 1.1, P<0.01), local procurement team (OR 10.9, P<0.01), and high BMI (OR 1.4, P<0.01) had a positive correlation with islet recovery. In contrast, hyperglycemia (all blood glucose >10 mmol/L) (OR 0.63, P<0.01), frequency and duration of cardiac arrest (OR 0.7, P<0.01), and increased duration of cold storage before islet isolation (OR 0.83, P<0.01) had negative correlation. Using these combinations of factors, the prediction of success was 85% accurate. By donor age, success was 13% for 2.5- to 18-year-old donors (n=23), 37% for 19- to 28-year-old donors (n=30), 65% for 29- to 50-year-old donors (n=70), and 83% for 51- to 65-year-old (n=29) donors. However, when vitro function was assessed by perifusion, the insulin secretory capabilities of islets isolated from the >50-year-old donor group was significantly reduced as compared with the 2.5- to 18-year-old group (P<0.02). Multiple regression analysis using postdigestion and postpurification islet recovery as outcome variables identified BMI, procurement team, pancreas weight, and collagenase digestion time factors tht can affect the recovery of human islets. Locally procured pancreases and donors with elevated minimum blood glucose levels were identified as factors that affect the insulin secretory capabilities of the isolated islets. This review of parameters suggests an improved approach to the prediction of successful islet isolation from human pancreases. Selection of suitable pancreases for processing may improve consistency in human islet isolation and thereby decrease costs.     

Expression of Gal alpha(1,3)gal on neonatal porcine islet beta-cells and susceptibility to human antibody/complement lysis

Neonatal porcine pancreases may be a potential source of islets for transplantation into patients with type 1 diabetes; however, whether these cellular grafts will be susceptible to damage by human natural antibody-mediated rejection remains controversial. Although we and others have demonstrated that porcine islets bind human IgG and IgM, it remains unknown if they express the xenoreactive antigen Gal alpha(1,3)Gal beta(1,4)GlcNAc-R (Gal epitope). In this study, by using the Gal-specific lectin IB4 for immunohistochemistry and fluorescence-activated cell sorter (FACS) analysis, we determined which cell types present in porcine neonatal islet cell (NIC) aggregates express the Gal epitope and which ones are susceptible to lysis by activation of the human complement. After FACS analysis, 30.0 +/- 3.0% of porcine NICs were shown to express Gal, whereas 70.0 +/- 2.0% did not. Histological assessment of Gal-expressing cells revealed that 54.9 +/- 8.8% stained positive for either insulin or glucagon. In contrast, 68.8 +/- 8.4% of the Gal-negative population stained positive for the pancreatic hormones insulin and glucagon. Incubation of either the Gal-positive or -negative cells with human AB serum plus complement for 1.5 h resulted in the lysis of >90% of the cells. These results demonstrate that porcine NIC aggregates are composed of Gal-expressing cells and that expression of Gal is not restricted to nonendocrine cells. Furthermore, both Gal-positive and Gal-negative cells are susceptible to human antibody/complement-mediated cytolysis, suggesting that this form of immunological destruction is an obstacle that will need to be overcome before porcine NIC aggregates can be used clinically.