Virus-induced diabetes mellitus. VI. Genetically determined host differences in the replicating of encephalomyocarditis virus in pancreatic beta cells.

Beta cells were isolated from strains of mice that were susceptible and resistant to encephalomyocarditis (EMC) viral-induced diabetes mellitus. Beta cells from susceptible mice that were infected in vivo with EMC virus showed higher viral titers, more severe degranulation, and lower concentrations of immunoreactive insulin than beta cells from resistant mice. Immunofluorescence and infectious center assays revealed that pancreas from susceptible mice contained at least 10 times more infected cells than pancreas from resistant mice. Beta cell cultures prepared from susceptible mice and infected in vitro also showed higher viral titers and more severe cytopathologic changes than beta cell cultures from resistant mice. In contrast to beta cell cultures, virus replicated equally well in primary embryo and kidney cell cultures from susceptible and resistant strains of mice. It is concluded that the development of EMC virus-induced diabetes is related to genetically determined host differences in the capacity of the virus to infect beta cells.

Freeze-fracture cytochemistry: localization of wheat-germ agglutinin and concanavalin A binding sites on freeze-fractured pancreatic cells.

The combined application of thin-section and critical-point-drying "fracture-label" is used to determine the pattern of distribution and partition of wheat-germ agglutinin and concanavalin A binding sites on the membrane faces of freeze-fractured exocrine and endocrine rat pancreatic cells. Whereas the exoplasmic face of plasma membrane is preferentially labeled by both lectins, the endoplasmic reticulum and nuclear envelope are strongly and uniformly labeled by concanavalin A but not by wheat-germ agglutinin. The results support current views in the glycosylation of membrane proteins and do not support the backflow of sialidated glycoproteins to the endoplasmic reticulum.

Myosins of secretory tissues.

Myosin has been purified from the principal pancreatic islet of catfish, hog salivary gland, and hog pituitary. Use of the protease inhibitor Trasylol (FBA Pharmaceuticals, New York) was essential in the isolation of pituitary myosin. Secretory tissue myosins were very similar to smooth muscle myosin, having a heavy chain of 200,000 daltons and light chains of 14,000 and 19,000 daltons. Salivary gland myosin cross-reacted with antibodies directed toward both smooth muscle myosin and fibroblast myosin, but not with antiskeletal muscel myosin serum. The specific myosin ATPase activity measured in 0.6 M KCl was present. Tissues associated with secretion of hormone granules contained substantial amounts of this ATPase, rat pancreatic islets having 4.5 times that of rat liver. Activation of low ionic strength myosin ATPase by actin could not be demonstrated despite adequate binding of the myosin to muscle actin and elution by MgATP. The myosins were located primarily in the cytoplasm as determined by cell fractionation and were quite soluble in buffers of low ionic strength.

Studies on proinsulin and proglucagon biosynthesis and conversion at the subcellular level. I. Fractionation procedure and characterization of the subcellular fractions.

Anglerfish islets were homogenized in 0.25 M sucrose and separated into seven separate subcellular fractions by differential and discontinuous density gradient centrifugation. The objective was to isolate microsomes and secretory granules in a highly purified state. The fractions were characterized by electron microscopy and chemical analyses. Each fraction was assayed for its content of protein, RNA, DNA, immunoreactive insulin (IRI), and immunoreactive glucagon (IRG). Ultrastructural examination showed that two of the seven subcellular fractions contain primarily mitochondria, and that two others consist almost exclusively of secretory granules. A fifth fraction contains rough and smooth microsomal vesicles. The remaining two fractions are the cell supernate and the nuclei and cell debris. The content of DNA and RNA in all fractions is consistent with the observed ultrastructure. More than 82 percent of the total cellular IRI and 89(percent) of the total cellular IRG are found in the fractions of secretory granules. The combined fractions of secretory granules and microsomes consistently yield >93 percent of the total IRG. These results indicate that the fractionation procedure employed yields fractions of microsomes and secretory granules that contain nearly all the immunoassayable insulin and glucagons found in whole islet tissue. These fractions are thus considered suitable for study of proinsulin and proglucagon biosynthesis and their metabolic conversion at the subcellular level.

Localization of the pancreatic beta cell glucose transporter to specific plasma membrane domains.

Immunocytochemical techniques revealed that the "liver-type" glucose transporter is present in the insulin-producing beta cells of rat pancreatic islets but not in other islet endocrine cells. Ultrastructural analysis of the transporter by the protein A-gold technique showed that it is restricted to certain domains of the plasma membrane, its density being sixfold higher in microvilli facing adjacent endocrine cells than in the flat regions of the plasma membrane. These results support a possible role for this glucose transporter in glucose sensing by beta cells and provide evidence that these cells are polarized.

Rat insulin genes: construction of plasmids containing the coding sequences.

Recombinant bacterial plasmids have been constructed that contain complementary DNA prepared from rat islets of Langerhans messenger RNA. Three plasmids contain cloned sequences representing the complete coding region of rat proinsulin I, part of the preproinsulin I prepeptide, and the untranslated 3' terminal region of the mRNA. A fourth plasmid contains sequences derived from the A chain region of rat preproinsulin II.

Differences in adrenergic recognition by pancreatic A and B cells.

The adrenergic control of glucose homeostasis is mediated in part through variations in the release of pancreatic hormones. In this study, purified pancreatic A and B cells were used to identify the recognition and messenger units involved in the adrenergic regulation of glucagon and insulin release. Catecholamines induced beta-adrenergic receptor activity in A cells and alpha 2-adrenergic receptor activity in B cells. The two recognition units provoked opposite variations in the production of cellular cyclic adenosine monophosphate, the beta-adrenergic unit enhancing the nucleotide's permissive effect on amino acid-induced glucagon release and the alpha 2-adrenergic unit inhibiting that upon glucose-induced insulin release. In both cell types, catecholamines interact powerfully with the synergistic control of hormone release by nutrient- and (neuro)hormone-driven messenger systems.

Minimal chronic hyperglycemia is a critical determinant of impaired insulin secretion after an incomplete pancreatectomy.

We now describe experiments that allow one to determine the consequences of B cell reduction alone vs. those that result from superimposed mild hyperglycemia. Male CD rats underwent a 60% pancreatectomy (Px); controls were sham operated. 1 wk later, either 10% sucrose (SUC) was substituted as fluid supply or tap water was continued (WAT). Plasma glucose and insulin values in Px-WAT remained equal to the sham groups; in Px-SUC the values were euglycemic for 25 d, but then nonfasting plasma glucose rose 15 mg/dl. After 6 wk, B cell mass in Px-WAT was reduced by 45% and non-B cell mass by 57%. In contrast, in Px-SUC both masses were comparable to the sham groups. The insulin response to 27.7 mM glucose was measured using the in vitro perfused pancreas. The responses were reduced in Px-WAT but in proportion to their reduced B cell mass; in contrast, it was 75% less than expected in Px-SUC. Also, the response to arginine given at 16.7 mM glucose was reduced only in Px-SUC. These results show that a lowering of B cell mass that does not result in hyperglycemia has no adverse effect on the remaining B cells. On the other hand, if even mild hyperglycemia develops, B cell function becomes impaired and results in inappropriately reduced insulin stores and insulin responses to marked stimuli.

Functional differences between rat islets of ventral and dorsal pancreatic origin.

Do functional linkages between islet endocrine cells exist? The effect of differences in frequency and distribution of islet endocrine cells on B cell function was examined in islets from the ventral (ventral islets) and dorsal (dorsal islets) areas of the rat pancreas. Dorsal islets contained 10 times as much glucagon as ventral islets, whereas insulin and total protein contents were similar. Basal rates of insulin secretion and proinsulin biosynthesis were similar in the two types of islet, but, under conditions of glucose stimulation, both insulin secretion and proinsulin biosynthesis were significantly greater in the glucagon-rich dorsal islets. Similarly, glucose utilization rates an ATP levels were greater in dorsal islets. In contrast, the rates of processing of newly synthesized proinsulin were similar in ventral and dorsal islets. That the islet glucagon content may have affected B cell function is inferred from two independent findings. Firstly, basal and glucose-stimulated cyclic AMP contents of glucagon-rich dorsal islets were greater than those of ventral islets. Secondly, in the presence of excess exogenous glucagon (1 microgram/ml), the differences in glucose-induced insulin secretion and proinsulin biosynthesis rates between the two types of islets were eliminated. These results strongly suggest that changes in the relative proportions of the different islet endocrine cells exert marked effects on islet function. In particular, a greater A cell and glucagon content is associated with higher rates of glucose-induced insulin secretion and biosynthesis.

Isolation and characterization of immunoreactive somatostatin from fish pancreatic islets.

Using a radioimmunoassay with labeled synthetic tetradecapeptide somatostatin, a large amount of immunoreactive somatostatin was found in the principal pancreatic islet of the channel catfish (Ictalurus punctata). The purpose of these experiments was to isolate and characterize the somatostatin-like material. Extracts of islets were chromatographed on a Bio-Gel P-30 column, and over 90% of the immunoreactive somatostatin migrated with proteins at least twice the size of synthetic tetradecapeptide somatostatin. This fraction was further purified by ion-exchange chromatography on carboxymethyl-cellulose and DEAE-cellulose columns. Two peptides were obtained with identical immunoreactivity, which was approximately 25% that of the synthetic somatostatin. Each peptide was judged to be >95% pure by thin-layer electrophoresis, polyacrylamide gel electrophoresis at pH 8.9, and highpressure liquid chromatography. Further criteria of purity included amino-terminal analysis of fraction IV yielding only aspartic acid. A total of 1.3 mg of fraction II, and 3.8 mg of fraction IV somatostatin-like peptides were obtained from 10 g of fresh frozen islets. Characterization of the two peptides revealed both peptides slightly more acidic than synthetic tetradecapeptide somatostatin. Fraction II had an isoelectric point of 8.0-8.3, and fraction IV 8.3-9.0. Molecular weight estimation by sodium dodecyl sulfate-urea polyacrylamide gel electrophoresis revealed similar mobility of both peptides, between pancreatic polypeptide (mol wt 4,500) and glucagon (mol wt 3,500). The mobility was not altered by reduction, and was approximately twice the size of synthetic tetradecapeptide somatostatin (mol wt 1,800). This confirmed that the peptides were single polypeptide chains and not aggregates, or somatostatin bound to larger proteins. Molecular weight determination by gel filtration chromatography on Bio-Gel P-6 in 8 M urea gave an estimated mol wt of 3,700. Amino acid analysis of the two immunoreactive somatostatins indicated that they were very similar in composition. Both pancreatic somatostatins (1 muM) had full biological activity relative to synthetic somatostatin measured as inhibition of growth hormone release from rat anterior pituitary cells.

In situ analysis of pancreatic islets in rats developing diabetes. Appearance of nonendocrine cells with surface MHC class II antigens and cytoplasmic insulin immunoreactivity.

Aberrant expression of MHC class II molecules on endocrine cells has been proposed to induce autoimmune reactions in thyroid and endocrine pancreas. The present study examines whether MHC class II positive insulin-containing islet cells occur at the onset of diabetes in rats, in analogy to the findings in man. At the onset of diabetes, both streptozotocin-treated and diabetes-prone BB rats exhibited numerous class II positive islet cells that presented ultrastructural features of monocytes and were surrounded by class II negative islet B cells. These class II positive cells were characterized by vacuoles that contained insulin immunoreactive granules and disrupted membranes. Similar cells also appeared positive for the monocyte marker OX-42. The presence of class II positive monocytes with insulin-containing vacuoles may indicate a removal of damage B cells by infiltrating leukocytes. A similar electron microscopical study in man will be necessary to distinguish the putative endocrine pancreatic B cells with aberrant class II expression from infiltrating nonendocrine class II positive cells with insulin-containing phagosomes.

Insulin and glucose as modulators of the amino acid-induced glucagon release in the isolated pancreas of alloxan and streptozotocin diabetic rats.

The hyperglucagonemia that occurs in vivo in animals made diabetic with alloxan or streptozotocin is not suppressed by high glucose but is suppressed by exogenous insulin. These observations together with other studies suggested that insulin-dependent glucose transport and metabolism by the alpha-cells serves as the primary mechanism controlling glucagon secretion. This hypothesis was tested in the present investigation. The possible interactions between glucose, insulin, and a mixture of 20 amino acids at physiological proportions were examined in the isolated-perfusin diabetic rats. Release of insulin and glucagon were used as indicators of theta-cell and alpha-cell function. According to rigid criteria the diabetic animals entering the study were severely diabetic. It was found that in vitro: (a) basal glucagon release (measured in the absence of an alpha-cell stimulus or inhibitor) was extremely low, even lower (i.e. 10%) than the basal rates seen in controls; (b) the alpha-cells of alloxanized- and streptozotocin-treated rats responded with a biphasic glucagon release to stimulation by an amino acid mixture; (c) this alpha-cell response was reduced after both streptozotocin and alloxan; (d) glucose at 5 mM was a potent inhibitor of amino acid-induced glucagon secretion in both types of experimental diabetes; (e) in alloxan diabetes alpha-cell stimulation by amino acids can be curbed by exogenous insulin, whereas glucagon secretion by the perfused pancreas of streptoxotocin diabetic rats appeared to be resistant to insulin action. The data indicate that the modulation of glucagon secretion by glucose in vitro is indipendent of insulin and that other unknown factors extrinsic to the pancreatic islets are responsible for the hyperglucagonemia observed in vivo.

Diabetogenic action of streptozotocin: relationship of dose to metabolic response.

The relationship between the dose of intravenously administered streptozotocin (a N-nitroso derivative of glucosamine) and the diabetogenic response has been explored by use of the following indices of diabetogenic action: serum glucose, urine volume, and glycosuria, ketonuria, serum immunoreactive insulin (IRI), and pancreatic IRI content. Diabetogenic activity could be demonstrated between the doses of 25 and 100 mg/kg, all indices used showing some degree of correlation with the dose administered. Ketonuria was only seen with the largest dose, 100 mg/kg. The most striking and precise correlation was that between the dose and the pancreatic IRI content 24 hr after administration of the drug, and it is suggested that this represents a convenient test system either for both related and unrelated beta cytotoxic compounds or for screening for modifying agents or antidiabetic substances of a novel type. Ability to produce graded depletion of pancreatic IRI storage capacity led to an analysis of the relationship between pancreatic IRI content and deranged carbohydrate metabolism. Abnormal glucose tolerance and insulin response were seen when pancreatic IRI was depleted by about one-third, while fasting hyperglycemia and gross glycosuria occurred when the depletion had reached two-thirds and three-quarters, respectively. The mild yet persistent anomaly produced by the lowest effective streptozotocin dose, 25 mg/kg, exhibits characteristics resembling the state of chemical diabetes in humans and might thus warrant further study as a possible model. Finally, the loss of the diabetogenic action of streptozotocin by pretreatment with nicotinamide was confirmed and was shown to be a function of the relative doses of nicotinamide and streptozotocin and of the interval between injections.

A comparison of the structure of hamster pancreatic insulin and insulin extracted from a transplantable hamster islet-cell carcinoma.

Insulin has been isolated from pancreases of the Syrian hamster and from a transplantable islet-cell tumor of the hamster. Acid/ethanol extraction, ether precipitation, ion exchange and gel filtration chromatography gave preparations of suitable purity for structural studies. Using trypsin cleavage, automatic Edman degradation and manual Edman degradation, a complete sequence of the pancreatic insulin B chain was determined. By automatic Edman degradation, the amino-terminal 10 residues of the pancreatic A chain were assigned and the sequence of carboxy-terminal eleven residues could be deduced by homology to other mammalian and avian insulins. The sequence assigned to hamster insulin A chain is identical to that of the rat, mouse and spiny mouse. The sequence of hamster insulin B chain is identical to rabbit and spiny mouse B chain. In terms of protein evolution, hamster insulin thus appears to occupy an intermediate position between rabbit and rat insulins. Amino acid composition, tryptic peptide composition and partial sequence analysis of the hamster tumor insulin showed no differences from hamster pancreatic insulin.

Accumulation of cadmium in pancreatic beta cells is similar to that of calcium in being stimulated by both glucose and high potassium.

The transport of Cd2+ and the effects of this ion on secretory activity and metabolism were investigated in beta cell-rich pancreatic islets isolated from obese-hyperglycemic mice. The endogenous cadmium content was 2.5 mumol/kg dry wt. After 60 min of incubation in a Ca2+-deficient medium containing 2.5 microM Cd2+ the islet cadmium content increased to 0.18 mmol/kg dry wt. This uptake was reduced by approx. 50% in the presence of 1.28 mM Ca2+. The incorporation of Cd2+ was stimulated either by raising the concentration of glucose to 20 mM or K+ to 30.9 mM. Whereas D-600 suppressed the stimulatory effect of glucose by 75%, it completely abolished that obtained with high K+. Only about 40% of the incorporated cadmium was mobilized during 60 min of incubation in a Cd2+-free medium containing 0.5 mM EGTA. It was possible to demonstrate a glucose-induced suppression of Cd2+ efflux into a Ca2+-deficient medium. Concentrations of Cd2+ up to 2.5 microM did not affect glucose oxidation, whereas, there was a progressive inhibition when the Cd2+ concentration was above 10 microM. Basal insulin release was stimulated by 5 microM Cd2+. At a concentration of 160 microM, Cd2+ did not affect basal insulin release but significantly inhibited the secretory response to glucose. It is concluded that the beta cell uptake of Cd2+ is facilitated by the activation of voltage-dependent Ca2+ channels. Apparently, the accumulation of Cd2+ mimics that of Ca2+ also involving a component of intracellular sequestration promoted by glucose.

Effects of depolarizing agents on the sodium content of rat pancreatic islets.

Rat pancreatic islets were used for studying the effects of depolarization on their sodium content. The islet sodium was markedly affected by small variations of extracellular K+. As with increased K+, the presence of low concentrations of glucose (5 mM) and arginine (2 mM) decreased the sodium content. The latter substances did not lower the sodium concentration below the value obtained by depolarization with excessive K+, nor was it possible to obtain a further decrease when 10 mM arginine was combined with 5 mM glucose. The sodium content was also reduced in the presence of 10 mM L-leucine, 10 mM 2-ketoisocaproate and 0.1 mM Ba2+. Tolbutamide differed from the other depolarizing agents in that it increased the sodium concentration, an effect manifested also in the presence of excessive K+. The observation that depolarizing agents other than sulfonylureas do not increase but actually reduce sodium implies that islet cells are exceptional among electrically excitable cells. The observed reduction of sodium may reflect activation of a voltage-sensitive carrier mechanism for outward transport of Na+.

Glucose-induced changes in cytosolic ATP content in pancreatic islets.

The cytosolic and mitochondrial contents in ATP, ADP and AMP were measured in islets incubated for 45 min at increasing concentrations of D-glucose and then exposed for 20 s to digitonin. The latter treatment failed to affect the total islet ATP/ADP ratio and adenylate charge. D-Glucose caused a much greater increase in cytosolic than mitochondrial ATP/ADP ratio. In the cytosol, a sigmoidal pattern characterized the changes in ATP/ADP ratio at increasing concentrations of D-glucose. These findings are compatible with the view that cytosolic ATP participates in the coupling of metabolic to ionic events in the process of nutrient-induced insulin release.

Islet cell antigens. Extraction studies and ELISA analysis.

Whole human and bovine pancreases were extracted in 20 mM Tris-HCl buffer without detergents and fractionated by high-speed centrifugation. The 80,000 x g supernatant was used to coat microtiter plates at a concentration of 5 micrograms protein/ml in phosphate-buffered saline. This solid-phase ELISA system was used for the detection of islet cell antigens defined by a series of monoclonal islet cell antibodies (HISL-1, -4, -5, -8, -14, and -19 and 4F2, 3G5, and A2B5). Both glycoprotein and glycolipid islet cell antigens in the total pancreatic extracts were detected by the monoclonal islet cell antibody in the ELISA system, indicating that epitope preservation had occurred during the extraction procedure. There was a good correlation between islet cell antigen quantitated by the ELISA system and the corresponding islet immunohistochemical reaction. Studies along these lines have the potential to facilitate the design of large-scale protocols for the purification of diabetes-related islet cell antigens to homogeneity.

Automated method for isolation of human pancreatic islets.

We describe an automated method for the isolation of human pancreatic islets. The procedure meets the following requirements: 1) minimal traumatic action on the islets, 2) continuous digestion in which the islets that are progressively liberated can be saved from further enzymatic action, 3) minimal human intervention in the digestion process, and 4) high yield and purity of the isolated islets. After purification on Ficoll gradients, an average of 164,600 islets/pancreas was obtained (2279 islets/g), with an average purity of 78.5% islets. The average volume and average insulin content of the final islet preparation were 348 mm3 and 93.4 U, respectively. The islets were morphologically intact with a normal degree of beta-granulation and responded to glucose stimulation with a fivefold increase of insulin secretion over basal levels. The procedure is now being used for the initiation of the second phase of clinical trials on human islet transplants.

Ganglioside expression in human pancreatic islets.

Recent biochemical studies have shown that the cytoplasmic islet cell-antibody autoantigen has properties of a monosialoganglioside (GM). To characterize islet glycolipids and ascertain whether islets express unique gangliosides, we determined the pattern of ganglioside expression in whole human pancreas and isolated human islets using high-performance liquid chromatography (HPLC) and high-performance thin-layer chromatography (HPTLC). The major gangliosides detected in glycolipid extracts of whole human pancreas were GM3, GD3 (disialoganglioside), and in a lesser amount, a GD1a-comigrating ganglioside. In contrast to whole human pancreas, isolated human islets were found to predominantly express GM3, an acidic glycolipid comigrating with GM2, and a ganglioside with mobility between GM2 and GM1 by both HPLC and HPTLC. Quantitation of the major ganglioside UV peaks seen on HPLC gave the following results. In whole pancreas, GM3 represented 66.7% of total gangliosides detected; an asialoglycolipid comigrating with GM2, 2.0%; a ganglioside migrating between GM2 and GM1, 2.6%; GD3, 22.6%; and a GD1a-comigrating ganglioside, 6.1%. In isolated islets, these components were found at the following levels: GM3, 14.9%; GM2-comigrating glycolipid, 74.2%; a ganglioside migrating between GM2 and GM1, 9.8%; GD3, 1.1%; and the GD1a-comigrating ganglioside, not detectable.