Insulin release from cultured fetal human pancreas.

Abstract

The pancreas from human fetuses of 55-140 days gestation was grown on grids in organ culture dishes. Insulin was shown to be released at no earlier than 63 days gestation, but the amount increased thereafter with gestational age. The insulin content of the explants varied considerably. However, it appeared clear that pancreas from younger fetuses (55 and 63 days gestation) had much less insulin content than those from older fetuses (99-140 days gestation). Whereas 16.7 mM glucose alone caused no increase in insulin release, when used with 10 mM theophylline, a marked release occurred. Tolbutamide (1.85 mivr) plus glucose (16.7 mM), and tolbutamide (3.7 mM) plus glucose (1.67 mM) plus theophylline (10 mM) were also effective. No effect upon insulin release was produced by glucose (1.67 mM) plus theophylline (10 mM), by glucose (1.67 HIM) plus tolbutamide (3.7 mM), or by glucose (16.7 mM) plus arginine (11 mM). {Endocrinology 91: 1133, 1972) T N THE HUMAN fetus, pancreatic beta cells -*• may be distinguished by the presence of aldehyde-fuchsin positive granules during the third month of gestation (1). During the fourth-fifth months of gestation, accumulation of secretory granules in the capillary poles of the beta cells suggests active insulin synthesis and probably release (2). The ultrastructural features of alpha, beta, delta and exocrine acinar cells of the fetal human pancreas have been described in a recent report (3). Circulating insulin has been measured in the human fetus as early as 12 weeks of gestation (4,5). Glucagon, but not glucose or tolbutamide, enhanced insulin release from pancreatic islets isolated from human fetuses of 12-16 weeks of gestation (6). Recently, it has been reported that insulin release was significantly enhanced from human fetal pancreatic slices by a number of agents (arginine, leucine, glucagon, theophylline, dibutyryladenosine cyclic 3',5'-monophosphate, etc.) but not by glucose (7). These fetuses were of 1624 weeks gestation. Immunoassayable insulin and glucagon have been demonstrated in slices of fetal human pancreas by the 50th day of gestation; neither hormone was found in the pancreas of a 2 5-day-old fetus (L. D. Schaeffer, M. W. Wilder, and R. H. Williams, in preparaReceived January 4, 1972. 1 Supported in part by PHS grants AM-O5O2O and AM-02456. 2 Supported in part by PHS Fellowship 1F03 AM-47142-01A1 and Research Career Development Award 1K04 AM-47142-01 from the National Institute of Arthritis and Metabolic Diseases. tion). We now report studies on organ cultures of fetal human pancreas as an approach to the investigation of insulin release in the developing human pancreas. Materials and Methods Human fetuses were obtained at therapeutic abortion, except for one fetus obtained from an ectopic pregnancy. Gestational ages, estimated from crown to rump length, ranged from 55-140 days (8). The pancreas was identified with a dissecting microscope, removed and immediately transported in isotonic saline to the laboratory. Excess peripancreatic tissue was removed. The pancreas was cut into pieces about 2 mm thick and placed on a stainless steel grid in an organ culture dish. Medium 199 (9), supplemented with fetal calf serum (15% v/v) and neomycin (50 ngm/ml), was added. As measured by a Technicon Auto Analyzer, glucose concentration in this medium was about 115 mg/100 ml. Cultures were maintained at 37 C in an atmosphere of 5% CO2 and 95% air. Culture medium was usually changed every other day and saved for cellulose radioimmunoassay of immunoreactive insulin (IRI), using a Novo human insulin standard (10). Insulin was extracted from some explants by the method of Davoren (11). Some explants were also examined after hemotoxylin-eosin and aldehyde-fuchsin stains. During the second week of culture, studies of insulin release in the presence of glucose, tolbutamide, arginine, or theophylline as stimulatory agents were performed in Krebs-Ringer phosphate buffer (KRP), pH 7.4, with 2% (v/v) fetal calf serum. In each study, the culture was incubated first in a baseline medium of KRP containing 1.67 mM glucose for 1530-min periods. The medium was then changed to KRP containing the test substance and incubation was continued for another 60 min (in a few cases 30 min, for which IRI release was not significantly