Insulin proteinase liberates from glucagon a fragment known to have enhanced activity against Ca2+ + Mg2+-dependent ATPase.

We find, contrary to previous reports, that substantial cleavage of glucagon by insulin proteinase occurs at only one region, namely the double-basic sequence -Arg17-Arg18-. Cleavage takes place almost exclusively between these two residues, liberating fragments glucagon-(1-17) and glucagon-(18-29). Others have shown that the fragment glucagon-(19-29) is 1000-fold more efficient compared with intact glucagon, at inhibiting the Ca2+-activated and Mg2+-dependent ATPase activity and the Ca2+ pump of liver plasma membranes. We show that this fragment is not liberated in detectable quantities by our insulin proteinase preparation. On the other hand, others have shown that glucagon-(18-29), though less active than glucagon-(19-29), was still 100-fold more active than glucagon itself in the above-mentioned system. Our observations represent the first demonstration of the release by insulin proteinase of a hormone fragment having enhanced activity, although it has yet to be shown that the activity of this fragment is important in vivo. Since the formation of glucagon-(19-29) from glucagon-(18-29) would involve merely removal of Arg18, a second enzyme might exist to provide the more active fragment.     

Glucagon-(19-29) exerts a biphasic action on the liver plasma membrane Ca2+ pump which is mediated by G proteins

We have recently shown that nanomolar concentrations of glucagon-(19-29), which can derive from native glucagon by proteolytic cleavage of the dibasic doublet Arg17-Arg18, inhibit the Ca2+ pump in liver plasma membrane vesicles independently of adenylyl cyclase activation (Mallat, A., Pavoine, C., Dufour, M., Lotersztajn, S., Bataille, D., and Pecker, F. (1987) Nature 325, 620-622). We report here that the regulation of the Ca2+ pump by glucagon-(19-29) is dependent on guanine nucleotides. In the presence of 10 microM guanosine 5'-3-O-(thio) triphosphate (GTP gamma S) or 75 microM GTP, glucagon-(19-29) caused a biphasic regulation of the Ca2+ pump. ATP-dependent Ca2+ transport was inhibited in the presence of 10 pM to 1 nM glucagon-(19-29), while higher concentrations of the peptide (1-100 nM) reversed the inhibition caused by lower ones. GTP gamma S alone, at high concentrations (100 microM), reproduced the inhibitory effect of glucagon-(19-29) and induced a 40% inhibition of the basal activity of the Ca2+ pump which was reversed by low concentrations of glucagon-(19-29) (10 pM to 1 nM). Treatment of rats with cholera toxin resulted in a 70% increase in the basal activity of the Ca2+ pump, a loss of sensitivity to GTP gamma S and to the biphasic regulation by glucagon-(19-29). Treatment with pertussis toxin did not affect the response of the Ca2+ pump to GTP gamma S and glucagon-(19-29). We conclude that glucagon-(19-29) can exert a biphasic effect on the Ca2+ pump which is mediated by G protein(s) sensitive to cholera toxin.     

Glucagon is processed to the (19-29) fragment at the level of the hepatocyte membrane

Upon incubation with hepatic plasma membranes, glucagon is processed into its (19-29) C-terminal fragment. This suggests that, in physiological conditions, glucagon is processed in a target tissue at the level of its Arg17-Arg18 basic doublet, leading to the production of a fragment which is known to display an original biological specificity, namely the modulation of the calcium pump present in hepatocyte plasma membrane.     

Immunoreactive glucagons purified from dog pancreas, stomach and ileum

Previous studies have shown that pig intestine contains a 69 amino acid glucagon (glicentin) as well as a 37 amino acid glucagon (oxyntomodulin). In pig pancreas the 29 amino acid glucagon predominates. Since glucagon is thought to be expressed from a single gene in mammals, these differences in molecular forms indicate differential posttranslational processing of the glucagon precursor by different tissues. In the current study glucagon immunoreactivity (IR) was separately purified from dog pancreas, stomach mucosa and ileum mucosa. Purification and sequence analysis of the different tissue glucagons show that dog pancreas and stomach mucosa contain glucagon-29 while ileum mucosa contains glucagon-37 and glucagon-69. The latter is the major form present with glucagon-37 accounting for only 10-20% of the total ileum glucagon content. The N-terminal 32 amino acid portion of dog glucagon-69 differs at 6 sites from pig glucagon-69: RSLQDTEEKSRSFSAPQTEPLNDLDQMNEDKR... The C-terminal glucagon-37 is identical to pig oxyntomodulin.     

Structure-function relationships in glucagon. Re-evaluation of glucagon-(1-21)

Glucagon-(1-21) was prepared fully synthetically as well as by carboxypeptidase A digestion of natural porcine glucagon. Neither of the two preparations had glucagon agonistic effects with regard to receptor binding or adenylate cyclase activation in purified rat liver plasma membranes. Nor did these preparations contain lipolytic activity in isolated free fat cells. A preliminary batch of glucagon-(1-21) prepared by carboxypeptidase A digestion did, however, contain 1-2% glucagon bioactivity. This activity was separated from glucagon-(1-21) by high-performance liquid chromatography and quantitatively recovered in four minor hind peaks which eluted close to but not in a position identical to the elution position of native glucagon.     

Production of anti-glucagon sera with a C-terminal fragment of pancreatic glucagon.

The C-terminal region-sepcific anti-glucagon sera were raised in rabbits using as immunogen, and conjugate of BSA and a C-terminal fragment of pancreatic glucagon. The hapten was prepared by trypsin digestion of the glucagon, which was proved to be a 1:3 mixture of glucagon (18--29) and (19--29). Six rabbits were immunized by subcutaneous injection of an emulsion of the conjugate with complete Freund's adjuvant and five of the rabbits produced antibodies to the glucagon (GC-1, GC-2, GC-3, GC-5 and GC-6). For comparison, rabbit antisera were also produced against glucagon polymer (GA-10) and syrupy glucagon fibrils (PGA-2). All these antisera as well as the pancreatic glucagon-specific antiserum 30 K were characterized with dog gut-extract (gut-GLI) and glucagon-related peptide fragments in the radioimmunoassay systems. The assay systems utilized 125 I-monosubstituted pancreatic glucagon as tracer and human mono-component glucagon as standard. All sera of the GC-series crossreacted with the dog gut-extract very weakly and antisera GC-5 and GC-6 exhibited the lowest crossreactivities with the extract, which were shown to be as low as that of 30k. Characterization of the antiserum GC-5 with purified glucagon related fragments indicated that the major antigenic determinant located exactly in the C-terminal region of glucagon. The present results clearly showed high efficiency of the use of the glucagon C-terminal fragment as hepatenic immunogen in obtaining the C-terminal region-specific, i.e., pancreatic glucagon-specific antisera.     

Effect of glucagon-(1-21)-peptide on secretin-stimulated pancreatic exocrine secretion in anesthetized dogs

The effects of glucagon-(1-21)-peptide on pancreatic exocrine secretion and plasma glucose levels were studied and compared with those of native glucagon in anesthetized dogs. Intravenous bolus administration of 1 nmol or 10 nmol/kg of glucagon-(1-21)-peptide evoked a significant inhibition of secretin-stimulated pancreatic juice secretion and protein output in a dose-dependent manner, as equimolar doses of glucagon did. Native glucagon induced an immediate and transient increase in pancreatic juice volume, which was followed by a significant inhibition. However, glucagon-(1-21)-peptide showed only the inhibitory action. Glucagon-(1-21)-peptide had no effect on plasma glucose levels even when a dose of 10 nmol/kg was given. The results suggest that the N-terminal amino-acid residues of glucagon play an important role in the inhibition of pancreatic exocrine secretion.     

Evidence for the occurrence of two forms of β-lipotropin in rat pituitary

A peptide isolated from porcine gut according to its glucagon-like activity in liver (bioactive enteroglucagon) has been characterized immunologically, biologically and chemically: its potency relative to pancreatic glucagon in interacting with an antiglucagon antibody, hepatic glucagon-binding sites and hepatic adenylate cyclase was ～100%, 20% and 10%, respectively. In contrast, it is ～20-times more potent than glucagon in oxyntic glands, justifying the term ‘oxyntomodulin’. Chemically, it consists in the 29 amino acid-peptide glucagon elongated at its C-terminal end by the octapeptide Lys—Arg—Asn—Lys—Asn—Asn—Ile &;—Ala; accordingly, it is called ‘glucagon-37’     

Glucagon-(1-21) fails to inhibit meal size in rats

Rats were intraperitoneally injected with 115 or 230 nmol/kg pancreatic glucagon or equimolar doses of glucagon-(1-21) just before refeeding at the beginning of the dark phase after a 12 hr period of food deprivation. Glucagon-(1-21) and pancreatic glucagon have been reported to have similar visceral effects, but glucagon-(1-21) does not have pancreatic glucagon's metabolic effects. In this experiment, both doses of pancreatic glucagon, but neither dose of glucagon-(1-21), significantly decreased meal size. This indicates that the C-terminal octapeptide of pancreatic glucagon is necessary for its satiety effect, just as it is for its metabolic effects.     

Conformational and biological properties of glucagon fragments containing residues 1-17 and 19-29

The 29 amino acid polypeptide hormone glucagon was cleaved into two large fragments by the enzyme clostripain. The conformational properties of these two fragments were monitored by circular dichroism at pH 2 and 12 in both the presence and absence of sodium dodecyl sulfate. Both glucagon (1-17) and glucagon (19-29) have reduced abilities to fold in aqueous solution. However, both fragments can take on structure of higher apparent helical content in acidic solution in the presence of sodium dodecyl sulfate but only the glucagon (19-29) retains this conformation at high pH. Neither of the two fragments react with dimyristoylphosphatidylcholine as the intact peptide does. Only the carboxyl terminal fragment was capable of reacting with an antibody specific for glucagon. The glucagon (1-17) has markedly reduced affinity for binding to the glucagon receptor as well as markedly reduced ability to stimulate adenylate cyclase activity which is not affected by the presence of glucagon (19-29). It is proposed that the intact sequence provides specific groups required for activity as well as the potential for forming a stable amphipathic helix, both of which are necessary for full biological activity at low peptide concentrations.     

Cell-specific post-translational processing of preproglucagon expressed from a metallothionein-glucagon fusion gene

Glucagon is a peptide hormone of 29 amino acids encoded by a preprohormone which contains in tandem the sequences of glucagon and two additional glucagon-like peptides (GLPs) structurally related to glucagon and separated by intervening peptides. Glucagon arises by cleavage from the prohormone within the A cells of the pancreatic islets but in the intestine remains as part of a partially processed precursor (glicentin). To determine whether additional glucagon-like peptides are processed from preproglucagon and to analyze for potential cellular specificity in the processing of preproglucagon, we introduced and expressed a metallothionein-glucagon fusion gene in a fibroblast and two endocrine (pituitary and pancreatic islet) cell lines. Chromatographic analyses of cell extracts utilizing specific radioimmunoassays to chemically synthesized peptides demonstrate the liberation of intact glucagon, glicentin, GLP-I(1-37), GLP-I(7-37), GLP-II, and an intervening peptide amidated at its carboxyl terminus. The peptides were present in distinct yet different patterns in the two endocrine but not the fibroblast cell lines. The cell-specific liberation of the glucagon-like and intervening peptides suggests their potential as new bioactive peptides. The cellular specificity in the processing of preproglucagon indicates that the genetic determinants of the processing activity are complex and are expressed in a cell-specific manner.     

Production of a specific antiserum by synthetic C-terminal fragment of glucagon

Seven rabbits were immunized with a synthetic C-terminal glucagon fragment [15--29] conjugated with bovine serum albumin by means of glutaraldehyde. Antisera for glucagon were produced in all the animals after six injections of the conjugate. One of them revealed a higher titer antiserum (G42), which did not cross react with gut glucagon-like immunoreactive material, secretin, insulin, gastric inhibitory polypeptide or vasoactive intestinal peptide. From the results of inhibition of 125 I-glucagon in binding with the antiserum by various glucagon-related fragments the immunogenic determinant of the antiserum was proved to be in the C-terminal residue of the glucagon molecule, although peptide [17--29] or [21--29] reacted weakly with the antiserum. The plasma glucagon levels measured by antiserum G 42 during an arginine test in five normal subjects were superposed on those obtained by other antiserum (G21), specific for pancreatic glucagon. Furthermore, a comparable standard curve for glucagon was obtained using antiserum G42, when a labelled p-hydroxyphenylacetylated glucagon fragment [15--29] was employed as a tracer. The present study clearly demonstrated that the C-terminal glucagon fragment could yield a specific antiserum for pancreatic glucagon, supporting the proposal that the C-terminal fragment of glucagon is responsible for such specific antisera. Furthermore, it is concluded that immunoassay for glucagon could be performed using the labelled glucagon fragment as a tracer.     

Enzyme immunoassay of pancreatic glucagon at the picogram level using beta-D-galactosidase as a label.

An enzyme immunoassay of pancreatic glucagon was established by using E. coli beta-D-galactosidease [EC 3.2.1.23] as a marker. In order to increase the sensitivity of the immunoassay, different peptides obtained from glucagon fragments were used to produce the enzyme conjugate and the immunogen. Antiserum N6E raised against C-terminal fragment peptide (15-29) could be diluted to more than 1 : 100,000 in the assay and was highly specific for pancreatic glucagon. The antiserum reacted well with the C-terminal fragment peptide (21-29) as well as another fragment peptide (15-29) and pancreatic glucagon. The enzyme immunoassay using antiserum N6E and fragment peptide (21-29)-enzyme conjugate could detect as little as 1 to 2 pg of glucagon. The mean recovery of glucagon added to serum specimens was 104% and the coefficients of variation were 3.7-14.5% (within assay) and 9.0-18.5% (between assay).     

Glucagon1-6 binds to the glucagon receptor and activates hepatic adenylate cyclase.

A fragment of glucagon encompassing its first six NH2-terminal residues (His-Ser-Gln-Gly-Thr-Phe) binds to the glucagon receptor and stimulates adenylate cyclase activity in rat liver plasma membranes. Glucagon1-6 is a partial agonist since it stimulates, at saturating concentrations, to the extent of 75% of the maximal activity given by the native hormone. The binding affinity and potency of glucagon1-6 are 0.001% the native hormone. Discussed are the implications of these findings on the structure-function relationships required for the action of glucagon and for preparing clinically useful analogs of the hormone.     

Glucose metabolism in the newborn rat. Hormonal effects in vivo

1. The concentrations of liver glycogen and plasma d-glucose were measured in caesarian-delivered newborn rats at time-intervals up to 3h after delivery after treatment of the neonatal rats with glucagon, dibutyryl cyclic AMP, cortisol or cortisol+dibutyryl cyclic AMP. Glycogenolysis was promoted by glucagon or dibutyryl cyclic AMP in the third hour after birth but not at earlier times. Cortisol and dibutyryl cyclic AMP together (but neither agent alone) promoted glycogenolysis in the second hour after birth, but no hormone combination was effective in the first postnatal hour. 2. The specific radioactivity of plasma d-glucose was measured as a function of time for up to 75 min after the intraperitoneal injection of d-[6-(14)C]glucose and d-[6-(3)H]glucose into newborn rats at delivery and after treatment with glucagon or actinomycin D. Glucagon-mediated hyperglycaemia at this time was due to an increased rate of glucose formation and a decreased rate of glucose utilization. Actinomycin D prevented glucose formation and accelerated the rate of postnatal hypoglycaemia. 3. The specific radioactivity of plasma l-lactate and the incorporation of (14)C into plasma d-glucose was measured as a function of time after the intraperitoneal injection of l-[U-(14)C]lactate into glucagon- or actinomycin D-treated rats immediately after delivery. The calculated rates of lactate formation were unchanged by either treatment, but lactate utilization was stimulated by glucagon administration. Glucagon stimulated and actinomycin D diminished (14)C incorporation into plasma d-glucose. 4. The factors involved in the initiation of glycogenolysis and gluconeogenesis in the rat immediately after birth are discussed.     

Metabolism of glycine- and hydroxyproline-containing peptides by the isolated perfused rat kidney.

Thiol and glutathione (GSH) efflux across the sinusoidal plasma membrane in isolated perfused rat liver was stimulated by addition of hormones such as vasopressin, phenylephrine and adrenaline, whereas glucagon or dibutyryl cyclic AMP were without effect. Phenylephrine and adrenaline effects were sensitive to prazosin and phentolamine, respectively. The increase in thiol efflux was largely accounted for by an increase in GSH efflux. Thiol efflux and the hormone effects were abolished in GSH-depleted liver. Biliary GSH efflux was diminished upon hormone addition. The newly discovered hormone-dependence of GSH release across the sinusoidal plasma membrane may explain the known loss of GSH during conditions of experimental shock (traumatic or endotoxin) and stress and peripheral inflammation.     

Stabilization of soluble active rat liver glucagon receptor

Active glucagon receptor was solubilized with 3-(3-cholamidopropyl)dimethylammonio-1-propanesulfonate (Chaps) from rat liver plasma membranes but rapidly (less than 8 h) lost activity. Either inclusion of 1X Hanks' balanced salt solution in the 3 mM Chaps solubilization buffer or its addition after solubilization increased the percentage of total binding attributable to specific glucagon binding from approximately 10 to greater than 80%; of great importance, it increased the stability from near zero binding at 8 h to 50% binding at 48 h (4 degrees C). Of the Hanks' solution components, either NaCl (137 mM) or CaCl2 (1.26 mM) was effective in increasing specific binding to approximately 70 and 60% respectively: Mg salts were ineffective. Soluble receptor binding activity was assayed by dextran-coated charcoal adsorption of free hormone. The assay is rapid, simple, and reproducible. It is suitable for monitoring receptor activity during purification and molecular characterization. Competition binding studies gave an IC50 value of 10-20 nM (slope factor approximately 1), with or without GTP. Dissociation assays revealed GTP sensitivity when receptors were solubilized either as glucagon-receptor complexes or free receptor. Active glucagon-receptor complexes could be eluted from wheat germ lectin-agarose: neither concanavalin A-agarose nor soybean agglutinin-agarose bind receptor. A glucagon degrading activity which co-solubilized with the receptor but did not require detergent for extraction was distinguishable from the soluble receptor not only by solubility but also by its heat stability (30 degrees C), its inhibition by bacitracin, its affinity for glucagon, its retention of activity for at least 1 week at 4 degrees C, and its size.     

Glucagon: structure-function relationships investigated by sequence deletions

A series of glucagon analogues, des-(1-4)-glucagon, des-(5-9)-glucagon, des-(10-15)-glucagon, des-(16-21)-glucagon, des-(22-26)-glucagon and des-(27-29)-glucagon, were prepared by condensation of synthetic fragments and characterized biologically and immunologically. Fully synthetic glucagon was also characterized. The potencies with regard to glucagon receptor binding in purified rat liver plasma membranes were, in decreasing order: synthetic glucagon 108%, des-(1-4)-glucagon 5.7%, des-(27-29)-glucagon 0.92%, des-(5-9)-glucagon 0.47%, des-(10-15)-glucagon 0.0028%, des-(16-21)-glucagon 0.0017% and des-(22-26)-glucagon 0.00060% relative to that of natural porcine glucagon. Des-(27-29)-glucagon was the only analogue that activated the adenylate cyclase in rat liver plasma membranes or stimulated the lipolysis in isolated free fat cells from rat epididymal fat pad. The potencies were 0.16% and 0.20% of that of glucagon, respectively. Des-(1-4)-glucagon was a glucagon antagonist in the adenylate cyclase assay. The immunoreactivities of the glucagon analogues were determined with two commonly used anti-glucagon sera, K 5563 and K 4023, directed towards the C-terminus and some segment in the sequence 2-23, respectively. In the K 5563 assay, des-(27-29)-glucagon and des-(22-26)-glucagon had potencies of 0.0009% and less than 0.09% of that of glucagon, respectively. The remaining analogues had potencies varying from 45% to 141% of that of glucagon. In the K 4023 assay, the analogues showed a non-linear dilution effect. The combined results indicate a partition within the glucagon molecule with regard to receptor binding and adenylate cyclase activation. The region 10-26 appears to be the most important for receptor binding, whereas 1-4 is essential for adenylate cyclase activation. The C-terminal segment 27-29 is important for the maintenance of full receptor binding but non-essential for adenylate cyclase activation.     

N-ethylmaleimide uncouples the glucagon receptor from the regulatory component of adenylyl cyclase

125I-Glucagon binding to rat liver plasma membranes was composed of high- and low-affinity components. N-Ethylmaleimide (NEM) and several other alkylating agents induced a dose-dependent loss of high-affinity sites. This diminished the apparent affinity of glucagon receptors for hormone without decreasing the binding capacity of membranes. Solubilized hormone-receptor complexes were fractionated as high molecular weight (Kav = 0.16) and low molecular weight (Kav = 0.46) species by gel filtration chromatography; NEM or guanosine 5'-triphosphate (GTP) diminished the fraction of high molecular weight complexes, suggesting that NEM uncouples glucagon receptor-N-protein complexes. Exposure of intact hepatocytes to the impermeable alkylating reagent p-(chloromercuri)benzenesulfonic acid failed to diminish the affinity of glucagon receptors on subsequently isolated plasma membranes, indicating that the thiol that affects receptor affinity is on the cytoplasmic side of the membrane. Hormone binding to plasma membranes was altered by NEM even after receptors were uncoupled from N proteins by GTP. These data suggest that a sensitive thiol group that affects hormone binding resides in the glucagon receptor, which may be a transmembrane protein. Alkylated membranes were fused with wild-type or cyc- S49 lymphoma cells to determine how alkylation affects the various components of the glucagon-adenylyl cyclase system. Stimulation of adenylyl cyclase with fluoride, guanylyl 5'-imidodiphosphate, glucagon, or isoproterenol was observed after fusion of cyc- S49 cells [which lack the stimulatory, guanine nucleotide binding, regulatory protein of adenylyl cyclase (Ns)] with liver membranes alkylated with 1.5 mM NEM.(ABSTRACT TRUNCATED AT 250 WORDS)     

The effect of chronic ethanol ingestion on synthesis and degradation of soluble, contractile and stromal protein fractions of skeletal muscles from immature and mature rats.

Both insulin and glucagon from the pancreas of the holocephalan cartilaginous fish Callorhynchus milii (elephantfish) have been isolated and purified. Two reverse-phase h.p.l.c. steps enabled recovery of sufficient material for gas-phase sequencing of the intact chains as well as peptide digestion products. The elephantfish insulin sequence shows 14 differences from pig insulin, including two unusual substitutions, Val-A14 and Gln-B30, though none of these is thought likely to influence receptor binding significantly. The insulin B-chain contains 31 residues, one more than mammalian insulins, but markedly less than that of the closely related ratfish with which it otherwise exhibits high sequence similarity. Elephantfish and pig glucagons differ at only four positions, but there are six changes from the ratfish glucagon-36 (normal glucagon contains 29 residues) sequence. It is apparent that different prohormone proteolytic processing mechanisms operate in the two holocephalan species.