Role of hepatocyte plasma membrane in insulin degradation

An important role of the cell membrane in insulin degradation by cultured rat hepatocytes is supported by studies using the surface-active antibiotic bacitracin. Bacitracin inhibited degradation of cell-associated insulin (both randomly and A14 labeled) by 80-90% at 15 degrees C and by 60% at 37 degrees C. At 37 degrees C, inhibition of degradation was observed only with bacitracin present during dissociation and was accompanied by a compensatory increase in release of trichloroacetic acid (TCA)-precipitable insulin. This profile suggests inhibition of insulin degradation on the membrane after either primary binding or diacytosis (endocytosis-reverse endocytosis). In contrast, at 15 degrees C, bacitracin's inhibitory effect was greater with the antibiotic present during association and was not accompanied by a compensatory increase in TCA-precipitable insulin. This profile was compatible with inhibition of partial cleavage of insulin on the membrane. Internalization and degradation through chloroquine-sensitive pathways may be required to complete degradation at this temperature because chloroquine exhibited an inhibitory effect on insulin degradation equally potent to that of bacitracin at 15 degrees C (no effect at 37 degrees C).

The neuroendocrine and behavioral response to dextroamphetamine in normal individuals

Dextroamphetamine 30 mg and placebo were administered by mouth in a double-blind randomized cross-over trial to ten subjects. Behavior was assessed and blood samples analyzed for growth hormone (GH), prolactin (PRL), homovanillic acid (HVA), and amphetamine. There was a statistically significant increase in well-being following d-amphetamine as compared to baseline and placebo on both the Amphetamine Interview Rating Scale and the Hopkins Mood Scale. No subject became psychotic. There was a statistically significant increase in GH from baseline to peak following d-amphetamine ingestion. When compared to placebo, the increase in GH was invariable and statistically significant for the 90-min sampling. PRL decreased from baseline following both d-amphetamine and placebo and there was no significant drug-placebo difference. Serum HVA was measured at baseline and 120 min, for eight subjects. Six subjects had an increase in HVA following d-amphetamine but there was no significant drug effect.

Insulin processing and dissociation--effect of temperature and lysosomotropic agents

Dissociation of intact and degraded insulin from hepatocytes in monolayer culture was examined under conditions in which processing of insulin was altered by either temperature or pharmacologic agents. Conditions which increased insulin degradation or processing decreased equilibrium insulin binding whereas those conditions which inhibited processing increased equilibrium binding. The effect of lysosomotropic agents on processing was markedly temperature dependent. Not only was processing increased at higher temperatures (37 degrees C, but the effect of lysosomal inhibitors (chloroquine and methylamine) on insulin processing was abolished at this temperature. The temperature dependency of this effect may explain discrepancies between laboratories on the effect of these inhibitors in hepatocytes.

Failure of a midnocturnal insulin infusion to suppress the increased insulin need for breakfast in insulin-dependent diabetic patients

Insulin requirements for meals were measured in eight insulin-dependent diabetic patients, using a closed-loop insulin infusion system. Patients required more insulin for breakfast than for an isocaloric lunch (35.7 +/- 5.5 mU/kcal/3 h versus 26.9 +/- 5.1 mU/kcal/3 h, P less than 0.02) or an isocaloric supper (35.7 +/- 5.5 mU/kcal/3 h versus 26.6 +/- 6.6 mU/kcal/3 h, P = 0.05). To determine whether this insulin resistance at breakfast might be due to low basal insulin levels overnight, the insulin needs for breakfast were compared after an overnight fast (day 1) and after a midnocturnal (0200 h-0500 h) insulin infusion (day 2). Breakfast insulin requirements were similar on both days (35.7 +/- 5.5 mU/kcal/3 h versus 37.7 +/- 5.1 mU/kcal/3 h, P = NS). Whereas nonobese diabetic patients required approximately 60% more insulin for breakfast than for other meals, obese diabetic patients in this study did not demonstrate insulin resistance at breakfast. These findings provide a basis for the common clinical practice of allocating more insulin for breakfast than for other meals. The absence of an increased insulin need at breakfast in our obese patients cautions against a similar algorithm for obese diabetic patients. We postulate that growth hormone may be a cause for morning insulin resistance.

Role of insulin receptors in obesity-related diabetes

Insulin binding to monocytes was assessed before and after plasma insulin suppression by diazoxide in 14 obesity-related diabetic subjects. Four of the five patients with mild carbohydrate intolerance (FBS less than 150 mg%) and hyperinsulinism exhibited low monocyte insulin binding. Despite an increase in insulin binding after 7 days of diazoxide therapy, no improvement in carbohydrate tolerance could be demonstrated. Lack of improvement may have been related to persistent diazoxide effect. An additional group of 4 patients with low plasma insulin values and more severe carbohydrate intolerance (FBS greater than 150 mg%) had high monocyte insulin binding. This group, as well as a group of patients with intermediate insulin responses, tolerated diazoxide poorly and developed moderate ketonuria or severe hyperglycemia (plasma glucose greater than 350 mg%) necessitating discontinuation of the drug after 3-6 days. The studies in these patients suggest that obesity-related diabetes may be characterized early by mild elevation of plasma glucose, hyperinsulinism and impaired monocyte insulin binding. As beta cell exhaustion occurs, more severe hyperglycemia intervenes and insulin binding to monocytes increases.

A cyproheptadine-reversible defect in ACTH control persisting after removal of the pituitary tumor in Cushing's disease

We studied two phases of cortisol feedback suppression of ACTH in nine patients who had had adrenalectomy for Cushing's disease. Four had been treated by adrenalectomy alone and presumably had ACTH-secreting pituitary tumors. Five others were studied two or more years after transsphenoidal removal of an ACTH-secreting microadenoma. In both groups, cortisol-ACTH feedback during the first 30 minutes of cortisol infusion was abnormal; plasma ACTH fell only 2.7 +/- 2.6 per cent (mean +/- S.E.), as compared with 28.0 +/0 10.1 per cent in five hypoadrenal controls (p less than 0.01). The fall in ACTH during the second phase of cortisol infusion was similar in the patients and the controls. Cyproheptadine corrected the feedback abnormality occurring during the first phase in both groups of patients with Cushing's disease; ACTH fell by 24.4 +/- 4.8 per cent (P less than 0.005). Persistence of a cortisol-ACTH feedback abnormally after removal of the pituitary tumor in Cushing's disease, as well as the correction by cyproheptadine, suggests that higher centers have a role in the pathophysiology of Cushing's disease.

Effect of Anti-Insulin Receptor Antibody on Insulin Dissociation from IM-9 lymphocytes

Although anti-insulin receptor antibodies mimic insulin in many of its actions, antireceptor antibodies in serum from a diabetic patient did not share the property of insulin to produce negative cooperativity as defined by the Demeyts experiment. In contrast, an inhibition of insulin dissociation from IM-9 lymphoblastoid cells was observed after exposure to the antireceptor antibody. The capacity of this particular antireceptor antibody to attach to cells whose insulin receptors have been saturated suggests that the antibody may trap insulin by molecular hindrance.

Effects of thyrotropin-releasing hormone and metoclopramide in patients with phenothiazine-induced hyperprolactinemia

TRH and metoclopramide tests were performed in 10 female patients with presumed phenothiazine-induced hyperprolactinemia to define the serum PRL response to these agents. Our results show that the serum PRL response to metoclopramide is blunted in most patients with phenothiazine-induced hyperprolactinemia, and the serum PRL response to TRH is exaggerated in most patients during and 3 weeks after stopping phenothiazines.

Galactorrhea-amenorrhea syndrome: follow-up of forty-five patients after pituitary tumor removal

Forty-five patients with galactorrhea-amenorrhea were followed during a period of 1 to 8 years (mean 3.1) after transsphenoidal prolactinoma removal. The ratios of patients who appear to be cured to the total numbers treated were 20 patients of 27 with grade I tumors; six of 10 with grade II; two of five with grade III; and none with grade IV tumors. Six patients with normal prolactin levels one week postoperatively had relapse later, as did three with normal prolactin levels 2 months postoperatively. A normal prolactin level 6 months postoperatively predicted ultimate cure. The 19 pregnancies that occurred in 15 patients, four with high prolactin levels, were uneventful. Prolactin rose normally with pregnancy and returned to prepregnancy level in all but one patient. Prolactin responses to stimulation tests were blunted for 6 months after successful tumor removal. By 1 year, responses to thyrotropin releasing hormone and metoclopramide tests were returning to normal, although responses to chlorpromazine and hypoglycemia remained blunted. The postoperative inhibition of normal lactotropes for 6 months is suggested. Ultimate cure cannot be determined before 6 months and conception should be deferred until then.

Effect of down-regulation of insulin receptors on receptor-antibody binding

The extraction of insulin receptor antibody (ARAB) by IM-9 lymphoblastoid cells was measured using nonisotopic ARAB, ARAB extraction was expressed as a clearance, with ARAB determinations made indirectly by a method dependent on its inhibition of insulin binding. Control cells and down-regulated IM-9 cells (cells exposed to 10(-7) M insulin for 16 h) at a cell density of 8 X 10(7) cells/ml were incubated in the presence of a 1:75 dilution of serum from a patient with insulin resistance due to insulin receptor antibodies. After 60 min at room temperature, control cells cleared 42 +/- 3% of ARAB compared to only 18 +/- 2% by the down-regulated cells. ARAB clearance was directly proportional to maximal tracer insulin binding, as determined in parallel experiments. ARAB clearance was similarly reduced in a dose-related fashion by 16-h exposure to Concanavalin A. These results indicate that ARAB extraction by IM-9 cells is reduced by chronic exposure to insulin. The observation that ARAB binding to insulin down-regulated cells is reduced despite the ability of ARAB to attach normally to cells whose receptors have been saturated with insulin supports the notion that down-regulation is due to a loss of membrane receptor rather than to simple receptor occupancy. (Endocrinology 108: 478, 1981)

Persistent defect in regulation of prolactin secretion after successful pituitary tumor removal in women with the galactorrhea-amenorrhea syndrome

TRH, metoclopramide (MCP), chlorpromazine (CPZ), and insulin (ITT) stimulation tests of PRL secretion were carried out in age-matched controls and before and after successful removal of pituitary prolactinomas in women with the galactorrhea-amenorrhea syndrome. In preoperative patients there was a blunted or absent PRL response to TRH in 87%, to MCP in 100%, to CPZ in 100%, and to ITT in 93%. Two to 6 months after successful tumor removal, serum PRL rose 2-fold (the usual criterion for a normal response) in 73% after TRH, in 100% after MCP, but in only 13% after CPZ and in only 14% on ITT. However, the PRL increment with all four tests was significantly lower than that in normal controls. One to 8 yr after successful surgery, the PRL increments after TRH and MCP were returning to normal, but the PRL responses to CPZ and ITT remained blunted. GH, ACTH, and TSH reserves were intact in all patients. The diminished PRL response to all stimulation tests observed up to 6 months postoperatively might be explained by the persistence of a negative feedback effect from high PRL levels associated with the tumor. The more persistent impairment of the PRL response to CPZ and ITT is unexplained but suggests a hypothalamic defect.

Inhibition of insulin binding by concanavalin-A

Prolonged incubation of IM-9 lymphoblastoid cells with concanavalin-A (Con-A) reduces insulin binding at concentrations of the lectin that neither agglutinate cells nor acutely inhibit insulin binding. Scatchard plot analysis suggested that Con-A treatment reduces insulin receptor number. Con-A-induced reduction in insulin binding resembles insulin "down regulation" in the time required for appearance and reversal of the binding alteration. The effect of Con-A on insulin binding was inhibited when the sugar alpha-methyl mannoside was incorporated into the incubation medium at the same time as Con-A. Even when the sugar was added after hours of Con-A exposure, the depressed insulin binding was reversed proportional to the reduction of Con-A binding to the cells by the sugar. This latter observation suggests steric hindrance as the mechanism for Con-A-induced reduction in insulin binding. The slow association rate of Con-A to the cells under the conditions of the experiments probably explains the prolonged exposure time to Con-A required for inhibition of insulin binding. Additional support for steric hindrance was observed in experiments in which insulin was "trapped" on cells after exposure to Con-A. Bivalent succinyl Con-A was only 10% as effective as the natural tetravalent lectin in reducing insulin binding. Cytoskeleton and protein synthesis inhibitors failed to alter the Con-A-induced reduction in insulin binding.

The treatment of acromegaly by transsphenoidal surgery

Pituitary tumors were removed transsphenoidally in 32 patients with acromegaly. Ten patients had high prolactin (PL) levels as well as elevated growth hormone (GH) levels. In 24 patients, GH levels were lowered to 5 ng/mL or less and remained low. The PL level was reduced to normal in eight patients. Of three patients with postoperative GH levels of 6 to 10 ng/mL, one patient remains in this range but two have relapsed. Five patients showed only a partial lowering in GH level. Pituitary irradiation lowered the level further. Hypopituitarism developed in four patients. Permanent diabetes insipidus occurred in one patient and meningitis developed in another. They subsequently recovered. There were no deaths. Abnormal GH responses to hyperglycemia, hypoglycemia, and levodopa returned to normal following surgery. Transsphenoidal tumor removal appears to be an effective treatment for acromegaly.

Central nervous system insulin receptors in normal and diabetic rats

The presence of specific insulin receptors in homogenates of various parts of the rat brain was demonstrated. Hypothalamic preparations exhibited the greatest insulin binding. Scatchard plots of the binding data were typically curvilinear. Insulin binding to the brain homogenates was influenced by pH, incubation time, temperature, and ionic milieu. The specificity of insulin binding to the brain homogenates was demonstrated by partial displacement of labeled insulin by insulin analogs according to their biological activities and by inhibition of insulin binding by serum containing an antibody to insulin receptors. Insulin binding to brain homogenates from streptozotocin-induced diabetic rats was similar to that from control animals and failed to indicate up-regulation of these receptors.

Characterization of circulating insulin and proinsulin-binding antibodies in autoimmune hypoglycemia

Five patients with fasting and(or) postprandial hypoglycemia were found to have insulin antibodies in the absence of previously documented immunization. Studies on the equilibrium-binding of insulin to the autoantibodies revealed two classes of binding sites with association constants and binding capacities analogous to those of insulin antibodies from insulin-treated diabetic patients. Similarly, no consistent differences in these parameters were found in both groups of patients with insulins of bovine, porcine, and human origin. Proinsulin (C-segment directed) antibodies capable of binding bovine or porcine proinsulin were present in 10 of 10 and 9 of 10 insulin-treated diabetics serving as controls, respectively, and, when present, provide incontrovertible evidence of exogenous insulin administration. No such antibodies could be detected in the hypoglycemic patients with autoimmune insulin antibodies. The kinetics of dissociation of the insulin-antibody complexes were consistent with the existence of two classes of antibody sites. The corresponding dissociation rate constants were large enough to predict that significant amounts of free hormone may be generated by this mechanism and provide a plausible pathogenesis for the hypoglycemia in these patients.

Downregulation of insulin receptors in obese man

In an attempt to determine whether the decreased number of insulin's receptors in obesity is a result of downregulation of the receptors, diazoxide (5 mg/kg/d) was given to 10 obese subjects. Insulin's suppression by diazoxide in these 10 subjects resulted in a mild glucose intolerance and an increase in insulin's receptors in seven of the 10 subjects. The subjects could be divided into three groups by analyzing the Scatchard plots of their insulin receptor studies before and after diazoxide. Four subjects exhibited an increase in both high affinity and low affinity receptors, three showed an increase only in high affinity receptors, and three failed to demonstrate any change in receptors in response to diazoxide. These studies support the concept that the decreased number of insulin's receptors observed in obesity is a result of the downregulation of the receptors and is not the primary, underlying cause of insulin resistance in obesity, although a contributory role cannot be ruled out.

Effect of dietary composition on insulin receptors in normal subjects

Six normal subjects were placed on a high carbohydrate diet (80%) and a high fat diet (60%) for 2 weeks each. Glucose tolerance testing with plasma immunoreactive insulin levels was performed along with insulin receptor quantitation after a control period and after each of the dietary manipulations. Despite improved carbohydrate tolerance and decreased plasma immuno-reactive insulin after the high carbohydrate diet (evidence for increased insulin sensitivity) insulin receptor number and affinity were unchanged. These studies suggest that the increased insulin sensitivity induced by a high carbohydrate diet is due to some adaptive change in postreceptor activity. Manipulations of dietary composition fail to alter insulin binding to peripheral mononuclear cells.

Down regulation of insulin receptors in primary cultures of adult rat hepatocytes in monolayer

Down regulation of insulin receptor number by insulin at physiological concentrating (10(-9) M) has been demonstrated in primary cultures of adult rat liver parenchymal cells in monolayer. However, because insulin is rapidly degraded in this culture system, special efforts are necessary to maintain the concentration of insulin in the medium in order to evoke the down regulation phenomenon.

Diabetes and hypoglycemia due to insulin antibodies

A patient with "autoimmune diabetes" with reactive hypoglycemia due to anti-insulin antibodies is described. The pathophysiology of the impaired carbohydrate tolerance and reactive hypoglycemia is illustrated by measurements of C-peptide, and free and antibody-bound insulin. Detailed analysis of her antibodies did not provide any evidence in favor of exogenous insulin immunization. Nevertheless, a possible cause of unsuspected insulin immunization was present in her history which should be considered in other reported cases of "autoimmune diabetes" due to insulin antibodies in which exposure to insulin cannot be documented.

Effect of lithium on pancreatic islet insulin release

Lithium exerts an inhibitory effect on glucose and amino acid-induced insulin release. The inhibitory effect of Li+ persists even in subsequent Li+-free conditions, indicating an only slowly reversible effect. Lithium fails to inhibit glucagon-induced insulin release. The exact mechanism of the lithium effect is as yet undetermined, but interference with calcium flux and/or microtubular function is an attractive hypothesis. The inhibitory effect of lithium on insulin release cannot be reversed by alteration in ionic (Ca++, Mg++, K+) concentrations in the incubation media. Studies involving the effect of low sodium on insulin release in which lithium had been used as an osmotic replacement for sodium must be carefully reassessed because of these findings.

Hyperinsulinism in endotoxin shock dogs

Extreme hyperinsulinism was observed in endotoxin-shock dogs made hyperglycemic by glucose infusion. Qualitatively (at least in terms of gel filtration characteristics), the insulin secreted under these conditions was normal. Hyperinsulinism was not observed in endotoxin-shock dogs not given glucose. Thus hyperinsulinism does not explain the hypoglycemia so frequently observed in endotoxin-treated dogs. Hyperinsulinsm could not be impaired degradation of insulin as disappearance of labeled insulin as well as cold insulin was comparable in control and endotoxin-treated animals. An adrenergic mechanism (either beta receptor stimulation or postadrenergic hyperresponsiveness of the beta cells) probably does not explain the hyperinsulinism observed in endotoxin-shock dogs given glucose as beta blockade failed to inhibit the hyperinsulinsm. Hyperinsulinism was not observed in endotoxin-shock dogs given tolbuamide. A tenfold rise in plasma IRG was observed in endotoxin-treated dogs whether glucose was infused or not. The persistently low IRI levels in endotoxin-treated dogs not given glucose suggest that hyperresponsiveness of the beta cell to glucagon was not present in these animals. Extreme hyperinsulinsm in response to induced hyperglycemia in endotoxin-shock dogs is unexplained. Hyperresponsiveness of the beta cell to glucose during endotoxin shock seems likely.