Insulin inhibition of calcium binding by human placental membrane.

Insulin and its analogues displaced membrane-bound calcium within a physiological range of insulin concentration, in proportion to both biological potency and ability to displace porcine 125 I-labelled insulin from the insulin receptor. Mild tryptic digestion of the membrane reduced insulin binding but did not reduce specific calcium binding. Displacement of membrane-bound calcium by insulin was dependent on insulin binding to its intact receptor. These studies suggest that Ca2"ay exert a controlling influence on insulin-receptor binding in vivo.

Purification of the insulin receptor from human placental membranes.

Insulin receptors were purified from human placental microsomal membranes by solubilisation with Triton X-100 followed by Sepharose 6B chromatography, phosphate gradient elution from hydroxyapatite and affinity chromatography on concanavalin A-Sepharose. 2000-fold purification was achieved with 63% overall recovery. The purified receptor gave a single band on 3.75% polyacrylamide (0.1% Triton X-100) gel electrophoresis. On sodium dodecyl sulphate-polyacrylamide gel electrophoresis there was a major band at 75,000 and a minor band at 80,000 daltons. The purified receptor rechromatographed on Sepharose 6B with an apparent molecular weight of 300,000.

Control of insulin receptor affinity by a Ca2+-sensitive binding site.

Calcium (Ca2+) increased insulin-receptor binding in both membrane and solubilised receptor preparations. Ca2+ increased both receptor affinity and initial rate of association of [125I]insulin to the receptor preparations. Ca2+ had no effect on insulin receptor number in either receptor preparation. The effect of Ca2+ on affinity could be mimicked by ions with similar ionic radii and properties (e.g., Ba2+, Mg2+ and Sr2+). EDTA and oleic acid reduced insulin binding and receptor affinity and these effects were reversed by the addition of Ca2+. These studies suggest that Ca2+ and Ca2+-like ions may bind to a site on or near the receptor and may be responsible for a conformational change with a consequent increase in receptor affinity.

The dissociation of insulin from human insulin antibodies.

The dissociation of insulin from human insulin antibodies has been investigated using a technique that is rapid and does not require addition of excess unlabelled insulin. A slow (k1 = 2-10(-3) min-1) and a fast k2 = 4-10(-2) min-1) dissociating antibody component were identified in all studies. These have been shown to correspond, respectively, to the high and low affinity antibody components of equilibrium binding studies. The range of k1 and k2 values and their response to temperature change is small. Insulin resistance and stability of diabetes are not related to properties of antibody dissociation. Dissociation is faster in the presence of high (6-850 nM) insulin concentration due to increased binding to the fast dissociating component without change in the dissociation rate constants. When incubation time is increased beyond achievement of maximal binding there is a time-depent rise in binding to the slow dissociating component, with a concomitant fall in k1. The traditional concept that equilibrium is established at maximum binding requires further examination.

Terbium, a fluorescent probe for insulin receptor binding. Evidence for a conformational change in the receptor protein due to insulin binding.

Terbium (Tb3+), an ion of the lanthanide series that has been used as a fluorescent probe for calcium (Ca2+) binding sites in proteins, binds to the proteins in both solubilized and purified human placental insulin receptor preparations. Tb3+ fluorescence was determined directly and the effect of insulin binding on Tb3+-enhanced fluorescence was studied without the need to separate bound and free ligands. Tb3+ behaved similarly to, but was more potent (50-fold) than, Ca2+ in increasing the insulin bound to its receptor. When insulin bound to its receptor, the Tb3+ fluorescence of the receptor preparation decreased. When various insulin analogues were tested, the decrease in Tb3+ fluorescence was proportional to the biologic activity of the insulin analogues. In addition, Tb3+ could be displaced from insulin-sensitive sites by Ca2+, indicating that there were Ca2+ (and Tb3+) binding sites on or near the insulin receptor. These sites, when filled, were responsible for the increased insulin receptor affinity. The decrease in Tb3+ fluorescence after insulin binding may be indicative of a conformational change in the insulin receptor precipitated by the binding of insulin. This conformational change may be related to the release of Ca2+ by insulin binding, is associated with a decrease in insulin receptor affinity, and suggests that an allosteric mechanism involving both Ca2+ and insulin binding sites may be responsible for the observed changes in insulin receptor affinity.

Clinical trial of an aldose reductase inhibitor in diabetic neuropathy.

A single-blind, nonrandomized, placebo crossover clinical trial of an aldose reductase inhibitor, Alrestatin (AY 22, 284) was performed over a 4-mo period in nine patients with diabetic peripheral neuropathy. Most patients had subjective benefit, but objective measures of conduction were essentially unchanged. Substantial toxicity was evident, particularly photosensitive skin rash.

Antigenicity of "monocomponent" pork insulin in diabetic subjects.

"Single-peak," "single-component," and "monocomponent" insulins have been produced in an attempt to eliminate insulin antigenicity. Recently "single-peak insulin" has been shown to be antigenic. From animal experiments and preliminary human studies it has been claimed that monocomponent (MC) insulin is nonantigenic or only negligibly so. In this study the antigenicity of MC insulin was determined in two groups of diabetic patients. In group 1, seven patients treated with insulin for the first time were given MC insulin for seven to fifteen months. Four of the seven patients developed significant IgG insulin antibodies after four to ten months. In one patient the IgG insulin antibody concentration was high (8.51 mU./ml.). In two patients, IgG proinsulin-specific antibodies were detected. In group 2, fourteen patients with unstable diabetes, insulin allergy, or resistance were changed from conventional to MC insulin. Treatment with MC insulin did not decrease insulin requirement or improve diabetic control when assayed by the M factor. After seven to eleven months of therapy there was no significant fall in insulin antibodies except in two patients in whom corticosteroids had been administered simultaneously. These results differ significantly from those previously reported and could be interpreted as suggesting that insulin itself is antigenic. When the purity of the MC insulin was determined, significant contaminants could be demonstrated in all of ten separate batches of MC insulin. Gel chromatography, polyacrylamide gel electrophoresis, and proinsulin radioimmunoassay were used to identify the presence of nonconvertible insulin dimer, proinsulin, and monodesamido insulin in antigenically significant concentrations. The generation of IgG insulin antibodies in MC-insulin-treated patients cannot be interpreted as a true indication that insulin itself is antigenic. The problem of insulin antigenicity has not been resolved and will not be until a highly purified insulin is available. Unfortunately, the MC insulins do not meet these requirements.

Growth and hormonal content of human fetal pancreas passaged in athymic mice.

While the usefulness of the human fetal pancreas transplanted into diabetic humans has yet to be realized, its transplantation into nude mice has revealed some of its potential. In this animal the organ grows in size, during which time differentiation of its endocrine component and maturation of the insulinogenic response to glucose occurs. The results reported in this article expand these results by providing data on the weight and both the insulin and glucagon content of these passaged organs. Human fetal pancreata of gestational age 14-19 wk were implanted subcutaneously (s.c.) in nude mice and maintained in vivo for 5-54 wk (absolute age 19-68 wk). The implants were then removed and their insulin and glucagon content determined. Both the weight of these implants and their insulin content were positively correlated with the absolute age of the tissue (weight: r = 0.61, P less than 0.001; insulin content: r = 0.62, P = 0.0003). The glucagon content bore no relationship to the age. The maximum level of insulin extracted from 30 passaged human fetal pancreata was 0.8 U, a level far below the daily insulin requirements in adult humans. It is suggested that an explanation for this is the site of transplantation used. To compare the s.c. site with the renal subcapsular space, the explants of four fetal pancreata were evenly divided between these two sites. After 11-13 wk, the implants were removed. Those beneath the renal capsule were larger and contained a greater amount of insulin and glucagon than those transplanted s.c.(ABSTRACT TRUNCATED AT 250 WORDS)

The measurement of glycosylated hemoglobin in man and animals by aminophenylboronic acid affinity chromatography.

Glycosylated hemoglobin (GHb) was estimated in normal and diabetic human, rat, and dog hemolysate by m-aminophenylboronic acid (PBA) affinity chromatography and the results compared with the values determined using two ion-exchange (ION-E) methods and a colorimetric thiobarbituric acid (TBA) method. There was a good correlation between the values estimated by PBA and both ION-E chromatography methods for the human samples (r = 0.83, P less than 0.0002, r = 0.86, P less than 0.002). In diabetic rat and dog hemolysates, PBA chromatography demonstrated higher glycosylated hemoglobin than in normal hemolysate. In both species, there was an excellent correlation between the PBA-estimated GHb and plasma glucose levels (rat r = 0.79, P less than 0.001; dog r = 0.67, P less than 0.003). The ION-E and TBA methods were not as effective in separating diabetic from normal samples and correlated less well with plasma glucose levels. PBA chromatography relies on the interaction of m-amino phenylboronate with the hydroxyl groups of the glucose residues attached to hemoglobin. It is not affected by intra- or interspecies variations in the hemoglobin moiety and should be adaptable for measurement of GHb in a number of laboratory animals and in patients with hemoglobinopathy. It is not affected significantly by temperature and may offer advantages over the ION-E method in the routine determination of human glycosylated hemoglobin.

Histologic differentiation of human fetal pancreatic explants transplanted into nude mice.

The transplantation of human fetal pancreas has been suggested as a means of treatment of insulin-dependent diabetes in man. We have obtained human fetal pancreata during the second trimester of pregnancy and transplanted 1-mm3 explants subcutaneously (s.c.) into both diabetic and nondiabetic nude mice, some of the tissue being cultured in vitro before implantation. These implants coalesced and grew. They were removed at intervals up to 37 wk later and showed selective differentiation of endocrine tissue that normally occurs in the fetus and neonate, with formation of bipolar, mantle, and mature islets. There was growth of this endocrine tissue with significantly more islets than in the freshly stained fetal pancreas assuming an average dimension larger than 150 micron, which is the reported mean diameter of a neonatal islet. Duct and fibrous tissue remained viable, but there was no definitive acinar tissue seen. The pancreata uncultured before implantation reached a larger size than that attained by those implants cultured before being transplanted, the difference probably being the amount of ductular and mesenchymal tissue still present. Of those glands cultured before transplantation, the longer the period of culture, the smaller the size the implants reached. Culture beyond 3 wk in vitro made it difficult to macroscopically locate the implant. These data show that, in human fetal pancreas removed from its usual environment, both selective differentiation of the endocrine component and growth of the islets can occur.

High glucose concentration causes a decrease in mesangium degradation. A factor in the pathogenesis of diabetic nephropathy.

Mesangium enlargement is a constant feature of diabetic nephropathy and is likely to be important in the pathogenesis of this diabetic complication. Whether decreased degradation of mesangium plays any role in causing the enlargement is uncertain. We developed a system of preparing radioactively labeled mesangium matrix from mesangial cell cultures to be used as substrates for studies of mesangium degradation. Degradation is commenced by growing mesangial cells on the labeled matrix and monitored by the release of radioactivity into the culture medium. High glucose concentration (30 mM), whether present 1) when the matrix is being made or 2) when the degradation is taking place, reduces the rate of mesangium degradation. The second but not the first of these two phenomena was abolished by aminoguanidine. Phorbol 12-myristate 13-acetate, added in a manner to antagonize the action of protein kinase C, inhibited mesangium degradation and was not able to nullify the effect of high glucose. Thus it appears unlikely that a high glucose concentration inhibits mesangium degradation by increasing mesangial cell protein kinase C activity. We conclude that decreased degradation of mesangium as a result of hyperglycemia may play a role in causing the mesangium enlargement that occurs in diabetic nephropathy.

The effect of aldose reductase inhibition on motor nerve conduction velocity in diabetic rats.

This study examined the effects of an aldose reductase inhibitor (CP 45634, Sorbinil, Pfizer, New York, New York) on the neuropathy of streptozotocin-induced diabetic rats. Sorbinil treatment for 4 wk reduced sciatic nerve sorbitol concentration and improved motor nerve conduction velocity in diabetes of 2-9 mo duration. It remains to be determined whether Sorbinil can prevent chronic diabetic neuropathy.

Perifusion and culture of human fetal pancreas.

Human fetal pancreas has been obtained after the therapeutic termination of pregnancy with prostaglandin F2 alpha. Pancreatic explants were studied in a perifusion system and maintained in organ culture for up to 3 wk. Insulin biosynthesis was stimulated by glucose; however, the incorporation of 3H-L-leucine into proinsulin was surprisingly high (52% after 3 h) vs. insulin (48%), which suggests a possible block in the conversion of proinsulin to insulin. Insulin secretion was stimulated during perifusion with 19.3 mM glucose (1.6 X prestimulation level), 1.5 microM glucagon (2.4), 5mM leucine (2.4), 10 mM arginine (2.7), and 10 mM theophylline (10.0). Insulin biosynthesis and secretion were maintained in organ culture. After 12 days there was a 3.4-fold increase in insulin secretion in the presence of 22 mM glucose compared with 5.5 mM glucose, and the explant insulin content rose in the presence of high glucose levels by 89%. The acute insulin secretory response to 10 mM theophylline was maintained after culture. These studies suggest that human fetal pancreas should be considered as a potential source of donor tissue for pancreas transplantation in human diabetes.

Measurement of insulin binding by erythrocyte ghosts. A method that allows storage without loss of binding characteristics.

The insulin binding of red cell ghosts was studied and found to be similar to insulin binding of the original erythrocyte. Red cell ghosts were stable on storage, and serial samples can be measured in one assay, thus eliminating the problem of interassay variation. Erythrocyte ghosts may be preferable to whole red cells for long-term insulin receptor studies.

Ascorbic acid metabolism and polyol pathway in diabetes.

It has been reported previously that the plasma concentration of ascorbic acid (AA) is reduced in streptozocin-induced diabetic rats and can be normalized by treatment with the aldose reductase inhibitor tolrestat. This study was designed to investigate further the relationship between the polyol pathway and AA metabolism in diabetic rats. Disturbance of AA metabolism was demonstrable after 1 wk of diabetes. Dietary myo-inositol supplementation was effective in normalizing plasma AA levels, as was treatment with tolrestat. In untreated diabetes, despite low plasma AA concentration, there was increased urinary excretion of AA that was reversed by treatment with either tolrestat or myo-inositol. In contrast, AA supplementation normalized plasma AA concentrations while further increasing urinary AA excretion. The abnormality of AA metabolism was less severe in galactose-fed rats, which had normal plasma AA levels and only minor increases in urinary AA excretion. These studies demonstrated a disturbance in the regulation of plasma and urinary AA concentration in experimental diabetes and confirmed the relationship of AA with the polyol pathway. Because AA has many important biological functions, abnormalities of AA metabolism could be important in the pathogenesis of some diabetic complications. The interaction of the polyol and AA pathways suggests that this could be another site of action for aldose reductase inhibitors.

The effect of salicylates on nonenzymatic glycosylation and thermal stability of collagen in diabetic rats.

The effects of glucose on the nonenzymatic glycosylation and thermal rupture time of rat tail collagen were examined by (1) in vitro incubation of collagen fibers in glucose and (2) in vivo in diabetic rats. In vitro, glucose caused a dose-dependent rise in both nonenzymatic glycosylation and thermal rupture time and there was a good correlation between these two parameters (r = 0.68, P less than 0.0001). Aspirin, a known inhibitor of nonenzymatic glycosylation, was effective in preventing the glucose-induced rise in nonenzymatic glycosylation and thermal rupture time when present at concentrations of 0.78, 1.56, and 3.12 mM. Sodium salicylate at concentrations of 1.56 and 3.12 mM was also effective. In vivo, the nonenzymatic glycosylation and thermal rupture time of collagen fibers were both increased by 2-3-fold in rats with streptozotocin-induced diabetes of 4 wk duration. Aspirin or sodium salicylate treatment for 4 wk (240 mg/kg/day) from the onset of diabetes was able to prevent the rise in thermal rupture time without affecting nonenzymatic glycosylation of collagen or glycosylated hemoglobin levels. Aspirin or sodium salicylate treatment did not have detectable effect on properties of collagen in normal rats. The in vitro findings are consistent with the hypothesis that nonenzymatic glycosylation leads to the increased thermal stability of collagen fibers. The significance of nonenzymatic glycosylation in vivo is less certain, as thermal rupture time can be altered independently. The action of aspirin and sodium salicylate in vivo suggests new therapeutic options in the prevention and treatment of diabetic complications.

Deficiency of ascorbic acid in experimental diabetes. Relationship with collagen and polyol pathway abnormalities.

The plasma and tissue concentration of ascorbic acid (AA) is reduced in diabetes. This study was designed to investigate the mechanism and significance of this phenomenon. The low plasma AA concentration of diabetic rats can be normalized by dietary AA supplement (20-40 mg/day), a dosage approximately equal to the maximal synthetic rate of this substance in the rats. Treatment of diabetic rats with this regime prevented the decrease in activity of granulation tissue prolyl hydroxylase (PRLase), an AA-dependent enzyme required for maintaining the normal properties of collagen. The decreased plasma AA concentration and granulation tissue PRLase activity in diabetes can also be normalized by the aldose reductase inhibitor tolrestat. We conclude that in diabetic animals there is a true deficiency of AA that may be responsible for some of the changes of collagen observed in diabetes. Treatment with AA or an aldose reductase inhibitor may prevent some of the diabetic complications with underlying collagen abnormalities.

The effects of cyclooxygenase and lipoxygenase inhibitors on the collagen abnormalities of diabetic rats.

The importance of cyclooxygenase and lipoxygenase pathways in the determination of collagen abnormalities in diabetes was investigated. Pharmacological agents with antiprostaglandin activity, such as indomethacin, naproxen, and aspirin, were able to prevent the rise in thermal rupture time of tail collagen in diabetic rats. Paracetamol was without effect. The action of indomethacin on diabetic collagen was abolished by concurrent administration of sodium benoxaprofen, an inhibitor of lipoxygenase, to the diabetic rats. Collagen abnormalities in diabetes may be regulated by a balance of the cyclooxygenase and lipoxygenase pathways. Antiprostaglandin agents may have a role in the prevention of some diabetic complications.

Changes in hepatic glutathione metabolism in diabetes.

Glutathione is important in the regulation of the redox state, and a decline in its tissue level has often been considered to be indicative of increased oxidative stress in diabetes. In this study of diabetic rats, the level of hepatic glutathione was normal unless food intake was restricted. Thus, the previous report of a reduction in hepatic glutathione in diabetes is likely to be the result of food deprivation rather than diabetes alone. In contrast to changes characteristic of oxidative stress, the efflux of glutathione in bile from diabetic animals was significantly decreased, whereas hepatic mixed disulfides were unchanged, and the hepatic gamma-glutamyltransferase activity was considerably increased. These changes were not reproduced by food deprivation. The decrease in biliary excretion of glutathione in diabetes may reflect an attempt to conserve glutathione by activation of the hepatic gamma-glutamyl cycle. We conclude that the disturbances of glutathione metabolism in diabetes are not typical of those seen in oxidative stress or food restriction.

Interaction of ascorbic acid and glucose on production of collagen and proteoglycan by fibroblasts.

Collagen and proteoglycans are two major constituents of the extracellular matrix, and their abnormalities have been incriminated in the pathogenesis of diabetic complications. A decrease of plasma ascorbic acid has been reported in diabetes and thus may play a role in the collagen and proteoglycan abnormalities in diabetes. Ascorbic acid and glucose share structural similarity, and their metabolism may interact at the level of membrane transport and cellular action. In this study, we used a fibroblast culture system to explore this possibility. Ascorbic acid increased collagen and proteoglycan both in the culture medium and the cell layer. This stimulatory action of ascorbic acid was inhibited by the presence of glucose at a concentration of 25 mM. The effect of high glucose concentration was not mediated by inhibition of ascorbic acid uptake by fibroblasts. Insulin is able to abolish this inhibitory action of glucose on collagen production, but the precise mechanism is unclear. These results show that the high glucose concentration in diabetes can impair the action of ascorbic acid at the cellular level. This may further accentuate the problem of decreased availability of this vitamin as a result of its low plasma concentration.