Glucose stimulation of the proinsulin synthesis in isolated pancreatic islets without increasing amount of proinsulin mRNA.
Explore the source record for details and available documents.
SEARCH · Search PubMed
Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Explore the source record for details and available documents.
Altered proinsulin levels in β-cells and bloodstream are hallmarks of diabetes and other diseases, but our knowledge about the proinsulin regulators remains limited. Here we perform a genome-wide CRISPR screen to identify 84 proinsulin regulators that alter intracellular proinsulin/insulin ratio in a mouse β-cell line. The proinsulin regulators are distinct from the insulin regulators from a previous orthogonal CRISPR screen. Functional annotation of the proinsulin regulators highlights Golgi as the primary organelle for proinsulin storage and regulation. Trafficking towards the Golgi increases the intra-cellular proinsulin/insulin ratio, while trafficking away from the Golgi, including exocytosis and Golgi-to-ER retrograde transport, decreases the intracellular proinsulin levels. We also map mouse quantitative trait loci (QTLs) associated with plasma proinsulin levels and use the CRISPR screen results to pinpoint the causal genes within the QTL loci. Interestingly, protein disulfide isomerase Pdia6 is the strongest hit from both CRISPR screen and the in vivo QTL mapping. Knocking down Pdia6 significantly reduce proinsulin accumulation in Golgi and secretory granules. Intriguingly, Pdia6-depletion in both human and mouse β-cells does not affect the folding status of proinsulin but causes significantly impaired proinsulin production through a UPR-independent mechanism. Taken together, our genetic profiles provide mechanistic insights into the regulation of proinsulin/insulin homeostasis.
Familial hyperproinsulinemia is an autosomal dominant defect that is associated with strikingly elevated levels of serum proinsulin-like material. Our studies show that trypsin converts familial hyperproinsulinemia proinsulin to insulin more slowly than it converts a 131I-labeled porcine proinsulin marker. Molar yields of insulin indicated that the material may be an intermediate proinsulin. Studies with two human C-peptide antisera that differ in their relative immunoreactivity with human C-peptide and proinsulin showed that the two antisera reacted equally with familial hyperproinsulinemia proinsulin, suggesting that it is a partially cleaved proinsulin intermediate. Sulfitolysis of highly purified material to break the inter- and intra-chain disulfide bridges and subsequent adsorption on a specific B-chain antibody covalently bound to Sepharose beads showed that the C-peptide was still connected to the B-chain. These data indicate that familial hyperproinsulinemia proinsulin is normally cleaved at the C-peptide-A-chain linkage site. A structural abnormality appears to underlie familial hyperproinsulinemia proinsulin, which impairs its cleavage at the B-chain-C-peptide linkage site.
This work addressed the problem of heterogeneity of immunoreactive insulin (IRI) in human plasma. Subjects with normal glucose tolerance were given 75g of an oral glucose solution, followed in 30 min by an intravenous infusion of 30g of arginine over 30 min. At the end of the infusion blood was withdrawn for analysis. IRI was extracted from plasma of individual subject by immunosorbent columns and was fractionated by gel filtration, disc gel electrophoresis and isoelectric focusing. Human IRI components were identified by molecular size, immunoreactivity with a human proinsulin antibody, sensitivity to trypsin, and by comparison of electrophoretic mobility and isoelectric point with porcine pancreatic products, after suitable correction for electric charge and molecular weight differences. The pattern of IRI heterogeneity was the same among six healthy subjects. Heterogeneity of proinsulin-size IRI in circulation was more marked than that of insulin-size material. Proinsulin and desdipeptide proinsulin were present in approximately equal amounts accompanied by minor amounts of split proinsulin and monodesamido-desdipeptide proinsulin. Insulin-size IRI contained over 80% insulin. Minor amounts of monodesamidoinsulin and diarginylinsulin were observed in some cases. The types of IRI components observed in plasma are evidence in support of a physiologic role of trypsin-and carboxypeptidase B-like enzymes in the conversion of proinsulin to insulin. Moreover, this study provides a base line for investigation of abnormalities in proinsulin-to-insulin conversion that may be associated with certain pathologic states.
An indirect two-site immunoradiometric assay is described for the measurement of human proinsulin in plasma. Polyethylene tubes coated with purified guinea-pig antibodies to insulin were used to extract proinsulin and insulin from plasma. Rabbit antibody to human C peptide was then added to react with the C-peptide moiety of the bound proinsulin. The uptake of this antibody was measured by the subsequent binding of 125I-sheep antibody to rabbit IgG. The binding of radioactivity to the tubes was a function of the proinsulin concentration in the sample. The sensitivity of the assay was 0.006 pmol/ml. Only 200 microliters of plasma was required in the assay and the 125I-labelled antibody was produced from readily available reagents. The polyethylene tubes remained stable for at least 5 months after coating. The mean fasting proinsulin level was 0.009 pmol/ml in sixteen normal subjects and 0.025 pmol/ml in twelve maturity onset diabetics. Oral glucose produced an 8 fold increase in proinsulin concentration but a decline in the plasma proinsulin/insulin molar ratio. Four patients with insulinoma had extremely elevated proinsulin levels and proinsulin/insulin ratios.
Porcine proinsulin, related intermediates and plasma immunoreactive insulin components have been studied by radioreceptor assay. Using the purified rat liver membrane or cultured human lymphocyte radioreceptor assay, porcine proinsulin is %5, split proinsulin 6% (54-55 split in connecting peptide) desdipeptide proinsulin 20% (deletion of amino acids 62 and 63 of connecting peptide) and desnonapeptide proinsulin 27% (deletion of amino acids 55-63 of connecting peptide) as active as porcine insulin in both assay systems; these values closely parallel the in vitro bioactivity of these preparations. In the lymphocyte radioreceptor assay the human plasma immunoreactive insulin-like component has the same potency as porcine insulin per immunoreactive unit, whereas the plasma immunoreactive proinsulin-like component is only 15% as active. Since both plasma immunoreactive components are somewhat less reactive than would be expected from puriified human insulin and proinsulin, the data suggest that both plasma components contain immunoreactive molecules that do not react in the radioreceptor assay.
Proinsulin synthesis, insulin release and intracellular ATP concentrations were measured in isolated rat islets of Langerhans under control conditions in vitro incubation and after treatment with several concentrations of streptozotocin for different periods of time. It was found that streptozotocin inhibited proinsulin synthesis, as well as insulin release, in a time and concentration dependent manner. The characteristics of the inhibition of these two processes were similar in general terms, but one dissimilarity was noted, i.e. after 60 min exposure to a high concentration of streptozotocin, proinsulin synthesis was inhibited more than insulin release. ATP content was reduced by high concentrations of streptozotocin, but it was found that proinsulin synthesis and insulin release could be inhibited without any effect on ATP content by a low (6.22 mM) concentration of streptozotocin. The effect of streptozotocin on proinsulin synthesis was judged to be the result of a target specificity for the B-cell rather than a specific effect on proinsulin relative to total protein synthesis.
Insulin, proinsulin, glucagon and gastrin were determined in extracts of tumors of 27 patients with pancreatic islet cell neoplasia of pancreas, in one patient with nesidioblastosis, in extracts of uninvolved portions of the pancreas in 11 of the tumor patients and of 15 control pancreases. Mean insulin concentration in solitary adenomas and in adenomas of patients with adenomatosis was higher than in control pancreases; however, in all but 1 patient the insulin concentration in neoplastic islet tissue was lower than in islet tissue of control pancreas, assuming islet volume is 1% of pancreas. The percentage of proinsulin was elevated in 52% of tumors. Adenoma insulin content correlated with increments of plasma insulin after tolbutamide administration. Insulin and proinsulin concentrations in pancreas uninvolved by tumor were not suppressed. Fasting plasma glucagon was elevated in patients with islet cell adenomatosis and in patients with islet cell carcinoma some of whom had multiple endocrine adenomatosis. The mean concentration of glucagon in tumors was lower than in control pancreases. Elevated concentration of gastrin was found in some adenomas. The data indicate: 1) insulin-secreting islet cell tumors have decreased storage capacity for insulin, 2) elevated concentration of proinsulin in tumors may be due to decreased capacity to store insulin and in some to decreased conversion of proinsulin to insulin as well, 3) tolbutamide stimulates the exaggerated release of a relatively constant fraction of insulin stored in adenomas. 4) solitary adenomas may contain excess amounts of pancreatic hormones in addition to insulin, 5) elevated plasma glucagon in patients with organic hyperinsulinism may indicate malignancy, microadenomatosis or multiple endocrine adenoma syndrome, and 6) chronic hyperinsulinism and hypoglycemia due to adenoma do not suppress insulin and proinsulin content of uninvolved pancreas.
Proinsulin biosynthesis was stimulated progressively to a plateau level at 45 min by 5.5 mmol/1 glucose. The rate of biosynthesis promptly decreased following cessation of either 15 or 60 min exposure to this concentration of glucose. Actinomycin D had no effect on the rate of decline. In contrast, exposure to 20 mmol/1 glucose caused a prolonged increase in proinsulin biosynthesis which was still apparent at 75 min. After removal of the stimulus there was an initial decrease in the rate of proinsulin biosynthesis followed by persistently elevated rates. When proinsulin biosynthesis was stimulated by 20 mmol/1 glucose in the presence of actinomycin D, there was a rapid switch off of biosynthesis and no persistent effect. Thus, two controls for regulation of proinsulin biosynthesis can be characterised: a prompt, rapidly reversible stimulation in response to low or high glucose concentrations and a persistent stimulation in response to high glucose concentrations. These two effects may represent the translational and transcriptional effects of glucose respectively.
Inosine, guanosine and adenosine strongly stimulated proinsulin biosynthesis and insulin secretion in isolated mouse pancreatic islets. None of the purine ribonucleosides stimulated insulin secretion in rat islets, although as reported [jain & Logothetopoulos (1977) Endocrinilogy 100, 923-927] inosine and guanosine, but no adenosine, were potent stimulants of proinsulin biosynthesis in this species. The purine bases had no effect in either species. D-Ribose, which enhanced proinsulin biosynthesis at 0.3 and 0.6 mM but not at 5mM in rat pancreatic islets [jain & Logothetopoulos (1977) Endocrinology 100, 923-927], produced no secretory signals in rat islets and was without any effect on proinsulin biosynthesis and insulin secretion in mouse islets. The rates of oxidation of 14C-labelled purine ribonucleosides and D-ribose in islets of the two species correlated well with their effectiveness as inducers of insulin secretion and proinsulin biosynthesis. Specific inhibitors of purine ribonucleoside phosphorylase, adenosine deaminiase and of purine ribonucleoside transport suppressed the stimulatory effects of nucleosides in pancreatic islets without altering the effect of D-glucose. The same inhibitors also markedly diminished the oxidation rats of the labelled purine ribonucleosides. The experiments clearly indicate that porinsulin biosynthesis and insulin secretion are modulated through metabolic signals and not through interactions of intact substrate molecules with cell receptors.
Inosine and guanosine were potent stimuli of proinsulin biosynthesis ([3H]leucine incorporation) in isolated pancreatic islets of the rat. The effect was nearly abolished by formycin B, an inhibitor of purine nucleoside phosphorylase, but not by D-mannoheptulose. The corresponding bases had no effect on the rate of proinsulin biosynthesis. D-ribose enhance proinsulin biosynthesis at low concentrations )0.3-0.6mM) but concentrations above 5 mM were ineffective. The effect of all three compounds was highly specific for proinsulin biosynthesis, since incorporation of [3H]leucine into other islet proteins was not significantly stimulated. The data strongly indicate that metabolic signals regulate modulation of proinsulin biosynthesis in the beta cells.
(1) A system is described for studying the short-term effects of agents on proinsulin synthesis in vitro, as measured by the incorporation of [3H]leucine into isolated proinsulin. (2) Of the agents tested, glucose has the most marked, and apparently earliest, effect on proinsulin synthesis. (3) The adenyl cyclase system participates in the regulation of proinsulin synthesis since exogenous cyclic AMP, glucagon, and caffeine are stimulatory. When cyclic AMP is added to the medium in the presence of glucose, it is the most potent agent acting on the adenyl cyclase-phosphodiesterase system. (4) The addition of NADPH to isolated rat islets inhibits proinsulin and Bulk Protein synthesis in vitro.
The secretory pattern of insulin and the rate of conversion of proinsulin to insulin were studied in isolated pancreatic islets from normoglycemic (buffer-infused for 24 hours) and hyperglycemic (glucose-infused for 24 hours) rats. The profiles of insulin secretion obtained during one hour of perifusion were markedly different in the two groups. The rate of insulin secretion by islets from the hyperglycemic rats was initially very high but progressively declined during the late period of the perifusion. The reverse pattern was found with the islets from buffer-infused rats. For the estimation of the rate of proinsulin conversion, islets were pulse-labeled with L-[4,5-3H]-leucine for 15 minutes and "chase"-incubated for 30 and 60 minutes. Labeled rat proinsulin and rat insulins in the medium and in the islet extracts were separated by a validated SDS-urea electrophoretic acrylamide procedure following immunoprecipitation. The conversion rate was estimated from the radioactivity in the insulin band, expressed as a per cent of the radioactivity in the proinsulin + insulin bands. Islets from hyperglycemic rats converted newly synthesized proinsulin to insulin at significantly higher rates than did control islets.
The antisera using at final dilution of 1 : 10,000 have been prepared by immunizing synthetic human proinsulin connecting peptide to rabbits for human proinsulin C-peptide radioimmunoassay. The cross reactivities of human proinsulin C-peptide derivatives with the prepared antisera were reduced by leaving amino acid residues from N terminal, although this phenomenon was a little different among antisera. Those results suggested that main antigen determinant in N terminal 31-38 of human proinsulin connecting peptide. The cross reactivities of other animal proinsulin C-peptide and other peptide hormones with the prepared antissera were not recognized at 10(3) p mole/ml.
The contribution of proinsulin to the total serum immunoreactive insulin (IRI) was measured in 59 patients with maturity onset diabetes (23 being treated with diet alone and 36 with oral sulfonylurea agents) and compared to that in 44 control subjects. The percentage of proinsulin was increased in 11 patients and correlated with plasma glucose, but not with IRI. There was no difference between the drug-treated group and diet-treated group, or between patients taking different sulfonylurea agents. Sequential studies in one patient showed normalization of the proportion of proinsulin following lowering of the plasma glucose level. It is probably that the increased circulating proportion of proinsulin in hyperglycemic diabetic patients is secondary to beta cell exhaustion with release of less mature granules.
1. Rabbit islets of Langerhans were disrupted by ultrasonic methods and the sonicated preparations were used to study proinsulin biosynthesis. 2. When [3h]leucine is incubated in such preparations, incorporation takes place into proinsulin, as evidenced by characterization on polyacrylamide gels, and by the conversion of this labelled material into insulin, by using trypsin. 3. The labelled proinsulin may also be purified by antiinsulin antibody bound to Sepharose. 4. With the broken-cell preparation it was shown that incorporation of leucine is accelerated by increasing the glucose content of the medium from 2mM to 16mM. However, 16mM-galactose or -sucrose did not stimulate incorporation significantly from basal values. This effect of glucose was abolished by cycloheximide. 5. The significance of these findings in relation to the mechanism of glucose stimulation of proinsulin biosynthesis is discussed.
Increasing concentrations of pyruvate failed to stimulate proinsulin biosynthesis and insulin release in freshly isolated islets. Glycolytic flux (3H2O from [5-3H]glucose) decreased by 80-85%, but decarboxylation of [1(-14)C]pyruvate was unaffected in islets tested immediately after alloxan exposure. This strongly suggested that in freshly isolated islets, beta-cells, in relation to other islet cells, hardly contribute to the decarboxylation of pyruvate. Non-alloxan-treated cultured islets decarboxylated 2-2.5 times as much pyruvate as did alloxan-treated islets cultured for 15-18h. Thus the contribution of beta-cells to the metabolism of pyruvate after culturing markedly increased. Concomitantly beta-cells became responsive to pyruvate. At 20mM-pyruvate, release of prelabelled proinsulin and insulin and incorporation of [3H]leucine into proinsulin reached values approximately half of those obtained with 20mM-glucose. Lactate was as effective as pyruvate in inducing responses in cultured islets. The experiments indicate that a critical degree of substrate utilization is necessary for the generation of signals for insulin release and proinsulin biosynthesis.
The ability of insulin and proinsulin to stimulate tyrosine transaminase in dexamethasone-treated cultured rat liver cells was compared. Insulin increased this enzyme whereas proinsulin was seemingly without effect. Since proinsulin was not degraded by these cells, failure of stimulation of hepatic tyrosine transaminase could not be due to rapid destruction of the prohormone. This is the first demonstration of an action of insulin that cannot be duplicated by proinsulin.