Intramolecular Disulphide Bonds Research Articles

Solution synthesis of human midkine, a novel heparin-binding neurotrophic factor consisting of 121 amino acid residues with five disulphide bonds.

Human midkine (hMK), a novel heparin-binding neurotrophic factor consisting of 121 amino acid residues with five intramolecular disulphide bonds, was synthesized by solution procedure in order to demonstrate the usefulness of our newly developed solvent system, a mixture of dichloromethane or chloroform and trifluoroethanol. The final protected 121-residue peptide was assembled from two large fully protected intermediates, Boc-(1-59)-OH and H-(60-121)-OBzl, in CHL/TFE(3:1, v/v) using water-soluble carbodiimide in the presence of HOOBt as coupling reagents. After removal of the protecting groups by HF followed by treatment with Hg(OAc)2 in 50% acetic acid, the fully deprotected peptide was subjected to the oxidative folding reaction. The final product was confirmed to have the correct disulphide structure from its tryptic peptide mapping and to possess the same biological activities as those of the natural product. In order to clarify the active region of the hMK molecule, the N-terminal half domains [(1-59) and (60-121)] were also synthesized by the same procedure used for the hMK synthesis. The C-half domain was confirmed to show the full pattern of bioactivities except for the neuronal cell survival activity, while the N-half one showed much less activity in general.

Solution Synthesis of Human Midkine, a Novel Heparin‐binding Neurotrophic Factor Consisting of 121 Amino Acid Residues with Five Disulphide Bonds

Human midkine (hMK), a novel heparin-binding neurotrophic factor consisting of 121 amino acid residues with five intramolecular disulphide bonds, was synthesized by solution procedure in order to demonstrate the usefulness of our newly developed solvent system, a mixture of dichloromethane or chloroform and trifluoroethanol. The final protected 121-residue peptide was assembled from two large fully protected intermediates, Boc-(1-59)-OH and H-(60-121)-OBzl, in CHL/TFE(3:1, v/v) using water-soluble carbodiimide in the presence of HOOBt as coupling reagents. After removal of the protecting groups by HF followed by treatment with Hg(OAc)2 in 50% acetic acid, the fully deprotected peptide was subjected to the oxidative folding reaction. The final product was confirmed to have the correct disulphide structure from its tryptic peptide mapping and to possess the same biological activities as those of the natural product. In order to clarify the active region of the hMK molecule, the N-terminal half domains [(1-59) and (60-121)] were also synthesized by the same procedure used for the hMK synthesis. The C-half domain was confirmed to show the full pattern of bioactivities except for the neuronal cell survival activity, while the N-half one showed much less activity in general.

Purification and properties of the lipoate protein ligase of Escherichia coli

Lipoate is an essential component of the 2-oxoacid dehydrogenase complexes and the glycine-cleavage system of Escherichia coli. It is attached to specific lysine residues in the lipoyl domains of the E2p (lipoate acetyltransferase) subunit of the pyruvate dehydrogenase complex by a Mg(2+)- and ATP-dependent lipoate protein ligase (LPL). LPL was purified from wild-type E. coli, where its abundance is extremely low (< 10 molecules per cell) and from a genetically amplified source. The purified enzyme is a monomeric protein (M(r) 38,000) which forms irregular clusters of needle-like crystals. It is stable at -20 degrees C, but slowly oxidizes to an inactive form containing at least one intramolecular disulphide bond at 4 degrees C. The inactive form could be re-activated by reducing agents or by an as-yet unidentified component (reactivation factor) which is resolved from LPL at the final stage of purification. The pI is 5.80, and the Km values for ATP, Mg2+ and DL-lipoate were determined. Selenolipoate and 6-thio-octanoate were alternative but poorer substrates. Lipoylation was reversibly inhibited by the 6- and 8-seleno-octanoates and 8-thio-octanoate, which reacted with the six cysteine thiol groups of LPL. LPL was inactivated by Cu2+ ions in a process that involved the formation of inter- and intra-molecular disulphide bonds. Studies with lplA mutants lacking LPL activity indicated that E. coli possesses another distinct lipoylation system, although no such activity could be detected in vitro.

The location of disulphide bonds in α-type gliadins

Gliadin prepared from gluten of the cultivar Rektor by extraction with 70% (v/v) aqueous ethanol adjusted to pH 5.5 was separated by RP-HPLC. Amongst 23 components obtained, two α-type gliadins (α3- and α8-gliadin) were selected for the determination of disulphide bonds. After both proteins were digested with thermolysin, differential RP-HPLC (chromatography prior to and after reduction of disulphide bonds) was used for the detection of cystine peptides. Two cystine peptides from α3-gliadin and three cystine peptides from α8-gliadin were isolated by RP-HPLC. The resulting peptides were reduced and alkylated with 4-vinylpyridine, separated by RP-HPLC and their amino acid sequences determined. The cystine peptides from both α-type gliadins had similar structures, and the corresponding fragments had homologous sequences. One cystine peptide of each gliadin was composed of three fragments linked by two disulphide bonds. The second cystine peptide consisted of two fragments linked by one disulphide bond. The third cystine peptide derived from α8-gliadin was different from the second peptide in one position of the sequences (glutamic acid instead of glutamine). Comparing complete sequences of α-type gliadins described in the literature, the cystine peptides from α3- and α8-gliadins were identical with corresponding sequences of clones A1235 and A212, respectively 11. The structures of the cystine peptides analysed indicate one intramolecular disulphide bond within domain III of α-type gliadins and two disulphide bonds between domains III and V. The linkages found correspond to homologous linkages determined for low M r subunits of glutenin and glutenin-bound γ-type gliadins 6. Obviously, these intramolecular disulphide bonds are not linked randomly, but are strongly directed.

Cationic Bactericidal Peptides

This chapter provides a detailed overview of the known polycationic peptides, with emphasis on those of less than 100 amino acids and with a net charge greater than +2. Cationic peptides can be classified into several groups on the basis of sequence similarities, secondary and tertiary structure, function and origin. These peptides possess antimicrobial activity against many species, including bacteria (Gram-positive and -negative), fungi, and viruses and, in the case of the more potent peptides, can have a lytic activity on mammalian cells. The oxygen-independent microbicidal host defence mechanism of mammals involves several proteinaceous molecules that are cationic in nature. These include lysozyme, bactericidal/permeability increasing factor (BPI), cathepsin G, CAP-37, lactoferrin, defensins and the eosinophil-derived proteins: the major basic protein (MBP) and the eosinophil cationic protein (ECP). Members of the defensin family possess a secondary structure rich in P-pleated sheet, which is stabilized by these intramolecular disulphide bonds. Defensins kill a wide variety of bacteria, fungi, spirochaetes, and viruses. They exert not only microbicidal activity because of permeabilization of biological membranes but they also possess chemotactic and endocrine regulatory activities. A range of inducible antimicrobial cationic peptides has been isolated—including attacins, cecropins, coleoptericin, diptericins, drosocin, phormicins, sarcotoxins, sapecins, and insect defensins. One of the main groups of antibacterial components in the insect humoral response is the cecropins. These molecules constitute a class separate and differ from other bacteriolytic peptides produced by insects by not lysing mammalian cells. The chapter discusses occurrence of cationic peptides in nature, structure-function relationships of cationic bactericidal peptides, and interactions with lipids and membranes with these peptides.

Electrospray ionisation mass spectrometry of lens crystallins: Verification and detection of errors in protein sequences of bovine γ-crystallins

Individual bovine gamma crystallins have been isolated by gel permeation chromatography (GPC) followed by fast protein ion-exchange liquid chromatography (FPLC). Electrospray ionisation (ESI) mass spectrometric examination of the isolated proteins confirms the relative molecular masses (RMM) of three γ-crystallins, i.e. γS, γII (or γB) and γIIIb. The RMM of γIVa and γIIIa, however, are inconsistent with the reported sequences of these proteins. Reduction and carboxymethylation followed by ESI mass spectrometric analysis is shown to be a rapid and convenient method for determining the sulphydryl content of these proteins. In the case of γIIIa and γIVa, this showed these proteins each contain five cysteine residues, compared with ten and six, respectively, in the published sequences. The number of cysteine residues is significant since intramolecular disulphide bonds may have a marked effect on the tertiary structures of these proteins. In addition, the RMM of a protein that we have tentatively assigned as γIVb, which has not been sequenced, is reported here. This study demonstrates the considerable advantages offered by ESI mass spectrometry for confirming published amino acid sequences and detecting sequence errors.

The regulation of catalysis in ATP synthase

ATP synthase is regulated so as to prevent futile hydrolysis of ATP when the transmembrane proton electrochemical gradient, Δμ H +, falls. Mitochondria and chloroplasts have different mechanisms for inhibition of ATP synthase: by binding an inhibitor protein, and by stabilization of the ADP-inhibited state by making an intramolecular disulphide bond, respectively. The recently determined structure of bovine F 1-ATPase is locked in a conformation that probably represents the ADP-inhibited state of the enzyme.

Disulphide bonds in wheat gluten: further cystine peptides from high molecular weight (HMW) and low molecular weight (LMW) subunits of glutenin and from gamma-gliadins.

Glutenin was prepared from gluten of the wheat variety Rektor by extraction of gliadin with aqueous ethanol. It was cleaved successively into soluble peptides by the enzymes trypsin and thermolysin. Separation of the peptide mixtures was performed by gel permeation chromatography (GPC) on Sephadex G25 and reversed phase high performance liquid chromatography (RP-HPLC) on ODS-Hypersil. Cystine peptides were detected by differential chromatography of the samples prior to and after reduction. After isolation by multi-step RP-HPLC, the cystine peptides were reduced. The resulting cysteine peptides were alkylated with 4-vinylpyridine, separated by RP-HPLC and sequenced by means of the Edman degradation. The isolated cystine peptides represented a considerable portion of the total cysteine in glutenin: four out of seven cysteine residues of HMW subunits, and eight out of nine cysteine residues of LMW subunits are documented by at least one cystine peptide. Most of the peptides corresponded to known sequences of gluten protein components. From the structures of some tryptic peptides, inter- and intramolecular disulphide bonds for HMW subunits of glutenin have been proven. Cystine peptides from the thermolytic digest have been assigned to LMW subunits of glutenin and to gamma-gliadins. Other peptides have been closely related to partial sequences of these protein components. The results have allowed several conclusions about the arrangement of intra- and intermolecular disulphide bridges in gluten proteins.

Influence of other whey proteins on the heat-induced aggregation of α-lactalbumin

Abstract The thermally induced aggregation of α-lactalbumin in solution and in the presence of serum protein-free casein micelles has been studied using gel permeation chromatography. No aggregation of α-lactalbumin was detected after heating at 90°C for 24 min. However, addition of β-lactoglobulin or serum albumin to the serum protein-free casein micelles + α-lactalbumin system caused aggregation of the α-lactalbumin, the rate and extent of this aggregation being dependent upon the concentration of free sulphydryl groups present in the other whey protein. It is therefore the sulphydryl group which is important and which appears to function by inducing cleavage of intramolecular disulphide bonds in the α-lactalbumin. This leads to the formation of intermolecular disulphide bridges and hence to aggregation of the α-lactalbumin.

Identification of Human Immunodeficiency Virus Type 1 Glycoprotein gp120/gp41 Interacting Sites by the Idiotypic Mimicry of Two Monoclonal Antibodies

A sequence of four amino acid residues amino-terminal to the only intramolecular disulphide bond of the human immunodeficiency virus type 1 (HIV-1) transmembrane protein gp41 is recognized by an anti-idiotypic antibody (9G5A) raised against another monoclonal antibody (M38), which recognizes the C5 region of gp120. 9G5A is an Ab2 beta antibody (internal image of the M38 epitope) in that it inhibits the interaction of M38 to its antigen. The binding of 9G5A to gp41 can be inhibited by M38 showing that the two antibodies interact via their paratopes. 9G5A neutralizes HIV-1 infection and syncytia formation. Ab3 antibodies induced in mice and rabbits immunized with 9G5A also can neutralize virus in both assays. These data show that the M38-defined epitope of the carboxy-terminal region of gp120 interacts with the 9G5A-defined epitope of gp41, and that this interaction can be reproduced by the idiotypic mimicry of the two antibodies. The results are consistent with a proposed molecular model of the two env regions which predicts the presence, within the C5 region of gp120, of a large intramolecular pocket that is contacted by the gp41 cysteine loop.

A homologue of the Escherichia coli DsbA protein involved in disulphide bond formation is required for enterotoxin biogenesis in Vibrio cholerae.

A strain of Vibrio cholerae, which had been engineered to express high levels of the non-toxic B subunit (EtxB) of Escherichia coli heat-labile enterotoxin, was subjected to transposon (TnphoA) mutagenesis. Two chromosomal TnphoA insertion mutations of the strain were isolated that showed a severe defect in the amount of EtxB produced. The loci disrupted by TnphoA in the two mutant derivatives were cloned and sequenced, and this revealed that the transposon had inserted at different sites in the same gene. The open reading frame of the gene predicts a 200-amino-acid exported protein, with a Cys-X-X-Cys motif characteristic of thioredoxin, protein disulphide isomerase, and DsbA (a periplasmic protein required for disulphide bond formation in E. coli). The V. cholerae protein exhibited 40% identity with the DsbA protein of E. coli, including 90% identity in the region of the active-site motif. Introduction of a plasmid encoding E. coli DsbA into the V. cholerae TnphoA derivatives was found to restore enterotoxin formation, whilst expression of Etx or EtxB in a dsbA mutant of E. coli confirmed that DsbA is required for enterotoxin formation in E. coli. These results suggest that, since each EtxB subunit contains a single intramolecular disulphide bond, a transient intermolecular interaction with DsbA occurs during toxin subunit folding which catalyses formation of the disulphide in vivo.

Selective extracellular release of cholera toxin B subunit by Escherichia coli: dissection of Neisseria Iga beta-mediated outer membrane transport.

The C-terminal domain (Iga beta) of the Neisseria IgA protease precursor is involved in the transport of covalently attached proteins across the outer membrane of Gram-negative bacteria. We investigated outer membrane transport in Escherichia coli using fusion proteins consisting of an N-terminal signal sequence for inner membrane transport, the Vibrio cholerae toxin B subunit (CtxB) as a passenger and Iga beta. The process probably involves two distinct steps: (i) integration of Iga beta into the outer membrane and (ii) translocation of the passenger across the membrane. The outer membrane integrated part of Iga beta is the C-terminal 30 kDa core, which serves as a translocator for both the passenger and the linking region situated between the passenger and Iga beta core. The completeness of the translocation is demonstrated by the extracellular release of the passenger protein owing to the action of the E. coli outer membrane OmpT protease. Translocation of the CtxB moiety occurs efficiently under conditions preventing intramolecular disulphide bond formation. In contrast, if disulphide bond formation in the periplasm proceeds, then translocation halts after the export of the linking region. In this situation transmembrane intermediates are generated which give rise to characteristic fragments resulting from rapid proteolytic degradation of the periplasmically trapped portion. Based on the identification of translocation intermediates we propose that the polypeptide chain of the passenger passes in a linear fashion across the bacterial outer membrane.

The relationship between disulphide bond formation, processing and secretion of lipo-beta-lactamase in yeast.

The hybrid prokaryotic lipo-beta-lactamase mature and precursor proteins spontaneously form an intramolecular disulphide bond when oxidized in vitro. When expressed in Saccharomyces cerevisiae (in vivo) the lipo-beta-lactamase precursor is in a reduced form whereas the majority of the mature protein is oxidized. The results indicate that in yeast, the lipo-beta-lactamase precursor is first processed (the signal peptide is removed) and then oxidized to form a disulphide bond in the mature protein. Reduced-mature lipo-beta-lactamase was found to reach the yeast periplasm and the process depends on endoplasmic reticulum (ER) entry even though the protein is not oxidized. This result is remarkable since in eukaryotes, disulphide bond formation occurs in the ER. Oxidized mature lipo-beta-lactamase can also be released from the sphaeroplast into the yeast periplasm. Mutant lipo-beta-lactamase genes in which cysteine residue 131 was changed to either tyrosine or threonine, were efficiently processed and secreted in yeast, which is consistent with the finding that reduced-mature non-mutant lipo-beta-lactamase can be secreted. We discuss the possibility that the folding mechanism of lipo-beta-lactamase in vitro may be fundamentally different from the process in the eukaryotic system of S. cerevisiae.

A matrix protein of Mr 55 000 that accumulates in human articular cartilage with age

Adult human cartilage contains a protein of M r 55 000 which is deficient in newborn cartilage. In the adult the molecule represents one of the most abundant non-collagenous, non-proteoglycan molecules in 4 M guanidinium chloride extracts of the tissue. The molecular size of the protein on SDS-PAGE remains constant under reducing and non-reducing conditions, suggesting that it does not exist a disulphide-bonded multimer, nor do intramolecular disulphide bonds greatly influence its conformation. The protein has the ability to interact with some immunoglobulin preparations making its detection possible by Western blotting with some non-specific antisera. Labeling with [ 3H]leucine in organ culture indicates that protein of this size is being made by the chondrocytes. However, during purification the newly synthesized molecules do not behave as the resident protein on ion-exchange chromatography, suggesting that the protein may accumulate with age rather than being a major synthetic product of the adult chondrocytes. Amino terminal protein sequence analysis that the N-terminus of the protein is blocked. Sequences derived from peptides generated with cyanogen bromide do not show homology with previously characterized proteins. Molecules of a similar size and composition have been described in bovine cartilage.

Unusual fucoidin-binding properties of chymotrypsinogen and trypsinogen

Previous work (Jones, R. (1987) Cell Biol. Int. Reports 11, 833 and Jones et al. (1988) Development 102, 781–792) has shown that sperm proacrosin (the enzyme form of the acrosomal proteinase acrosin, EC 3.4.21.10) has the capacity to recognize and bind sulphated polysaccharides and that this property is important for the initial stages of fertilization in mammals. To investigate whether this behaviour is specific to proacrosin, a variety of other proteinases (chymotrypsinogen, trypsinogen, thrombin, elastase, plasminogen, pepsin, Streptomyces griseus proteinase and V8 proteinase from Staphylococcus aureus) were immobilized on nitrocellulose and probed with [125I]fucoidin. Only chymotrypsinogen and trypsinogen retained significant amounts of the probe with Kd values of 1.4 · 10−6 M and 3.0 · 10−5 M, respectively. Proteinase inhibitors were ineffective as blocking agents suggesting that enzymic activity is not involved in recognition. However, the tertiary structure of the proteins is important, since cleavage of intramolecular disulphide bonds with 2-mercaptoethanol reduced binding by 50–60%. Competition experiments with a variety of mono- and polysaccharides suggest that the number and disposition of sulphate groups is critical for interaction with basic residues on the protein. It is concluded that, like proacrosin, chymotrypsinogen and trypsinogen are bifunctional proteins.

Immunological relationships between receptors for insulin and insulin-like growth factor I. Evidence for structural heterogeneity of insulin-like growth factor I receptors involving hybrids with insulin receptors

The receptors for insulin and insulin-like growth factor-I (IGF-I) are closely related in primary sequence and overall structure. We have examined the immunological relationships between these receptors by testing the reactivity of anti-(insulin receptor) monoclonal antibodies with IGF-I receptors in various tissues and cell lines. Antibodies for six distinct epitopes reacted with a subfraction of IGF-I receptors, as shown by inhibition of 125I-IGF-I binding, precipitation of 125I-IGF-I-receptor complexes or immunodepletion of receptor from tissue extracts before binding assays. Both immunoreactive and non-immunoreactive subfractions displayed the expected properties of 'classical' IGF-I receptors, in terms of relative affinities for IGF-I and insulin. The proportion of total IGF-I receptors which was immunoreactive varied in different cell types, being approx. 40% in Hep G2 cells, 35-40% in placental membranes and 75-85% in IM-9 cells. The immunoreactive fraction was somewhat higher in solubilized receptors than in the corresponding intact cells or membranes. A previously described monoclonal antibody, alpha-IR-3, specific for IGF-I receptors, inhibited IGF-I binding by more than 80% in all preparations. When solubilized placental receptors were pretreated with dithiothreitol (DTT) under conditions reported to reduce intramolecular (class I) disulphide bonds, the immunoreactivity of IGF-I receptors was abolished although total IGF-I binding was little affected. Under the same conditions insulin receptors remained fully immunoreactive. When solubilized receptor preparations were fractionated by gel filtration, both IGF-I and insulin receptors ran as symmetrical peaks of identical mobility. After DTT treatment, the IGF-I receptor was partially converted to a lower molecular mass form which was not immunoreactive. The insulin receptor peak showed a much less pronounced skewing and remained fully immunoreactive in all fractions. It is concluded that the anti- (insulin receptor) antibodies do not react directly with IGF-I receptor polypeptide, and that the apparent immunoreactivity of a subfraction of IGF-I receptors reflects their physical association with insulin receptors, both in cell extracts and in intact cells. The most likely basis for this association appears to be a 'hybrid' receptor containing one half (alpha beta) of insulin receptor polypeptide and the other (alpha' beta') of IGF-I receptor polypeptide within the native (alpha beta beta' alpha') heterotetrameric structure.

2 The molecular biology and biochemistry of tissue factor

The cloning of tissue factor (TF) cDNAs has recently been carried out which has allowed the characterization of this cellular receptor for factor VII. The predicted primary structure indicates that it is a multidomain protein comprised of a hydrophilic extracellular domain, a membrane spanning structure and a cytoplasmic region. Post-translational modifications include four potential N-linked glycosylation sites of which three are occupied. There is evidence that the intracellular cysteine residue is thioester-linked to palmitic acid, and that this could enhance the anchoring of the receptor to the membrane. The four remaining cysteine residues are located in the putative extracellular domain and appear to form two intramolecular disulphide bonds. There is no significant homology between TF and any other published protein sequence in current databases. There is evidence that TF is the primary initiator of the coagulation cascade. Early models of coagulation tended to assign TF to a subordinate role, but now it is widely accepted that the extrinsic pathway is critical as TF/FVIIa activates both FX and FIX. It has been reported that zymogen FVII has catalytic activity and this would indicate that the simple complexing of FVII and TF is sufficient to initiate coagulation without an infinite regression of proteolytic events. Many cell types synthesize TF constitutively but, significantly, endothelial cells do not normally express TF on their surfaces, consistent with a quiescent haemostatic system. Recently a number of agonists known to be thrombogenic in vivo have been shown to induce de novo synthesis of TF by endothelial cells. TF thus operates at two levels: exposure of the subendothelium to blood results in the binding of FVII to TF and consequently to clot formation, and induction of TF synthesis by a variety of agonists results in the intact endothelium becoming procoagulant. In concert with TF expressed by monocytes and macrophages this endothelial cell procoagulant activity may play a role in the pathogenesis of thrombotic disease.

The occurrence of disulphide bonds in purified clathrin light chains.

Three forms of clathrin light chain contain two cysteine residues. These are the predominant brain-specific forms of LCa and LCb and the non-brain form of LCb. After purification in the absence of thiols they contain intramolecular disulphide bonds. The reduced and the oxidized forms show differences in electrophoretic mobility, explaining the variable and heterogeneous patterns observed on electrophoresis. Accessibility of the thiol groups in the free light chains is greater than when they are associated with the heavy chain. In contrast the cysteine residues of the clathrin heavy chain are completely inaccessible in the absence of denaturants and are not found in disulphide bonds. The antigenic properties of the oxidized and the reduced forms of the clathrin light chains are similar, as is their capacity to bind to the clathrin heavy chain. After isolation in the presence of 10 mM-iodoacetamide, the light-chain cysteine residues are fully alkylated. The results are consistent with the reduced form being the native state and the light-chain disulphide bonds an artifact of isolation.

Stufenweise Reduktion von Weizenglutelin mit Mercaptoethanol und Dithioerythrit

The fractions of glutenin, from the wheat variety Rektor obtained by gel permeation chromatography after stepwise reduction with increasing concentrations of mercaptoethanol (6–150 mmol/1) and with 35 mmol/1 dithioerythritol, [this journal (1988), 187:20], were further investigated by reversed-phase high performance liquid chromatography (RP-HPLC) after complete reduction with dithioerythritol in the presence of urea. Glutenin samples can be separated into high- (HMW), middle- (MMW) and low-molecular-weight (LMW) subunits [this journal (1987), 185:487] by this method. Only 16% of the HMW but 49% of the LMW subunits can be separated from the highly aggregated part of glutenin (fractions I and II) by treatment of the glutenin at pH 8 with 6mol/l urea. Without any reducing agent, they occur in the fractions III to Vb. After the reduction of glutenin by incubation with 6 mmol/l mercaptoethanol, the HMW and LMW subunits in fraction I decrease, leaving a residual protein in this fraction which is atypical for glutenin. In contrast, fraction II is more typical for glutenin since increasing amounts of HMW and decreasing amounts of LMW subunits are present. By reduction of glutenin with 24 mmol/1 or higher concentrations of mercaptoethanol, fraction II also becomes atypical for glutenin. After complete reduction of glutenin, most of the HMW subunits occur in fraction III, whereas fraction IV contains most of the LMW subunits. The distribution of the MMW subunits is different from that of the HMW and LMW subunits. In the presence of 6mol/1 urea, without any reducing agent, 55% of the MMW subunits occur in fraction III. Total reduction does not cause any significant change in fraction III, but most of the remaining MMW subunits are shifted from fractions I and II to fraction IV. The results show that inter- and/or intramolecular disulphide bonds are very important for glutenin aggregation, but obviously secondary forces are also involved. The aggregation of the HMW subunits is more dependent on disulphide bonds than the LMW and MMW subunits.

Defective co-translational formation of disulphide bonds in protein disulphide-isomerase-deficient microsomes.

The formation of disulphide bonds in mammalian secretory and cell-surface proteins occurs in the lumen of the endoplasmic reticulum and is believed to be catalysed by the enzyme protein disulphide-isomerase (PDI). The evidence for this physiological role for PDI is circumstantial and relates to the cell and tissue distribution of the enzyme, its developmental behaviour and its catalytic properties in vitro. A clear requirement for PDI in the correct folding or assembly of disulphide-bonded proteins during biosynthesis has not been demonstrated. We have prepared dog pancreas microsomes which are deficient in soluble lumenal proteins, including PDI, but which are still able to translocate and process proteins synthesized in vitro. Using the formation of intramolecular disulphide bonds during the in vitro synthesis of gamma-gliadin, a wheat storage protein, as a model, we have demonstrated that these microsomes are defective in co-translational formation of disulphide bonds. Reconstitution of these microsomes with purified PDI reverses this defect.