Disulfide bonds play crucial roles to predict the three-dimensional structure and the function of a protein. This thesis develops two algorithms to predict the disulfide bonding state of each cysteine in a protein sequence. These methods are based on the multi-stage framework and the multi-classifier of the support vector machine (SVM). The first algorithm achieves 94.0% accuracy of cysteine state prediction for dataset PDB4136, but in some datasets the results are not as good as our expectation. Thus the second algorithm is designed to improve the predicting ability for the proteins which have oxidized and reduced cysteines simultaneously. In addition, a new training strategy is also developed to increase the prediction accuracy. It appends the probabilities which are obtained from the SVM to the existing features and then starts a new training procedure repeatedly to get better performance. The experiments are performed on the datasets derived from well-known databases, such as Protein Data Bank and SWISS-PROT. It gets 94.3% accuracy for predicting disulfide bonding state on dataset PDB4136, which gets improvement 3.6% compared with the previously best result 90.7%.

Disulfide bonds are special covalent cross links between two cysteines in a protein. This kind of bonding state plays an important role in protein folding and stabilization. For connectivity pattern prediction, it is a very difficult problem because of the fast growth of possible patterns with respect to the number of cysteines. In this thesis, we propose a new approach to address this problem. The method is based on hybrid models with SVM. Via this strategy, we can improve the prediction accuracies by selecting appropriate models. In order to evaluate the performance of our method, we apply the method by 4-fold cross-validation on SP39 dataset, which contains 446 proteins. We achieve accuracies with 70.8% and 65.9% for pair-wise and pattern-wise prediction respectively, which is better than the previous works.

Disulfide bond information can help the prediction of protein secondary structure, tertiary structure and all-atom coordinates. Most of previous works focused on either state classification or connectivity prediction with some assumption that some constraints were added to make the problem solvable in reality. In this thesis, we propose a multi-phase approach to solve the problem. Our method can export the number of bonds and achieve 90.7% accuracy in the state classification. For the connectivity prediction problem, we use the number of bonds we predict as a base to decide bond pairs. For overcoming the ratio imbalance of samples, we propose a down-sampling method to reducing processing time. Finally, we perform the weighted graph matching algorithm to obtain the bonding pattern, which achieves 63.5% accuracy. We also achieve 48% accuracy for the thorough prediction. Our method is validated by the datasets derived from SWISS-PROT and PDB. The results are better than the previous works.

No description available.

The formation of aberrant disulfide bonds is a structural consideration for the manufacturing of the extracellular domain of human CD83 (hCD83ext), a potential therapeutic protein. In certain instances, hCD83ext protein products, even when stored frozen, tend to dimerize or even multimerize through the formation of aberrant intermolecular disulfide bonds. Herein, we discovered an analytical inconsistency and applied a modified sample preparation protocol for proper structural analysis of hCD83ext products which are heterologously expressed in Escherichia coli and subsequently purified. In addition, a mutant derivative with the Cys100Ser mutation was identified as an improved version which did not form dimers or multimers. The identification of this mutant variant as a more potent therapeutic protein than other hCD83ext species demonstrated that the structural variation associated with disulfide bond formation can be a critical issue for rigorous control of the quality and bioactivity of therapeutic proteins. The application of this mutant variant for protein therapeutic is currently under exploration. As a comparative study, the hCD83ext was expressed as a glutathione-S-transferase (GST) fusion in two E. coli B strains, i.e. BL21 and Origami B having a reductive and oxidative cytoplasm. The final therapeutic products of hCD83ext produced by the two expression hosts exhibited significant differences in protein conformation and molecular properties, which presumably resulted from different disulfide patterns. The study highlights the importance of developing proper host/vector systems and biomanufacturing conditions for the production of recombinant therapeutic proteins with a consistent product quality. Cys27 in the hCD83ext was identified as a target for molecular manipulation. Two E. coli strains of BL21(DE3) and Origami B(DE3) were used as the expression host to produce the Cys27 mutants. It was observed that Cys27 was involved in the in vivo formation of intramolecular disulfide bonds when hCD83ext was expressed in Origami B(DE3). The Origami-derived protein products had a higher tendency than the BL21-derived counterparts for multimerization via the in vitro formation of intermolecular disulfide bonds. Various analyses were conducted to identify the structural differences among these mutant variants. Most importantly, molecular stability was enhanced by the Cys27 mutations since the Cys27 mutants derived from either BL21 or Origami were much less susceptible to degradation compared to wild-type hCD83ext. This study highlights the implications of aberrant disulfide bond formation on the production of therapeutic proteins. To address an inconsistent bioactivity issue that is primarily due to the aberrant formation of disulfide bonds associated with the presence of five cysteine residues, i.e. AA 27, 35, 100, 107, and 129, the molecular role that each cysteine plays upon the formation of intramolecular or intermolecular disulfide bonds was characterized, using various hCD83ext mutant variants derived by two E. coli expression hosts, i.e. BL21(DE3) and Origami B(DE3). Among the five cysteines, Cys100 and Cys129 can act as a bridging cysteine for in vitro multimerization via the formation of intermolecular disulfide bonds. The multimerization can be alleviated to some extent with less free Cys129 residues, associated with the possible formation of Cys27-Cys129 intramolecular disulfide bond. As a result, introducing the Cys27 mutation can increase the multimerization presumably via freeing more Cys129 residues. In addition, protein stability can be improved in the presence of the Cys27 mutation. The formation of the Cys27-Cys129 intramolecular disulfide bond appears to be more effective in the presence of the Cys100 mutation, resulting in the suppression of multimerization. The two conserved cysteine residues, i.e. Cys35 and Cys107, can be potentially linked to form an intramolecular disulfide bond, particularly when the protein is produced in Origami B(DE3).

The formation of aberrant disulfide bonds is a structural consideration for the manufacturing of the extracellular domain of human CD83 (hCD83ext), a potential therapeutic protein. In certain instances, hCD83ext protein products, even when stored frozen, tend to dimerize or even multimerize through the formation of aberrant intermolecular disulfide bonds. Herein, we discovered an analytical inconsistency and applied a modified sample preparation protocol for proper structural analysis of hCD83ext products which are heterologously expressed in Escherichia coli and subsequently purified. In addition, a mutant derivative with the Cys100Ser mutation was identified as an improved version which did not form dimers or multimers. The identification of this mutant variant as a more potent therapeutic protein than other hCD83ext species demonstrated that the structural variation associated with disulfide bond formation can be a critical issue for rigorous control of the quality and bioactivity of therapeutic proteins. The application of this mutant variant for protein therapeutic is currently under exploration. As a comparative study, the hCD83ext was expressed as a glutathione-S-transferase (GST) fusion in two E. coli B strains, i.e. BL21 and Origami B having a reductive and oxidative cytoplasm. The final therapeutic products of hCD83ext produced by the two expression hosts exhibited significant differences in protein conformation and molecular properties, which presumably resulted from different disulfide patterns. The study highlights the importance of developing proper host/vector systems and biomanufacturing conditions for the production of recombinant therapeutic proteins with a consistent product quality. Cys27 in the hCD83ext was identified as a target for molecular manipulation. Two E. coli strains of BL21(DE3) and Origami B(DE3) were used as the expression host to produce the Cys27 mutants. It was observed that Cys27 was involved in the in vivo formation of intramolecular disulfide bonds when hCD83ext was expressed in Origami B(DE3). The Origami-derived protein products had a higher tendency than the BL21-derived counterparts for multimerization via the in vitro formation of intermolecular disulfide bonds. Various analyses were conducted to identify the structural differences among these mutant variants. Most importantly, molecular stability was enhanced by the Cys27 mutations since the Cys27 mutants derived from either BL21 or Origami were much less susceptible to degradation compared to wild-type hCD83ext. This study highlights the implications of aberrant disulfide bond formation on the production of therapeutic proteins. To address an inconsistent bioactivity issue that is primarily due to the aberrant formation of disulfide bonds associated with the presence of five cysteine residues, i.e. AA 27, 35, 100, 107, and 129, the molecular role that each cysteine plays upon the formation of intramolecular or intermolecular disulfide bonds was characterized, using various hCD83ext mutant variants derived by two E. coli expression hosts, i.e. BL21(DE3) and Origami B(DE3). Among the five cysteines, Cys100 and Cys129 can act as a bridging cysteine for in vitro multimerization via the formation of intermolecular disulfide bonds. The multimerization can be alleviated to some extent with less free Cys129 residues, associated with the possible formation of Cys27-Cys129 intramolecular disulfide bond. As a result, introducing the Cys27 mutation can increase the multimerization presumably via freeing more Cys129 residues. In addition, protein stability can be improved in the presence of the Cys27 mutation. The formation of the Cys27-Cys129 intramolecular disulfide bond appears to be more effective in the presence of the Cys100 mutation, resulting in the suppression of multimerization. The two conserved cysteine residues, i.e. Cys35 and Cys107, can be potentially linked to form an intramolecular disulfide bond, particularly when the protein is produced in Origami B(DE3).

Abstract Protein disulfide bonds are covalent links formed between the thiol groups of cysteine residues. In many proteins, they have an important role in stabilizing the three-dimensional conformation of the polypeptide chain. Usually proteins are physiologically active and functional only when they are correctly folded. Protein folding takes place very soon after the synthesis of a new polypeptide chain. Proteins which are to be secreted from the cell fold in a specialized compartment, the endoplasmic reticulum (ER). Folding and disulfide bond formation in the ER does not happen spontaneously, there are proteins which are specialized in assisting in these processes. Protein disulfide isomerase (PDI) is a multifunctional protein, which is capable of catalysing both of disulfide bond formation and folding of a protein. In addition, it has other functions: it is an essential part of two protein complexes: collagen prolyl 4-hydroxylase (C-P4H) and microsomal triglyceride transfer protein. C-P4H is an enzyme essential in the formation of collagens, proteins found in connective tissue. The function of C-P4H is to catalyse the hydroxylation of prolines, which is essential for the structural stability of collagens. C-P4H is a tetramer, formed of two catalytic α subunits and two β subunits, which are identical to PDI. The function of PDI in C-P4H is apparently to keep it in a soluble, functionally active conformation. In mammals there are several proteins similar to PDI, together forming a PDI family of proteins. They share both structural and functional similarities. One of these proteins is ERp57. It is specialized in assisting in the folding and disulfide bond formation of glycoproteins. PDI consists of four domains, two of which contain a catalytic site for disulfide bond formation. One domain is the main site of interaction with other proteins and one domain is of unknown function. In this study, the role of these domains in the activities of PDI was investigated. The peptide-binding domain was characterized in detail. In addition, structural similarities of PDI and ERp57 were studied by formation of hybrid proteins containing domains of both and comparing the activities of these recombinant proteins to those of PDI.

Abstract The correct three dimensional structures of proteins are essential for their ability to function properly. Proteins start to fold as soon as they are synthesized in the ribosomes from activated amino acids. Many secreted, cell-surface, secretory pathway and endoplasmic reticulum (ER) lumenal proteins have in their amino acid sequence cysteine residues which form intra- and intermolecular disulfide bridges that stabilize the overall fold of the proteins and protein complexes. The formation of correct disulfide bonds is a complex process which takes place within the ER. Protein disulfide isomerase (PDI) is the key enzyme in the formation and rearrangement of correct disulfide bonds in the ER. It is an archetypal and the best studied member of the PDI family, i.e. a group of ER proteins that resemble thioredoxin (TRX), a protein reductase, in their structure. PDI has a four domain a-b-b'-a' structure the a and a' domains having the catalytic activity and amino acid sequence similarity to TRX. In addition to its function as a thiol-disulfide oxidoreductase, PDI acts as the β subunit in two protein complexes: collagen prolyl 4-hydroxylase (C-P4H) and microsomal triglyceride transfer protein (MTP). The closest homologue of PDI is the multifunctional enzyme and chaperone ERp57 that functions in concert with two lectins, calnexin (CNX) and calreticulin (CRT) specifically in the folding of proteins that have sugar moieties linked to them. ERp57 is 56% similar to PDI in its amino acid sequence and has also the four-domain architecture. Despite the high similarity in their structures ERp57 cannot substitute for PDI as the β subunit of C-P4H. The minimum requirement for the C-P4H tetramer assembly is fulfilled by domains b' and a' of PDI, while domains a and b enhance this function and can be substituted in part by those of ERp57. Until very recently the structural information of any of the PDI family members, which contains the TRX active site was limited to solution structures of human PDI domains a and b. In this research the domain boundaries of the full length ERp57 were defined and the individual domains characterized. Furthermore the solution structures of the catalytically active domains a and a' of ERp57 were studied by nuclear magnetic resonance (NMR).

Master of Science / Biochemistry / John Tomich / One of the important keys to characterize the biological function of a protein is the study of post-translational modification (PTM). Formation of disulfide bond linkages between cysteine residues within a protein is a common PTM which not only contributes to folding and stabilizing the protein structure, but also to accomplishing its native function. Therefore, the study and discovery of structural-functional relationships of expressed proteins using an isolated proteomics approach has been one of the biggest advances within the field of structural biology in recent years. In this study, rapid disulfide bond mapping of freshly obtained equine serum albumin (ESA) was performed using matrix assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS). Highly sensitive MALDI-TOF MS is commonly used for the investigation of disulfide bond linkages in the proteomics field. However, it has also been known that the presence of disulfide bond linkages absorbs the energy which is created by the cysteine-cysteine kinetic vibration, resulting in a decrease of the instrumental sensitivity. To overcome this problem, the disulfide bond mapping method was optimized by applying a combination of chemical labeling, proteolytic enzymes, and matrices. With the optimized method, we were also able to achieve high protein sequence coverage. Obtaining higher sequence coverage of a protein provides more information about a protein which helps to identify the protein by peptide mass fingerprint (PMF) technique. These analyses eventually contribute to the estimation of the possible PTM sites.

Orientadores: Beatriz Gomes Guimarães, Jose Antonio Brum / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Fisica Gleb Wataghin / Made available in DSpace on 2018-09-27T18:06:10Z (GMT). No. of bitstreams: 1 Rinaldi_FabioCupri_D.pdf: 8466921 bytes, checksum: 8a88bf7cf4ccef10efbca8ec0412db74 (MD5) Previous issue date: 2008 / Resumo: As oxidoredutases de pontes dissulfeto da família DsbA são responsáveis pela catálise da formação de pontes dissulfeto em proteínas secretadas para o periplasma, participando do processo de enovelamento de fatores de virulência de diversos organismos. É a proteína com maior potencial de oxidação atualmente caracterizada e tal propriedade é associada às interações eletrostáticas envolvendo resíduos de seu sítio ativo, que apresenta um arranjo Cys-Pro-His-Cys altamente conservado. A bactéria fitopatogênica Xylella fastidiosa possui dois genes adjacentes que codificam duas oxidoredutases pertencentes à família das DsbAs (XfDbsA e XfDbsA2). Embora a XfDbsA conserve o arranjo CPHC, a XfDbsA2 possui a substituição do resíduo histidina, descrito como essencial à atividade da enzima, por alanina (CPAC). Visando a caracterização estrutural e funcional destas proteínas, a estrutura cristalográfica da XfDsbA foi determinada a 1,9 Å de resolução e um modelo por homologia da XfDsbA2 foi construído. Além disso os potenciais de oxidação das enzimas foram determinados por medidas de fluorescência. A estrutura da XfDsbA revelou a presença de um peptídeo ligado próximo a região do sítio ativo em um dos monômeros mostrando, pela primeira vez em uma estrutura a alta resolução, o provável modo de interação da DsbA com um substrato. Os ensaios funcionais revelaram que as DsbAs de X. fastidiosa apresentam potenciais redox similares e ligeiramente superiores ao da homóloga de Escherichia coli. Embora trabalhos sobre a importância do arranjo CPHC têm associado o alto potencial redox das DsbAs à presença do resíduo histidina no sítio ativo, os resultados obtidos para a XfDsbA2 mostraram que a substituição do resíduo de histidina por alanina não afeta seu potencial redox. A análise das interações envolvendo resíduos do sítio ativo mostrou diferenças importantes entre XfDsbA, XfDsbA2 e suas homólogas de E. coli e Vibrio cholerae. Ensaios funcionais com mutantes foram realizados em busca da identificação dos resíduos que possam compensar a ausência da histidina em XfDsbA2. Os resultados obtidos fornecem novas informações sobre o mecanismo molecular dessa família de enzimas / Abstract: Disulfide oxidoreductase DsbA catalyzes disulfide-bond formation in proteins secreted to the periplasm and has been related to the folding process of virulence factors in many organisms. It is the most oxidizing of the thioredoxin-like proteins and DsbA redox power is understood in terms of the electrostatic interactions involving the active site motif CPHC. The plant pathogen Xylella fastidiosa has two chromosomal genes encoding two oxidoreductases belonging to the DsbA family and, in one of them, the canonical motif CPHC is replaced by CPAC. Aiming at the structural and functional characterization of X. fastidiosa DsbAs, the crystal structure of XfDsbA was solved at 1.9 Å resolution and the XfDsbA2 homology model was calculated. We also determined the redox potential of both enzymes by means of fluorescence experiments. The crystal structure of the XfDsbA revealed an electron density corresponding to an 8-mer peptide interacting with the hydrophobic groove on the surface of the monomer C next to the active site. This modeled peptide shows at first time in a high-resolution crystal structure the probable mode of interaction between DsbA and a substrate. Furthermore, the results presented in this work surprisingly show that, despite the absence of the active site histidine in XfDsbA2, both proteins have similar redox potentials. In addition, the structure of XfDsbA revealed critical differences in the interactions involving the active site residues. Biochemical assays with XfDsbA mutants were performed in order to investigate the residues which may be responsible for compensate for the lack of the conserved histidine in XfDsbA2. The results presented contribute to the understanding of DsbA molecular mechanism / Doutorado / Física da Matéria Condensada / Doutor em Ciências