Responses of the pancreatic A cell during hypoglycemia and hyperglycemia are dependent on the B cell.

It is controversial whether stimulation of glucagon secretion by hypoglycemia or suppression by hyperglycemia is a direct effect of glucose on the A cell or whether it is mediated indirectly through the B cell. To determine the role of the B cell in the mediation of glucagon secretion adolescent male baboons were studied before and after successive injections of streptozocin designed to cause B-cell destruction in a series of stages. Following one dose of streptozocin, B-cell function was impaired but fasting blood glucose remained normal. B-cell function declined further with additional doses of streptozocin. As B-cell function declined, there was a corresponding reduction in the glucagon response to hypoglycemia (Ghyp). There were significant correlations between the percentage reduction in Ghyp and the percentage reduction in B-cell function (acute insulin response to arginine, r = .47; acute insulin response to glucose, r = .69; Kg, r = .79). In a second study the ability of hyperglycemia to suppress the acute glucagon response to arginine (AGRarg) was studied. Before streptozocin AGRarg was 128 +/- 26 pg/mL at a glucose level of 80 +/- 4 mg/dL and was suppressed to 74 +/- 20 pg/mL when glucose was raised and clamped at 209 +/- 14 mg/dL. After the initial dose of streptozocin, with mild B cell damage and fasting normoglycemia, AGRarg was not suppressed by hyperglycemia. With severe B cell dysfunction and fasting hyperglycemia, there was paradoxical enhancement of AGRarg by additional hyperglycemia. In conclusion, the ability of the A cell to respond appropriately to hypoglycemia and to arginine during hyperglycemia is dependent on normal B-cell function.

Defects in beta-cell function and insulin sensitivity in normoglycemic streptozocin-treated baboons: a model of preclinical insulin-dependent diabetes.

During the preclinical period of human insulin-dependent diabetes, both impaired pancreatic beta-cell function and increased insulin resistance are found, although normoglycemia is preserved. To better understand the changes in beta-cell function and insulin sensitivity that occur in preclinical insulin-dependent diabetes, we performed a panel of in vivo beta-cell function tests and measured insulin sensitivity in adolescent male baboons both in normal health and after a small dose of streptozocin which did not induce hyperglycemia. Nine animals were studied before (stage 1) and 1 week after receiving a low dose of streptozocin (stage 2). There was no change in fasting plasma glucose or insulin. The mean glucose disposal rate (Kg) remained within the normal range, but dropped from 2.0 +/- 0.2% +/- SE) to 1.2 +/- 0.1%/min (P less than 0.01), the acute insulin response to arginine (AIR(arg)) fell from 67.7 +/- 19.4 microU/mL (485.8 +/- 139.2 pmol/L) to 32.8 +/- 7.2 microU/mL (235.3 +/- 51.7 pmol/L; P less than 0.05), and the acute insulin response to glucose (AIR(gluc)) fell from 881 +/- 243 microU/mL.10 min (6321 +/- 1744 pmol/L.10 min) to 334 +/- 82 microU/mL.10 min (2396 +/- 588 pmol/L.10 min; P less than 0.01). The most dramatic change, however, was in the ability of hyperglycemia to potentiate AIR(arg) (expressed as the slope of potentiation). This was reduced by 94% from 1.8 +/- 0.5 to 0.1 +/- 0.1 (P less than 0.01), with almost no overlap in values between stages 1 and 2. Insulin sensitivity was also lower 1 week after streptozocin treatment. When the animals were restudied 8 weeks after streptozocin treatment (stage 3) most measures of beta-cell function were not significantly different from those in stage 1. The fasting plasma glucose level was 85.4 +/- 4.3 mg/dL (4.7 +/- 0.2 mmol/L), Kg was 1.8 +/- 0.3%/min, fasting plasma insulin was 35.9 +/- 8.5 microU/mL (257.6 +/- 61.0 pmol/L), AIR(arg) was 67.0 +/- 15.4 microU/mL (480.7 +/- 110.5 pmol/L), and AIR(gluc) was 615.3 +/- 265.3 microU/mL.10 min (4413 +/- 1901 pmol/L.10 min), and tissue insulin sensitivity was 2.7 +/- 0.4 x 10(4) min/microU.mL. These values show extensive overlap with those of stage 1, from which they are not significantly different. The slope of glucose potentiation, however, remained low in all animals at stage 3.(ABSTRACT TRUNCATED AT 400 WORDS)

Glucose counterregulation during recovery from neuroglucopenia: which mechanism is important?

Neuroglucopenia (NGP), which is a serious potential hazard for all insulin-treated diabetics, stimulates many neural and hormonal responses including increased glucagon secretion and activation of beta-adrenergic receptors of the autonomic nervous system. To determine which of these responses is important in recovery from NGP, we induced NGP in baboons by the intravenous (IV) injection of 2-deoxy-D-glucose with and without beta-adrenergic blockade (propranolol) and somatostatin. Thirty minutes after the induction of NGP the animals recovered, and the mean (+/- SEM) rise in arterial plasma glucose was 6.6 +/- 0.9 mmol/L, in glycerol 0.106 +/- 0.22 mmol/L, and in beta-hydroxybutyrate 0.091 +/- 0.22 mmol/L. Animal recovery and glucose rise were uninfluenced by the infusion of propranolol (mean 30 minute plasma glucose rise of 6.2 +/- 0.8 mmol/L) and somatostatin (6.8 +/- 0.8 mmol/L). However, the combined infusion of somatostatin and propranolol prevented animal recovery and glucose rise (1.0 +/- 0.1 mmol/L). The glycerol and beta-hydroxybutyrate rises were blocked by the propranolol infusion alone. Thus, recovery from NGP and the associated rise in plasma glucose, glycerol, and beta-hydroxybutyrate are prevented by the combination of the suppression of the glucagon and beta-adrenergic response to NGP. Furthermore, if the results of our study are extrapolated to insulin-dependent diabetic patients, most of whom have an impaired glucagon response to insulin-induced hypoglycemia/neuroglucopenia, they would be critically dependent on beta-adrenergic mechanisms for recovery from NGP.

The antibody response to insulin therapy. A prospective study in HLA-typed insulin-dependent diabetic subjects.

There is heterogeneity within insulin-dependent diabetes mellitus (IDDM), and it has been suggested that the presence of the HLA-DR specificities DR3 and DR4 define two subsets of IDDM with clear differences in their immune response to therapeutic insulin. To test this hypothesis, we have prospectively studies the development of insulin binding antibody (IBA) in 54 subjects with newly diagnosed, classical childhood IDDM, determined seven binding constants of their IBA, and measured the presence or absence of pancreatic polypeptide-binding antibodies after 1 yr of therapy with insulin. There were no relationships between insulin and pancreatic polypeptide antibodies and the DR3 or DR4 specificities whether these specificities were tested for alone or in combination, comparing the presence and absence of DR3 and DR4 and comparing DR3 with DR4, except that of the 33% of all subjects who developed antibodies binding pancreatic polypeptide by 1 yr, none possessed the DR3 specificity alone (P = 0.018). Thus, the hypothesis that the HLA-DR3 and -DR4 specificities are major determinants of IBA formation and, therefore, define important subsets of childhood IDDM in terms of immune response to therapeutic insulin is not substantiated by this study.

Evidence for noncholinergic ganglionic neural stimulation of B cell secretion.

Insulin levels increase after 2-deoxyglucose (2DG) administration in dogs. This observation is in contrast to the decrease in insulin level post-2DG in baboons and rabbits. To evaluate a possible neural mechanism mediating this increase in insulin level, we studied normal mongrel dogs with 2DG alone, 2DG during ganglionic blockade, beta-adrenergic blockade, and postganglionic parasympathetic blockade. There was an increase in plasma epinephrine, norepinephrine, pancreatic polypeptide, insulin, and glucose post-2DG alone. During ganglionic blockade, the increase in epinephrine, norepinephrine, and pancreatic polypeptide post-2DG was completely abolished, verifying ganglionic blockade of sympathetic and parasympathetic pathways, respectively. Despite this, ganglionic blockade failed to abolish the insulin rise after 2DG. Postganglionic parasympathetic blockade did not change the insulin rise after 2DG. However, beta-adrenergic blockade completely abolished the insulin rise after 2DG. The above data suggests that 1) the insulin rise post-2DG is beta-adrenergic but 2) the ganglionic neurotransmitter mediating the 2DG-induced insulin rise during ganglionic blockade is noncholinergic (possibly peptidergic).

HLA-related heterogeneity in seasonal patterns of diagnosis in Type 1 (insulin-dependent) diabetes.

HLA type, time of year of diagnosis, and age at diagnosis were studied in 52 new cases of Type 1 diabetes in Seattle, Washington. Diagnosis was found to be seasonal in diabetic patients positive for DR3 (p less than 0.005), with the expected marked reduction in new cases during the summer months. This seasonality was not age-related (p greater than 0.13). Cases who were DR3-negative did not show significant seasonality of diagnosis (p greater than 0.5). However, when age at diagnosis was adjusted for, a seasonal effect was found in the DR3-negative group (p less than 0.006), with older cases favouring a spring onset and younger cases favouring an autumn onset. Thus, DR3-positive cases showed a seasonal diagnosis pattern that did not depend on age, while DR3-negative cases showed an age-dependent seasonal pattern. These differences may reflect the predominance of different aetiological mechanisms in these two genetic groups.

Insulin antibodies in insulin-dependent diabetics before insulin treatment.

A sensitive assay was used to measure the binding of iodine-125-labeled insulin in serum obtained from 112 newly diagnosed insulin-dependent diabetics before insulin treatment was initiated. Two groups of nondiabetics served as controls: children with a variety of diseases other than diabetes and nondiabetic siblings of insulin-dependent diabetics. Eighteen of the diabetics were found to have elevated binding and 36 were above the 95th percentile of control values. The insulin-binding protein is precipitated by antibody to human immunoglobulin G, has a displacement curve that is parallel and over the same concentration range as serum from long-standing insulin-dependent diabetics, and elutes from a Sephacryl S-300 column at the position of gamma globulin. These insulin antibodies are present in a large percentage of newly diagnosed, untreated diabetics and may be an immune marker of B-cell damage.