Kinetic analysis of the mechanism of insulin degradation by glutathione-insulin transhydrogenase (thiol: protein-disulfide oxidoreductase).

Kinetic studies have been made with glutathione-insulin transhydrogenase, an enzyme which degrades insulin by promoting cleavage of its disulfide bonds via sulfhydryl-disulfide interchange. The degradation of 125I-labeled insulin by enzyme purified from beef pancreas was studied with various thiol-containing compounds as cosubstrates. The apparent Km for insulin was found to be a function of the type and concentration of thiol; values obtained were in the range from 1 to 40 muM. Lineweaver-Burk plots for insulin as varied substrate were linear, whereas those for the thiol substrates were nonlinears: the plots for low molecular weight monothiols (GSH and mercaptoethanol) were parabolic; those for low molecular weight dithiols (dithiothreitol, dihydrolipoic acid, and 2,3-dimercaptopropanol) were apparently linear modified by substrate inhibition; and the plots for protein polythiols (reduced insulin A and B chains and reduced ribonuclease) were parabolic with superposed substrate inhibition. The nonparallel nature of the reciprocal plots for all substrates shows that the enzyme does not follow a ping-pong mechanism. Product inhibition studies were performed with GSH as thiol substrate. Oxidized glutathione was found to be a linear competitive inhibitor vs. both GSH and insulin. The S-sulfonated derivative of insulin A chain was also linearly competitive vs. both substrates. Inhibition by S-sulfonated B chain was competitive vs. insulin; the data eliminated the possibility that this derivative was uncompetitive vs. GSH. Experiments with the cysteic acid derivatives of insulin A and B chains similarly excluded the possibility that these were uncompetitive vs. either substrate. These inhibition studies indicate that the enzyme probably follows a randdom mechanism.     

Insulin and glucagon degradation by the kidney. II. Characterization of the mechanisms at neutral pH.

Examination of insulin and glucagon degradation by rat kidney subcellular fractions revealed that most degrading activity was localized to the 100 000 X g pellet and 100 000 X g supernatant fractions. Further characterization of the degrading activities of the 100 000 X g pellet and supernatant suggested that three types of enzymatic activity were present at neutral pH. From the cytosol an enzyme with characteristics of the insulin glucagon protease of skeletal muscle was purified. This enzyme appeared to be responsible for insulin degradation by the kidney at physiological insulin concentrations. This enzyme also contributed to glucagon degradation but was not the most active mechanism for this. In the 100 000 X g pellet at least two separate enzymatic activities were present. One of these had properties consistent with those described for glutathione insulin transhydrogenase and appeared to be responsible for insulin degradation at high insulin concentration. The other enzyme was associated with the brush border and had properties consistent with the brush border neutral protease. This enzyme appeared responsible for glucagon degradation at both low and high substrate concentrations. An apparent marked synergism between the 100 000 X g pellet and the 100 000 X g supernatant was noted for insulin degradation at physiological insulin concentrations. Pellet glucagon-degrading activity and soluble insulin-degrading activity were necessary for this. The mechanism was found to be limited insulin degradation by the soluble enzyme resulting in both trichloroacetic acid-precipitable trichloroacetic acid-soluble fragments followed by further degradtion of the fragments by the glucagon-degrading enzyme resulting in an additional increase in trichloroacetic acid-soluble products.     

Catalysis of thiol/disulfide exchange: single-turnover reduction of protein disulfide-isomerase by glutathione and catalysis of peptide disulfide reduction

Protein disulfide-isomerase, a protein localized to the lumen of the endoplasmic reticulum of eukaryotic cells, catalyzes the posttranslational formation and rearrangement of protein disulfide bonds. As isolated from bovine liver, the enzyme contains 0.8 free sulfhydryl group per mole of protein monomer and 3.1 disulfide bonds. Single-turnover experiments in which the disulfide bonds of the native enzyme are reduced by glutathione reveal three distinct reduction steps corresponding to the sequential reduction of the three disulfide bonds. The fastest disulfide to be reduced undergoes a change in the rate-determining step with increasing GSH concentration from a step which is second-order with respect to GSH concentration to a step which is first-order in GSH concentration. The disulfide which is reduced at an intermediate rate displays kinetics that are first-order in GSH concentration, and the slowest disulfide to be reduced exhibits kinetics which are second-order in GSH concentration. The enzyme catalyzes the steady-state reduction of a disulfide-containing hexapeptide (CYIQNC) by GSH. Initial velocity kinetic experiments are consistent with a sequential addition of the substrates to the enzyme. Saturation behavior is not observed at high levels of both substrates (Km for GSH much greater than 14 mM, Km for CYIQNC much greater than 1 mM). Only one of the three disulfides appears to be kinetically competent in the steady-state reduction of CYIQNC by GSH. The second-order thiol/disulfide exchange reactions catalyzed by the enzyme are 400-6000-fold faster than the corresponding uncatalyzed reactions.     

Analysis of catalytic properties of hen egg white lysozyme during renaturation from denatured and reduced material.

Catalytic properties of hen egg white lysozyme were analyzed during the renaturation of the enzyme from completely reduced and denatured material. The formation of intermediate folding products and the generation of native lysozyme was monitored by acetic acid/urea electrophoresis. The results showed that during the beginning of renaturation almost all reduced and denatured lysozyme is converted to forms possessing lower compactness than native lysozyme, probably as a result of formation of only one or two disulfide bonds. Kinetic analysis of lysozyme during renaturation showed that the generation of lysozyme with four disulfide bonds was not necessarily equivalent to the formation lysozyme with native-like catalytic properties. It appeared that the formation rate of the structures of the structures of the substrate binding site and of the catalytic site were limited by the generation of four disulfide bonds containing lysozyme. The catalytic properties of intermediate folding products made it evident that the final structures of the substrate binding site and of the catalytic site were formed after the generation of all disulfide bonds.     

Modification of the tritium-release assays for prolyl and lysyl hydroxylases using Dowex-50 columns.

The tritium release assay of Hutton et al. (Anal. Biochem.16, 384, 1966) for prolyl hydroxylase and of Miller (Anal. Biochem.45, 202, 1972) for lysyl hydroxylase have been modified. The reaction is carried out on a microscale, and tritiated water is collected after passage of the trichloroacetic acid-soluble reaction products through a small Dowex-50 (H⁺) column instead of using a vacuum distillation apparatus as described in the original procedures. When measured by the modified procedures, both the prolyl and lysyl hydroxlyase reactions showed regions of linearity with respect to enzyme concentration, time and substrate concentration and were almost completely dependent on ascorbate and α-ketoglutarate. In addition, both reactions were completely inhibited by the iron chelator, α,α′-dipyridyl. The results indicate that these assay procedures are valid means of measuring prolyl and lysyl hydroxylase activities.     

The steady-state kinetics of yeast phosphoglycerate kinase. Anomalous kinetic plots and the effects of salts on activity.

1. A re-investigation of the kinetics of yeast phosphoglycerate kinase in the direction of 1,3-bisphosphoglycerate formation has been carried out, covering a 1000-fold range in substrate concentrations. A variety of improved spectrophotometric and fluorimetric assay procedures have been used. 2. Kinetic plots proved to be non-linear for each variable substrate. A variety of checks have been carried out to show that this is not due to artifacts in the assay procedures or heterogeneity of the enzyme preparation. 3. The effects of a variety of salts on the activity of the enzyme have been examined. Most salts, especially those with multivalent anions, can cause activation of the enzyme, but inhibit at high concentration. 4. The salt effect is shown to be principally due to anions rather than cations, and not to ionic strength changes. Sulphate, as one of the most effective anions has been used in most comparisons. 5. Salt activation is steepest when the substrate concentrations are low; maximum activation has been about 5-fold with 0.2 mM MgATP and 0.2 mM 3-phosphoglycerate. Inhibition at the higher salt concentrations is strongest at the same substrate concentrations as when activation is steepest, indicating a link between the two effects. 6. The presence of 20 mM or more Na2SO4 converted non-linear kinetic plots to linear ones. A study of the kinetics in the presence of 40 mM Na2SO4 was interpreted in terms of a random sequential binding mechanism, with sulphate acting as a competitive inhibitor. 7. Possible explanations for these anomalous results are discussed in terms of several mechanisms which have been shown to apply in other systems.     

Correlation of Carbohydrate Catabolism and Synthesis of Macromolecules During Enzyme Synthesis in Pseudomonas fluorescens.

Kirkland, Jerry J. (Oklahoma State University, Stillwater), and Norman N. Durham. Correlation of carbohydrate catabolism and synthesis of macromolecules during enzyme synthesis in Pseudomonas fluorescens. J. Bacteriol. 90: 23-28. 1965.-Glucose, ribose, and fructose shorten the lag period required for synthesis of protocatechuate oxygenase. Radioactivity from uracil-2-C(14) is incorporated into the hot trichloroacetic acid-soluble fraction after a lag period of approximately 20 min after addition of protocatechuic acid. Addition of glucose or ribose simultaneously with the inducer shortens the lag period to approximately 5 min and increases the rate of uracil incorporation. The inducer must be present to initiate incorporation of radioactivity, and the exogenous carbon source accelerates incorporation but is not sufficient to initiate synthesis by itself. The addition of protocatechuic acid increases the rate and total incorporation of radioactivity from uniformly labeled glucose or ribose-1-C(14) into the hot trichloroacetic acid-soluble fraction. Ribose decreases the incorporation of radioactivity from uniformly labeled glucose into the hot trichloroacetic acid-soluble fraction, and glucose shows a similar effect on incorporation of radioactivity from ribose-1-C(14), indicating the two sugars are serving in the same capacity to enhance enzyme synthesis. Radioactivity from glucose-1-C(14) is not incorporated into the hot trichloroacetic acid-soluble fraction. The results suggest that glucose and ribose shorten the lag period for inducible enzyme formation by serving as a "specific" carbon source for synthesis of macromolecules such as ribonucleic acid.     

Rabbit erythrocyte purine nucleoside phosphorylase. Initial-velocity studies.

1. Concave-downward double-reciprocal plots were obtained for rabbit erythrocyte purine nucleoside phosphorylase when the concentration of Pi was varied over a wide range at a fixed saturating concentration of either inosine or deoxyinosine. Similar behaviour was also displayed by the calf spleen enzyme. 2. The degree of curvature of double-reciprocal plots was greatly modified by the presence of SO42-, introduced into the assay mixture with the linking enzyme xanthine oxidase; competitive inhibition by SO42- was observed over a narrow range of high Pi concentrations. 3. Partial inactivation with 5,5'-dithiobis-(2-nitrobenzoic acid) resulted in a marked alteration in the kinetic properties of the enzyme when Pi was the variable substrate. 4. Initial-velocity data are expressed in the form of Hill plots, and the significance of such plots is discussed.     

Characterization of redox state and reductase activity of protein disulfide isomerase under different redox environments using a sensitive fluorescent assay

In this study, dieosin glutathione disulfide (Di-E-GSSG) was synthesized by the reaction of eosin isothiocyanate with GSSG. Di-E-GSSG had low fluorescence which increased approximately 70-fold on reduction of its disulfide bond. The substrate was used to monitor the disulfide reductase activity of PDI. Di-E-GSSG is the most sensitive pseudo substrate for PDI reductase activity reported to date. This probe was further used as an analytical reagent to develop an end point assay for measuring the redox state of PDI. The reduction of Di-E-GSSG by reduced enzyme was studied in the absence of reducing agents and the redox state of PDI was monitored as a function of the stoichiometric changes in the amount of eosin-glutathione (EGSH) generated by the active-site dithiols of PDI. The redox state of PDI was also studied under variable [GSH]/[GSSG] ratios. The results indicate that PDI is in approximately 1/2-reduced state where the [GSH]/[GSSG] ratio is between 1:1 and 3:1, conditions similar to the lumen of endoplasmic reticulum or in the extracellular environment. On the other hand, [GSH]/[GSSG] ratios of > or =8:1, such as in cytosol, all active-site thiols would be reduced. The study was extended to utilize Di-E-GSSG to investigate the effect of variable redox ratios on the platelet surface PDI reductase activity.     

Effects of the aminonucleoside of puromycin on kidney cortex sulfhydryl and disulfide-containing components and their respective oxidoreductase activities.

Comparative study of glutathione reductase and of mixed disulfide reducing activity assayed with a bovine serum albumin-glutathione substrate in kidney cortex of normal and of aminonucleoside-nephrotic rats reveals significantly higher activities in the latter. Total acid-soluble sulfhydryl and reduced glutathione (GSH) were also found to be significantly higher in kidney cortex of the aminonucleoside-treated rats. Neither total mixed disulfides nor mixed disulfides in which glutathione is covalently bound by disulfide linkage to kidney cortex protein are significantly altered by the nephrosis-producing aminonucleoside.     

Catalysis of the oxidative folding of ribonuclease A by protein disulfide isomerase: pre-steady-state kinetics and the utilization of the oxidizing equivalents of the isomerase

At low concentrations of a glutathione redox buffer, the protein disulfide isomerase (PDI) catalyzed oxidative renaturation of reduced ribonuclease A exhibits a rapid but incomplete activation of ribonuclease, which precedes the steady-state reaction. This behavior can be attributed to a GSSG-dependent partitioning of the substrate, reduced ribonuclease, between two classes of thiol/disulfide redox forms, those that can be converted to active ribonuclease at low concentrations of GSH and those that cannot. With catalytic concentrations of PDI and near stoichiometric concentrations of glutathione disulfide, approximately 4 equiv (2 equiv of ribonuclease disulfide) of GSH are formed very rapidly followed by a slower formation of GSH, which corresponds to an additional 2 disulfide bond equiv. The rapid formation of RNase disulfide bonds and the subsequent rearrangement of incorrect disulfide isomers to active RNase are both catalyzed by PDI. In the absence of GSSG or other oxidants, disulfide bond equivalents of PDI can be used to form disulfide bonds in RNase in a stoichiometric reaction. In the absence of a glutathione redox buffer, the rate of reduced ribonuclease regeneration increases markedly with increasing PDI concentrations below the equivalence point; however, PDI in excess over stoichiometric concentrations inhibits RNase regeneration.     

Role of disulfide interchange enzyme in immunoglobulin synthesis

The role of disulfide interchange enzyme in protein biosynthesis was evaluated by studying the enzyme from mouse lymphoid tissue. The enzyme isolated from lymphoid cells was shown to have no tissue-specific characteristics. It was identical with the enzyme synthesized by mouse liver in its biochemical and immunological properties and its capacity to promote both disulfide bond formation and insulin degradation. In contrast to liver, the levels of enzyme in lymphoid tissues were found to vary with immunoglobulin secretory activity, Assays of lymphoid cells and their transformed counterparts showed that the enzyme contents of cells actively secreting immunoglobulin were 1-2 orders of magnitude higher than that of unstimulated B cells or non-immunoglobulin-producing T cells. The increase in enzyme levels paralleled the increase in immunoglobulin synthesis after antigen or mitogen stimulation and was independent of the class of immunoglobulin produced. This correlation indicated that the enzyme plays a critical role in the formation of intramonomer bonds common to all immunoglobulin molecules. Supporting data were obtained by assaying the ability of the enzyme to promote the polymerization of mouse pentamer IgM in vitro. The enzyme was found to catalyze the formation of the interchain bonds required for monomer IgM assembly but not the formation of the intermonomer bonds required for pentamer assembly. The sum of these results provides strong evidence that disulfide interchange enzyme functions in the in vivo synthesis protein disulfide bonds.     

Egg white sulfhydryl oxidase: kinetic mechanism of the catalysis of disulfide bond formation

The flavin-dependent sulfhydryl oxidase from chicken egg white catalyzes the oxidation of sulfhydryl groups to disulfides with reduction of oxygen to hydrogen peroxide. The oxidase contains FAD and a redox-active cystine bridge and accepts a total of 4 electrons per active site. Dithiothreitol (DTT; the best low molecular weight substrate known) reduces the enzyme disulfide bridge with a limiting rate of 502/s at 4 degrees C, pH 7.5, yielding a thiolate-to-flavin charge-transfer complex. Further reduction to EH4 is limited by the slow internal transfer of reducing equivalents from enzyme dithiol to oxidized flavin (3.3/s). In the oxidative half of catalysis, oxygen rapidly converts EH4 to EH2, but Eox appearance is limited by the slow internal redox equilibration. During overall turnover with DTT, the thiolate-to-flavin charge-transfer complex accumulates with an apparent extinction coefficient of 4.9 mM-1 cm-1 at 560 nm. In contrast, glutathione (GSH) is a much slower reductant of the oxidase to the EH2 level and shows a kcat/Km 100-fold smaller than DTT. Full reduction of EH2 by GSH shows a limiting rate of 3.6/s at 4 degrees C comparable to that seen with DTT. Reduced RNase is an excellent substrate of the enzyme, with kcat/Km per thiol some 1000- and 10-fold better than GSH and DTT, respectively. Enzyme-monitored steady-state turnover shows that RNase is a facile reductant of the oxidase to the EH2 state. This work demonstrates the basic similarity in the mechanism of turnover between all of these three substrates. A physiological role for sulfhydryl oxidase in the formation of disulfide bonds in secreted proteins is discussed.     

Assay of thiols and disulfides based on the reversibility of N-ethylmaleimide alkylation of thiols combined with electrolysis.

A simple and specific method for analyzing thiols and disulfides on the basis of the reversibility of N-ethylmaleimide (NEM) alkylation of thiols is described. When the adduct of NEM and glutathione (GSH) was electrolyzed at neutral pH, all of the GSH was recovered. When the adduct was exposed to pH 11.0 for 15 min at 30 degrees C before electrolysis, GSH was not detected. The same behavior was observed after protein thiols reacted with NEM. This pH-dependent production of thiol from the adduct was used to assay GSH and oxidized glutathione in yeast cells, to assay sulfhydryl groups and disulfide bonds in authentic proteins, and to protect thiols from oxidation during enzymatic digestion of protein. This method is useful for assay of thiols and disulfides of both small and large molecules and can be used to identify labile thiols in biological samples that are oxidized during extraction procedures.     

Reduced glutathione in chinese hamster ovary cells protects against inactivation of 3-hydroxy-3-methylglutaryl Coenzyme A reductase by 2-mercaptoethanol disulfide

When the disulfide of 2-mercaptoethanol (ESSE) is added to the medium of cultured Chinese hamster ovary (CHO) cells, a time and concentration dependent release of 2-mercaptoethanol to the medium is observed. The reduction of ESSE to 2-mercaptoethanol by cells is a saturable process, the rate being approximately 50 nmoles of 2-mercaptoethanol per mg cell protein for an hour upon exposure to 250 μM ESSE. Reduction rate of ESSE by cells attached to a substratum is independent of glucose and insulin for periods up to 4 hours. However, in detached cells, swirled in suspension, addition of glucose and insulin is necessary in order to obtain a linear reduction rate of ESSE. The rate limiting enzyme in the sterol biosynthetic pathway, 3-hydroxy-3-methyl-glutaryl Coenzyme A reductase (E.C. 1.1.1.34), is inhibited by ESSE when isolated from CHO cells but total nonsaponifiable lipids synthesis from [2?¹⁴C]-acetate in intact cells is not affected by ESSE at concentrations up to 500 μM. Cytosolic reduced glutathione can spontaneously exchange disulfide bonds with ESSE and thus prevent it from inhibiting the reductase. Cultured cells respond to ESSE administration by elevating their total and acid-soluble glutathione levels. The use of ESSE as a perturbant of the GSH Status in cells is discussed.     

A high-throughput turbidometric assay for screening inhibitors of protein disulfide isomerase activity

Protein disulfide isomerase (PDI) plays a key role in protein folding by catalyzing rearrangements of disulfide bonds in substrate proteins following their synthesis in eukaryotic cells. Besides its major role in the processing and maturation of secretory proteins in the endoplasmic reticulum, this enzyme and its homologs have been implicated in multiple important cellular processes; however, they have not served as targets for the development of therapeutic agents. The authors developed a high-throughput screening assay for PDI and its homologous enzymes in 384-well microplates. The method is based on the enzyme-catalyzed reduction of insulin in the presence of dithiothreitol and measures the aggregation of reduced insulin chains at 650 nm. This kinetic assay was converted to an end-point assay by using hydrogen peroxide as a stop reagent. The feasibility of this high-throughput assay for screening chemical libraries was demonstrated in a pilot screen. The authors show that this homogenous turbidometric assay is robust and cost-effective and can be applied to identify PDI inhibitors from chemical libraries, opening this class of enzymes for therapeutic exploration.     

Insulin degradation VII. Sequential degradation of insulin by rat liver homogenates at physiologic concentrations of insulin and in the absence of exogenous glutathione

Insulin is degraded by rat liver homogenates in a sequential manner: first, splitting of insulin at the disulfide bonds by GSH-insulin transhydrogenase, and second, proteolysis of the resultant A and B polypeptide chains to small molecular weight components. This sequential pattern of insulin degradation has been found to take plase not only at a high concentration of insulin (10^?5 M) in the presence of GSH, but also at low, physiologic concentations of insulin (10^?9 M) and in the absence of exogenous GSH. The study re-emphasizes the necessity of use of the proper relative concentrations of tissue protein and insulin for the demonstration of the sequential pattern. The data also indicate that the proteolysis of both A and B chains of insuli might be effected by the same enzyme system.     

Biosynthesis of hyaluronic acid by Streptococcus.

Synthesis of hyaluronic acid was investigated in a cell-free system derived from a strain of Group A streptococci. Preparative procedures were improved so that an enzyme system 70 times more active than that previously reported was obtained. The hyaluronic acid synthesized could be separated into trichloroacetic acid-soluble and -insoluble fractions. On the basis of pulse-chase experiments, it was shown that the trichloroacetic acid-insoluble fraction is a precursor of the soluble fraction. The release of the trichloroacetic acid-insoluble hyaluronic acid is specifically blocked with p-chloromercuribenzoate, without inhibition of chain elongation. The addition of butanol to trichloroacetic acid resulted in solubilization of all of the hyaluronic acid. No detectable difference in molecular size was observed between the two hyaluronic acid fractions, both of which were estimated to be more than one million daltons in size. Testicular hyaluronidase digestion of either one of the two types of hyaluronic acid yielded no high molecular weight fragments, indicating that hyaluronic acid is not bound covalently to protein. However, following incubation of enzyme assay mixtures with UDP-[14C]GlcUA, even in the absence of UDP-GlcNAc, radioactive high molecular weight hyaluronic acid was obtained which suggests that the enzyme system elongates rather than initiates hyaluronic acid chains. Tunicamycin did not inhibit hyaluronic acid synthesis, indicating lack of participation of an intermediate of pyrophosphorylpolyisoprenol type. The results obtained are consistent with the hypothesis that chain elongation of hyaluronic acid proceeds by alternate addition of monosaccharides from UDP-sugars by a membrane-bound synthesizing system followed by release of completed hyaluronic acid chains.     

Putative disulfide-forming pathway of porcine insulin precursor during its refolding in vitro

Although the structure of insulin has been well studied, the formation pathway of the three disulfide bridges during the refolding of insulin precursor is ambiguous. Here, we reported the in vitro disulfide-forming pathway of a recombinant porcine insulin precursor (PIP). In redox buffer containing L-arginine, the yield of native PIP from fully reduced/denatured PIP can reach 85%. The refolding process was quenched at different time points, and three distinct intermediates, including one with one disulfide linkage and two with two disulfide bridges, have been captured and characterized. An intra-A disulfide bridge was found in the former but not in the latter. The two intermediates with two disulfide bridges contain the common A20-B19 disulfide linkage and another inter-AB one. Based on the time-dependent formation and distribution of disulfide pairs in the trapped intermediates, two different forming pathways of disulfide bonds in the refolding process of PIP in vitro have been proposed. The first one involves the rapid formation of the intra-A disulfide bond, followed by the slower formation of one of the inter-AB disulfide bonds and then the pairing of the remaining cysteines to complete the refolding of PIP. The second pathway begins first with the formation of the A20-B19 disulfide bridge, followed immediately by another inter-AB one, possibly nonnative. The nonnative two-disulfide intermediates may then slowly rearrange between CysA6, CysA7, CysA11, and CysB7, until the native disulfide bond A6-A11 or A7-B7 is formed to complete the refolding of PIP. The proposed refolding behavior of PIP is compared with that of IGF-I and discussed.     

Role of disulfide bonds in the structure and activity of human insulin

Insulin contains two inter-chain disulfide bonds between the A and B chains (A7-B7 and A20-B19), and one intra-chain linkage in the A chain (A6-A11). To investigate the role of each disulfide bond in the structure, function and stability of the molecule, three des mutants of human insulin, each lacking one of the three disulfide bonds, were prepared by enzymatic conversion of refolded mini-proinsulins. Structural and biological studies of the three des mutants revealed that all three disulfide bonds are essential for the receptor binding activity of insulin, whereas the different disulfide bonds make different contributions to the overall structure of insulin. Deletion of the A20-B19 disulfide bond had the most substantial influence on the structure as indicated by loss of ordered secondary structure, increased susceptibility to proteolysis, and markedly reduced compactness. Deletion of the A6-A11 disulfide bond caused the least perturbation to the structure. In addition, different refolding efficiencies between the three des mutants suggest that the disulfide bonds are formed sequentially in the order A20-B19, A7-B7 and A6-A11 in the folding pathway of proinsulin.