Hepatic glucagon metabolism. Correlation of hormone processing by isolated canine hepatocytes with glucagon metabolism in man and in the dog.

We have found that canine and rat hepatocytes convert (125I)iodoTyr10-glucagon to a peptide metabolite lacking the NH2-terminal three residues of the hormone. The peptide is released into the cell incubation medium and its formation is unaffected by a variety of lysosomotropic or other agents. Use of specific radioimmunoassays and gel filtration demonstrated in both normal subjects and in chronic renal failure patients a plasma peptide having the properties of the hormone fragment identified by cell studies. Studies of the dog revealed a positive gradient of the fragment across the liver and no differential gradient of the fragment and glucagon across the kidney. We conclude that the glucagon fragment arises from the cell-mediated processing of the hormone on a superficial aspect of the hepatocyte, the glucagon fragment identified during experiments in vitro represents the cognate of a peptide formed during the hepatic metabolism of glucagon in vivo, and measurement of the fragment by COOH-terminal radioimmunoassays could lead to an understimulation of hepatic glucagon extraction.

Familial hyperproinsulinemia. Two cohorts secreting indistinguishable type II intermediates of proinsulin conversion.

Familial hyperproinsulinemia, a hereditary syndrome in which individuals secrete high amounts of 9,000-mol wt proinsulin-like material, has been identified in two unrelated cohorts. Separate analysis of the material from each of the two cohorts had suggested that the proinsulin-like peptide was a conversion intermediate in which the C-peptide remained attached to the insulin B-chain in one case, whereas it was a conversion intermediate in which the C-peptide remained attached to the insulin A-chain in the other. To reinvestigate this apparent discrepancy, we have now used chemical, biochemical, immunochemical, and physical techniques to compare in parallel the structures of the immunoaffinity chromatography-purified, proinsulin-like peptides isolated from the serum of members of both families. Our results show that affected individuals in both cohorts secrete two-chained intermediates of proinsulin conversion in which the COOH-terminus of the C-peptide is extended by the insulin A-chain and from which the insulin B-chain is released by oxidative sulfitolysis. Analysis of the conversion intermediates by reverse-phase high-performance liquid chromatography using two different buffer systems showed that the proinsulin-related peptides from both families elute at a single position very near that of the normal intermediate des-Arg31, Arg32-proinsulin. Further, treatment of these peptides with acetic anhydride prevented trypsin-catalyzed cleavage of the C-peptide from the insulin A-chain, a result demonstrating the presence of Lys64 and the absence of Arg65 in both abnormal forms. We conclude that individuals from both cohorts with familial hyperproinsulinemia secret very similar or identical intermediates of proinsulin conversion in which the C-peptide remains attached to the insulin A chain and in which Arg65 has been replaced by another amino acid residue.

Human insulin B24 (Phe----Ser). Secretion and metabolic clearance of the abnormal insulin in man and in a dog model.

We have already demonstrated that a hyperinsulinemic, diabetic subject secreted an abnormal insulin in which serine replaced phenylalanine B24 (Shoelson S., M. Fickova, M. Haneda, A. Nahum, G. Musso, E. T. Kaiser, A. H. Rubenstein, and H. Tager. 1983. Proc. Natl. Acad. Sci. USA. 80:7390-7394). High performance liquid chromatography analysis now shows that the circulating insulin in several other family members also consists of a mixture of the abnormal human insulin B24 (Phe----Ser) and normal human insulin in a ratio of approximately 9.5:1 during fasting. Although all affected subjects show fasting hyperinsulinemia, only the propositus and her father are overtly diabetic. Analysis of the serum insulin from two nondiabetic siblings revealed that normal insulin increased from approximately 2 to 15% of total serum insulin after the ingestion of glucose and that the proportion of the normal hormone plateaued or fell while the level of total insulin continued to rise. Animal studies involving the graded intraportal infusion of equimolar amounts of semisynthetic human [SerB24]-insulin and normal human insulin in pancreatectomized dogs (to simulate the secretion of insulin due to oral glucose in man) also showed both a rise in the fraction of normal insulin that reached the periphery and the attainment of a brief steady state in this fraction while total insulin levels continued to rise. Separate experiments documented a decreased hepatic extraction, a decreased metabolic clearance rate, and an increased plasma half-life of human [SerB24]-insulin within the same parameters as those determined for normal human insulin. These results form a basis for considering (a) the differential clearance of low activity abnormal insulins and normal insulin from the circulation in vivo, and (b) the causes of hyperinsulinemia in both diabetic and nondiabetic individuals who secrete abnormal human insulins.

C-peptide and insulin secretion. Relationship between peripheral concentrations of C-peptide and insulin and their secretion rates in the dog.

Estimation of the insulin secretory rate from peripheral C-peptide concentrations depends upon the following characteristics of C-peptide kinetics: (a) equimolar secretion of insulin and C-peptide by pancreatic beta cells; (b) negligible hepatic extraction of C-peptide; (c) constant metabolic clearance rate (MCR) of C-peptide over a physiological and pathophysiological range of plasma levels; and (d) proportional changes in the secretion rate of C-peptide and its peripheral concentrations under varying physiological conditions. In the present experiments, the relationship between a variable intraportal infusion of C-peptide and its concentration in the femoral artery was explored in 12 pancreatectomized dogs. As the infusion of C-peptide was rapidly increased, the magnitude of its peripheral concentration initially increased less than the infusion rate by 20-30%. After an equilibration period of approximately 30 min, however, further increases and decreases in the intraportal infusion were accompanied by nearly proportional changes in its peripheral concentration. Estimates of the amount of C-peptide infused during the experiment based on the steady state C-peptide MCR and its peripheral concentration were within 20% of the amount of C-peptide actually infused. These experiments demonstrate that the portal delivery rate of C-peptide can be calculated from its MCR and peripheral concentration in the dog. They also provide a basis for testing the validity of more complicated models of insulin secretion based on peripheral C-peptide concentrations in the dog as well as other species, including man. Finally, we have shown that the hepatic extraction of endogenously secreted C-peptide is negligible in the basal state (3.1 +/- 6.1%), and does not change after oral glucose ingestion. The MCR of exogenous dog C-peptide was similar whether measured by constant peripheral intravenous infusion (12.3 +/- 0.7 ml/kg per min), constant intraportal infusion (13.4 +/- 0.6 ml/kg per min), or analysis of the decay curve after a bolus injection (13.5 +/- 0.7 ml/kg per min).

Tumor hypoglycemia: relationship to high molecular weight insulin-like growth factor-II.

The mechanism of tumor-associated hypoglycemia was examined in 11 patients with hepatocellular carcinoma, 6 of whom presented with severe hypoglycemia and 5 in whom plasma glucose was persistently normal. Serum insulin levels in the hypoglycemic patients were low. Although total serum insulin-like growth factor II (IGF-II) levels in both groups of tumor patients were lower than normal, tumor tissue from hypoglycemic patients contained levels of IGF-II mRNA that were 10-20-fold higher than those present in normal liver. IGF-II immunoreactivity consisted in all cases of a mixture of both higher molecular weight forms and material having the character of IGF-II itself. The former comprised a greater proportion of total IGF-II, in patients with hypoglycemia. Studies to characterize the interactions of IGF-II with serum proteins showed that (a) the radiolabeled peptide bound to an approximately 40,000-D protein in sera from both hypoglycemic patients and normal subjects, (b) sera from hypoglycemic patients and normal subjects had similar capacity to bind the radiolabeled peptide, and (c) the apparent affinities of serum binding proteins for IGF-II were the same for both hypoglycemic patients and normal subjects. Whereas, acid extracted, tumor-derived IGF-II immunoreactive peptides with low or intermediate molecular weights bound to serum proteins in a manner indistinguishable from that of IGF-II itself, the highest molecular weight IGF-II immunoreactive peptide exhibited negligible ability to compete for radiolabeled ligand binding to serum proteins. The low affinity of serum binding proteins for this component suggests that high molecular weight IGF-II immunoreactivity might circulate free and be available for interaction with cell-surface receptors.

Biochemical and clinical implications of proinsulin conversion intermediates.

Since a complete map of insulin-related peptides in humans requires consideration of proinsulin, Arg32/Glu33-split proinsulin, Arg65/Gly66-split proinsulin, des-Arg31,Arg32-proinsulin, des-Lys64, Arg65-proinsulin, and insulin, we applied high performance liquid chromatography coupled with radioimmunoassay to investigate the formation of proinsulin conversion intermediates in vitro and in vivo. Kinetic analysis of proinsulin processing by a mixture of trypsin and carboxypeptidase B (to stimulate in vivo processes) revealed (a) a rapid decline in proinsulin concommitant with formation of conversion intermediates, (b) formation of des-Arg31, Arg32-proinsulin and des-Lys64,Arg65-proinsulin in the ratio 3.3:1 at steady state, and (c) complete conversion of the precursor to insulin during extended incubation. Studies on normal human pancreas identified a similar ratio of des-Arg31,Arg32-proinsulin to des-Lys64,Arg65-proinsulin (approximately 3:1), whereas two insulinomas contained sizable amounts of des-Arg31,Arg32-proinsulin, but barely detectable amounts of des-Lys64,Arg65-proinsulin. None of the tissues contained measurable quantities of Arg32/Glu33- or Arg65/Gly66-split proinsulin. Analysis of plasma from three diabetic subjects managed by the intravenous infusion of human proinsulin revealed less than 1% processing of the circulating precursor to conversion intermediates and no processing of the precursor to human insulin. Nevertheless, analysis of plasma from the same subjects managed by the subcutaneous infusion of proinsulin revealed 4-11% processing of the precursor to intermediates that had the properties of des-Arg31,Arg32-proinsulin and Arg65/Gly66-split proinsulin. We conclude that (a) processing of proinsulin to insulin in vivo as in vitro likely occurs by preferential cleavage at the Arg32-Glu33 peptide bond in proinsulin, (b) proinsulin is inefficiently processed in the vascular compartment, and (c) subcutaneous administration of the precursor can result in the formation of conversion intermediates with the potential for contributing to biological activity.

Lilly lecture 1983. Abnormal products of the human insulin gene.

Heritable abnormalities in the insulin gene have often been considered in terms of their potential for contributing to diabetes. It is only within the last 5 yr, however, that evidence has demonstrated the existence of insulin gene mutations in man and the secretion of abnormal human insulins in affected individuals. HPLC analysis of insulin purified from serum by immuno-affinity chromatography and detection of insulin by radioimmunoassay have documented abnormal insulins in subjects from three separate families. HPLC analysis of these natural insulins and of semisynthetic insulin analogues have identified the accompanying amino acid substitutions in two of these cases: in one, leucine replaces phenylalanine B25; in the other, serine replaces phenylalanine B24. Both substitutions occur in an invariant tetrapeptide sequence within the active site of the hormone. Studies of the biologic activities of these analogues further suggest that replacements at position B25 result in the loss of an important side chain contact between the hormone and its receptor, whereas those at position B24 result in conformational changes in the insulin molecule as a whole. Two additional individuals have been identified to secrete abnormal intermediates of proinsulin conversion in which the C-peptide remains attached to the insulin A-chain and in which Arg65 has been replaced by a nonbasic amino acid. This result emphasizes the importance of dibasic amino acid pairs at prohormone conversion sites and provides clues about the evolution of hormone precursors. Thus, studies of the products of abnormal human insulin genes have provided insights into subjects as varied as insulin biosynthesis, structure-activity relationships in insulin recognition by receptors, and the physiology of insulin action.

Identification of a point mutation in the human insulin gene giving rise to a structurally abnormal insulin (insulin Chicago).

Both insulin gene alleles of a diabetic patient with a mutant insulin were cloned in a lambda vector and their nucleotide sequences were determined. Nucleotide sequence analysis revealed, in one allele, a C (cytidylate) to G (guanylate) transversion in the codon for phenylalanine at position 25 of the insulin B-chain. This point mutation leads to the substitution of a leucine for phenylalanine accompanied by the loss of a restriction endonuclease Mboll recognition site and the creation of a new Rsal cleavage site at this position.

Products of therapeutic insulins in the blood of insulin-dependent (type I) diabetic patients.

The tendency of insulin in high concentrations to self-associate and the widespread presence of insulin-degrading enzymes suggest that fragments and/or aggregates of insulin may circulate in normal and insulin-dependent diabetic (IDDM) individuals. To examine this possibility, we have analyzed, by sensitive physicochemical methods, immunoreactive insulin (IRI) taken from the blood of 9 healthy volunteers and 12 insulin-dependent diabetic patients. IRI from the blood of the normal volunteers was composed of 6000 (91.0 +/- 1.4%) and 9000 (9.0 +/- 1.4%) molecular weight (mol wt) material. By 10% polyacrylamide disc gel electrophoresis (PAGE) and reverse-phase, high-performance liquid chromatography (HPLC), the 6000 mol wt material was indistinguishable from human insulin standards and insulin fragments were not found. C-peptide reactivity in the 9000 mol wt material confirmed the expected presence of proinsulin and intermediates of proinsulin conversion. IRI harvested from the blood of 12 C-peptide-negative IDDMs, using a variety of insulin preparations, also separated into 6000 (80.5 +/- 3.9%) and 9000-12,000 (19.5 +/- 3.9%) mol wt material. By HPLC, 6000 mol wt IRI was either pork insulin (in volunteers using pure pork insulin) or a mixture of beef (approximately 90%), pork (approximately 10%) and deamidated beef (trace) insulin in those using a beef-pork mixture. However, the 9000-12,000 mol wt material had characteristics entirely distinct from proinsulin of either human or animal origin: C-peptide reactivity was undetectable using any of three sensitive radioimmunoassay systems, on PAGE it migrated more rapidly than proinsulin-like material, and in contrast to proinsulin, it was unaffected by proteolytic degradation.(ABSTRACT TRUNCATED AT 250 WORDS)

Lessons learned from molecular biology of insulin-gene mutations.

Studies on naturally occurring and man-made mutations in the insulin gene have provided new insights into insulin biosynthesis, action, and metabolism. Ten families have been identified in which one or more members have single-point mutations in their insulin genes that result in amino acid substitutions within the proinsulin molecule. Six of these cause the secretion of biologically defective insulin molecules due to changes within the A or B chains. Replacing A3-Val with Leu, B24-Phe with Ser, or B25-Phe with Leu results in molecules that have essentially normal immunoreactivity but greatly reduced insulin-receptor-binding potency. Individuals with these mutations have a syndrome of mild diabetes or glucose intolerance, which is inherited in an autosomal-dominant mode and is associated with hyperinsulinemia and altered insulin-C-peptide ratios. Although affected individuals are heterozygous and coexpress both normal and abnormal molecules, the elevated circulating insulin consists mainly of the biologically defective form, which accumulates because it fails to be rapidly metabolized via receptor-mediated endocytosis. Four additional families have mutations that are associated with relatively asymptomatic hyperproinsulinemia. A point mutation affecting proinsulin occurs in 3 of the 4 families, leading to replacement of Arg-65 by His, which prevents recognition of the C-peptide-A-chain dibasic cleavage site by the appropriate beta-cell processing protease and results in the circulation of a type II proinsulin intermediate form (des 64, 65 HPI). Members of a fourth family with hyperproinsulinemia have a substitution of B10-His with Asp, resulting in a proinsulin that exhibits markedly altered subcellular sorting behavior.(ABSTRACT TRUNCATED AT 250 WORDS)

High titer glucagon antisera.

Antibody titers in rabbits immunized with glucagon conjugated to albumin using difluorodinitrobenzene rose rapidly. Under conditions of immunoassay, less than 2 nl of serum from two of four animals and approximately 4 nl from the other two was required to bind 50% of the 10 pg of [125I]iodoglucagon 100 days after immunization. The dissociation constants of the two higher titer antisera for glucagon were approximately 1 x 10(-10)M, and their binding capacities for the hormone, about 50 mug/ml. Competitive binding assays showed that neither of these antisera cross-reacts with the glucagon-like, immunoreactive peptides extracted from intestine to greater than 2.5%. In contrast, hens produced antisera which were reactive with the intestinal material and which bound only 0.3 mug of glucagon per ml. There were no consistent differences, however, in the abilities of specific and non-specific antisera to react with selected fragments of pancreatic glucagon.

Structural determinants of ligand recognition by type I insulin-like growth factor receptors: use of semisynthetic insulin analog probes.

We undertook a systematic analysis of the structural determinants necessary for ligand recognition by the type I insulin-like growth factor (IGF) receptor by investigating the binding of semisynthetic insulin analogs to IGF receptors from human placental cell membrane fragments. Analogs were prepared by synthetic and semisynthetic methods. Three groups of insulin analogs were synthesized: the first group contained insulin analogs modified at the amino-terminal position of the insulin A chain and included acetyl-insulin and human proinsulin; the second group included analogs in which B chain residues B26-B30 [despentapeptide insulin (DPI)], B25-B30 (deshexapeptide insulin), and B24-B30 (desheptapeptide insulin) were removed; the third group contained insulin analogs in which B chain residues B26-B30 were removed (DPI) and phenylalanine(B25) substituted with other amino acids, including alanine, serine, leucine, and tyrosine. Half-maximal inhibition of binding of radiolabeled IGF-I to placental cell membrane fragments was used as an index of relative binding affinity (K1/2). To determine further if semisynthetic insulin analogs bound to the type I IGF receptor, placental membrane fragments were affinity labeled with radiolabeled IGF-I in the presence and absence of submaximal concentrations of unlabeled hormone, insulin, or semisynthetic analogs, and the labeled proteins were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Insulin had a 500 times lower affinity for the type I IGF receptor than IGF-I [K1/2 = 140 +/- 69 nM (mean +/- SD)] whereas proinsulin and acetyl insulin had a more than 100 times lower affinity than insulin for this receptor type. Removal of insulin B chain amino acid residues 26-30 (DPI) did not negatively affect the binding of the insulin-derived peptide and actually increased the apparent affinity of ligand-receptor association approximately 2-fold. However, further removal of phenylalanine(B25) (deshexapeptide insulin) and phenylalanine(B24) (desheptapeptide insulin) decreased the binding of ligand to the type I IGF receptor progressively by several orders of magnitude. Substitution of phenylalanine(B25) of DPI with tyrosine, a substitution that actually increased the homology of this analog to IGF-I, resulted in a 4- to 5-fold increase in the relative apparent affinity of the analog for the type I IGF receptor (K1/2 = 31 +/- 4 nM). On the other hand, substitution of phenylalanine(B25) with alanine, serine, and leucine decreased the relative apparent binding affinity approximately 2- to 8-fold.(ABSTRACT TRUNCATED AT 400 WORDS)

Processing mechanisms in the biosynthesis of proteins.

Limited proteolysis is a widely occurring mechanism in protein biosynthesis. Protein precursors can be classified according to their functions, localization within cell compartments, and enzymic cleavage mechanisms. The presecretory proteins represent an important class of very rapidly turning over precursors which play an early role in the sequestration of secretory products and whose cleavage appears to be intimately associated with structures formed at the ribosome-membrane junction during protein synthesis. A model is proposed which predicts that the prepeptide forms a beta-pleated sheet structure with other components of the membrane which results in the transfer of a loop of peptide across the microsomal membrane. Proinsulin is representative of the general class of proproteins that are processed post-translationally within their secretory cells either during the formation and maturation of secretory granules (peptides hormones and neurotransmitters, serum albumins) or during the assembly of macromolecular structures (virus capsules, membrane-associated enzyme complexes). The former group are cleaved by Golgi-associated proteases having tryptic and carboxypeptidase B-like specificity. Some precursors are secreted as such and processed extracellularly either in the circulation or at special sites (procollagens, zymogens, provenoms, vitellogenins).

Proinsulin radioimmunoassay in the evaluation of insulinomas and familial hyperproinsulinemia.

Two new radioimmunoassays for human proinsulin (hPI) have been developed and used to study patients with islet cell tumors and familial hyperproinsulinemia. Both antisera were adsorbed against human C-peptide conjugated to Sepharose, following which cross-reactivity to insulin and C-peptide was less than 0.001%. Antiserum 18D recognized the junction between the insulin B-chain and C-peptide and provided fivefold greater sensitivity than our previously reported hPI assay. Antiserum 11E recognized a determinant which includes or is adjacent to the A-chain-C-peptide junction or which is specified by the tertiary structure. In all 20 patients studied with surgically confirmed islet cell tumors, fasting plasma proinsulinlike material (PLM) was abnormal (greater than 3 SD from the mean measured in either lean or obese subjects) in both assays. This provided better discrimination than has been reported for PLM measured by gel filtration (abnormal in 13 of 14 of the present samples) with a considerably less laborious procedure. Samples from two families in which a mutant proinsulin is present in the circulation have immunoreactivity in the two assays consistent with previous identification of the molecule as an A-chain-C-peptide-linked intermediate of proinsulin conversion. The immunoreactivity of a sample from another family in which large amounts of proinsulin circulate are consistent with an intact molecule being the predominant form. This assay will be useful for confirming the diagnosis of insulin-secreting tumor in patients suspected of recurrent fasting hypoglycemia and in physiologic studies of proinsulin secretion.

Implications of invariant residue LeuB6 in insulin-receptor interactions.

By the chemical synthesis of modified insulin B chains and the combination of the synthetic B chains with natural insulin A chains, we have prepared insulin analogs with natural and unnatural amino acid replacements of invariant residue LeuB6. Analogs have been investigated by reference to their potencies for interaction with the insulin receptor (as assessed by competition for 125I-labeled binding to isolated canine hepatocytes) and to their abilities to undergo the structural transitions that are characteristic of insulin self-aggregation (as assessed by the spectroscopic analysis of analog complexes with cobalt). Our results identify that (a) replacement of LeuB6 by glycine has nearly the equivalent effect as deletion of residues B1-B6 in decreasing receptor binding potency of the analog to only about 0.05% of that of insulin; (b) relative to the GlyB6 derivative, replacements that increase the relative hydrophobicity of the residue B6 side chain also increase the relative receptor binding potencies of the resulting analogs; (c) negative steric effects resulting from substitutions by valine, phenylalanine, and gamma-ethylnorleucine limit the potential for enhancing potency as the result of increased hydrophobicity; and (d) two analogs with disparate potency for receptor interaction (those with alanine and gamma-ethylnorleucine at position B6, analogs exhibiting about 1 and 48% of the potency of insulin, respectively) undergo the T6----R6 structural transition in the presence of Co2+ and phenol which is typical of insulin but result in hexameric complexes with greatly reduced stability. We conclude that leucine provides a closely determined best fit at insulin position B6, and we discuss our findings in terms of insulin conformations that may apply to the receptor-bound state of the hormone.