Abstract.Aminoguanidine (AG) is a potential therapeutic agent for preventing the generation of advanced glycation end products in diabetes mellitus.In this study, the effect of AG on insulin secretion was investigated in in vitro rat pancreatic islets. The islets were aseptically isolated and cultured in fissure culture medium 199 for 48 h with or without AG. After the culture, batches of 10 islets were incubated in Krebs-Ringer bicarbonate buffer containing 3.3 mM or 16.7 mM glucose. Islets previously exposed to 0.18 mM AG or 0.45 mM AG showed similar insulin release to control islets at a 16.7 mM glucose concentration, but high glucose-stimulated insulin release was inhibited in the islets exposed to 1.8 mM. In the perifusion experiment, insulin release caused by 16.7 mM glucose from the islets previously exposed to 1.8 mM AG was not significantly different from that of the control islets. However, culture of the islets with higher AG concentrations, 4.55 mM and 9.1 mM, significantly inhibited glucose-stimulated insulin release (<0.02 and 0.002, respectively). These results suggest that AG at high concentrations impairs pancreatic B-cell response to a high concentration of glucose.

To evaluate the role of pancreatic amylin in the pathogenesis of non-insulin-dependent diabetes mellitus (NIDDM), we determined the pancreatic amylin (IRA) and insulin (IRI) contents of pancreata obtained at autopsy from diabetics and nondiabetics. IRA was extracted from the tail of the pancreas using formic acid and assayed with a human amylin kit. Following gel filtration of amylin on a Sephadex G-50 column, it was eluted in a similar fraction to insulin. The pancreatic IRA content was significantly higher (p < 0.01) in NIDDM subjects compared with nondiabetics, with the mean values being 4.25 2 1.62 and 0.085 2 0.022 p.g/g, respectively. The IRA content of two IDDM pancreata was low. No significant relationship was found between the IRA and the IRI contents or between the IRA content and the duration of diabetes. However, there was a tendency for the IRA content to increase in longstanding diabetes. Men had a significantly higher pancreatic IRA content than women. The four subjects with very high IRA levels (>lo kg/g) were all elderly men with a long duration of diabetes. Thus, although the pancreatic amylin content was increased in NIDDM, no significant relationship to the clinical features of the disease was found.

Abstract. The amounts of immunoreactive proinsulin (IRP), immunoreactive insulin (IRI), and C-peptidelike immunoreactivity (CPR) in six insulinomas and one nesidioblastosis lesion were determined together with those in the surrounding pancreatic tissue. Four non-insulinoma and nondiabetic human pancreases were used as the control. The IRP in the seven tumors ranged from 5.85 µg/g to 65.45 µg/g (mean ± SEM, 28.70±8.01 µg/g), while the IRP in the surrounding pancreatic tissue ranged from 2.08 µg/g to 11.71 µg/g (5.32± 1.76 µg/g). Control pancreases had an IRP content of 12.01±2.36 µg/g. The IRI in the seven tumors ranged from 4.02 U/g to 47.97 U/g (14.40±6.35 U/g), while that in the surrounding pancreatic tissue ranged from 0.28 U/g to 3.64 U/g (2.32±0.63 U/g). Mean tumor CPR was 206.84±81.6 µg/g and it was 29.16±9.15 µg/g in the surrounding pancreatic tissue. The molar ratio of the IRP to IRI content was 6.83± 1.95% for tumor tissue and 6.24±2.18% for the surrounding pancreatic tissue. These levels were similar to the ratio in the control pancreases (7.67± 1.88%), in contrast to the higher serum IRP/IRI ratio in the tumor patients. studied--six with insulinoma and one with nesidioblastosis. Their clinical profiles are shown in Table 1. The diagnosis of insulinoma was confirmed histologically after tumor resection. Four non-insulinoma and non-diabetic autopsy pancreases were obtained within 6 h of death for use as controls. Fasting serum IRP and IRI levels in the seven insulinoma patients and 13 healthy subjects were also determined. Two of these patients were cases 2 and 3, and the other five patients were different from the patients listed in Table 3. To determine the extractable IRP, IRI and CPR, the insulinoma and the surrounding pancreatic tissue were minced and extracted with acidified alcohol [9], and the extract was assayed directly after dilution. Extractable IRI, IRP and CPR were determined immunologically. IRP was measured with a human proinsulin kit (Dainabott, Tokyo, Japan) with a range of 0.05 ng/ml to 3 ng/ml. This assay system was not affected by the addition of human insulin (1 U/ml) or CPR (50 ng/ml). No studies on the cross-reaction with

R e f e re n c e s 1 . Brange J (Ed.): Physical stability. In G a l e nics of Insulin. Berlin, Springer-Verlag, 1987, p. 52-53 2 . Attia N, Jones TW, Holcombe J, Ta m b o rlane WV: Comparison of human re g u l a r and lispro insulins after interruption of continuous subcutaneous insulin infusion and in the treatment of acutely decompensated IDDM. Diabetes Care 21:817-821, 1998 3 . Kappel M, Gyhrs A, Galbo H, Pedersen BK: The response on glucore g u l a t o ry hormones of in vivo whole body hypert h e rmia. Int J Hypert h e rm i a 13:413-421, 1997