Machine learning methods for prediction of disulphide bonding states of cysteine residues in proteins

The goal of this thesis work is to develop a computational method based on machine learning techniques for predicting disulfide-bonding states of cysteine residues in proteins, which is a sub-problem of a bigger and yet unsolved problem of protein structure prediction. Improvement in the prediction of disulfide bonding states of cysteine residues will help in putting a constraint in the three dimensional (3D) space of the respective protein structure, and thus will eventually help in the prediction of 3D structure of proteins. Results of this work will have direct implications in site-directed mutational studies of proteins, proteins engineering and the problem of protein folding. We have used a combination of Artificial Neural Network (ANN) and Hidden Markov Model (HMM), the so-called Hidden Neural Network (HNN) as a machine learning technique to develop our prediction method. By using different global and local features of proteins (specifically profiles, parity of cysteine residues, average cysteine conservation, correlated mutation, sub-cellular localization, and signal peptide) as inputs and considering Eukaryotes and Prokaryotes separately we have reached to a remarkable accuracy of 94% on cysteine basis for both Eukaryotic and Prokaryotic datasets, and an accuracy of 90% and 93% on protein basis for Eukaryotic dataset and Prokaryotic dataset respectively. These accuracies are best so far ever reached by any existing prediction methods, and thus our prediction method has outperformed all the previously developed approaches and therefore is more reliable. Most interesting part of this thesis work is the differences in the prediction performances of Eukaryotes and Prokaryotes at the basic level of input coding when ‘profile’ information was given as input to our prediction method. And one of the reasons for this we discover is the difference in the amino acid composition of the local environment of bonded and free cysteine residues in Eukaryotes and Prokaryotes. Eukaryotic bonded cysteine examples have a ‘symmetric-cysteine-rich’ environment, where as Prokaryotic bonded examples lack it.

Inhibition of AP-1 binding and transcription by gold and selenium involving conserved cysteine residues in Jun and Fos.

Gold(I) salts and selenite, which have diverse therapeutic and biological effects, are noted for their reactivity with thiols. Since the binding of Jun-Jun and Jun-Fos dimers to the AP-1 DNA binding site is regulated in vitro by a redox process involving conserved cysteine residues, we hypothesized that some of the biological actions of gold and selenium are mediated via these residues. In electrophoretic mobility-shift analyses, AP-1 DNA binding was inhibited by gold(I) thiolates and selenite, with 50% inhibition occurring at approximately 5 microM and 1 microM, respectively. Thiomalic acid had no effect in the absence of gold(I), and other metal ions inhibited at higher concentrations, in a rank order correlating with their thiol binding affinities. Cysteine-to-serine mutants demonstrated that these effects of gold(I) and selenite require Cys272 and Cys154 in the DNA-binding domains of Jun and Fos, respectively. Gold(I) thiolates and selenite did not inhibit nonspecific protein binding to the AP-1 site and were at least an order of magnitude less potent as inhibitors of sequence-specific binding to the AP-2, TFIID, or NF1 sites compared with the AP-1 site. In addition, 10 microM gold(I) or 10 microM selenite inhibited expression of an AP-1-dependent reporter gene, but not an AP-2-dependent reporter gene. These data suggest a mechanism regulating transcription factor activity by inorganic ions which may contribute to the known antiarthritic action of gold and cancer chemoprevention by selenium.

Proteasomal degradation of N-acetyltransferase 1 is prevented by acetylation of the active site cysteine - A mechanism for the slow acetylator phenotype and substrate-dependent down-regulation

Many drugs and chemicals found in the environment are either detoxified by N-acetyltransferase 1 (NAT1, EC 2.3.1.5) and eliminated from the body or bioactivated to metabolites that have the potential to cause toxicity and/or cancer. NAT1 activity in the body is regulated by genetic polymorphisms as well as environmental factors such as substrate-dependent down-regulation and oxidative stress. Here we report the molecular mechanism for the low protein expression from mutant NAT1 alleles that gives rise to the slow acetylator phenotype and show that a similar process accounts for enzyme down-regulation by NAT1 substrates. NAT1 allozymes NAT1 14, NAT1 15, NAT1 17, and NAT1 22 are devoid of enzyme activity and have short intracellular half-lives (similar to4 h) compared with wild-type NAT1 4 and the active allozyme NAT1 24. The inactive allozymes are unable to be acetylated by cofactor, resulting in ubiquitination and rapid degradation by the 26 S proteasome. This was confirmed by site-directed mutagenesis of the active site cysteine 68. The NAT1 substrate p-aminobenzoic acid induced ubiquitination of the usually stable NAT1 4, leading to its rapid degradation. From this study, we conclude that NAT1 exists in the cell in either a stable acetylated state or an unstable non-acetylated state and that mutations in the NAT1 gene that prevent protein acetylation produce a slow acetylator phenotype.

N-Glycosylation and conserved cysteine residues in RAMP3 play a critical role for the functional expression of CRLR/RAMP3 adrenomedullin receptor.

The calcitonin receptor-like receptor (CRLR) and receptor activity modifying protein-3 (RAMP3) can assemble into a CRLR/RAMP3 heterodimeric receptor that exhibits the characteristics of a high affinity adrenomedullin receptor. RAMP3 participates in adrenomedullin (AM) binding via its extracellular N-terminus characterized by the presence of six highly conserved cysteine residues and four N-glycosylation consensus sites. Here, we assessed the usage of these conserved residues in cotranslational modifications of RAMP3 and addressed their role in functional expression of the CRLR/RAMP3 receptor. Using a Xenopus oocyte expression system, we show that (i) RAMP3 is assembled with CRLR as a multiple N-glycosylated species in which two, three, or four consensus sites are used; (ii) elimination of all N-glycans in RAMP3 results in a significant inhibition of receptor [(125)I]AM binding and an increase in the EC(50) value for AM; (iii) several lines of indirect evidence indicate that each of the six cysteines is involved in disulfide bond formation; (iv) when all cysteines are mutated to serines, RAMP3 is N-glycosylated at all four consensus sites, suggesting that disulfide bond formation inhibits N-gylcosylation; and (v) elimination of all cysteines abolishes adrenomedullin binding and leads to a complete loss of receptor function. Our data demonstrate that cotranslational modifications of RAMP3 play a critical role in the function of the CRLR/RAMP3 adrenomedullin receptor.

Carbon monoxide inhibits L-type Ca2+ channels via redox modulation of key cysteine residues by mitochondrial reactive oxygen species

Conditions of stress, such as myocardial infarction, stimulate up-regulation of heme oxygenase (HO-1) to provide cardioprotection. Here, we show that CO, a product of heme catabolism by HO-1, directly inhibits native rat cardiomyocyte L-type Ca2+ currents and the recombinant alpha1C subunit of the human cardiac L-type Ca2+ channel. CO (applied via a recognized CO donor molecule or as the dissolved gas) caused reversible, voltage-independent channel inhibition, which was dependent on the presence of a spliced insert in the cytoplasmic C-terminal region of the channel. Sequential molecular dissection and point mutagenesis identified three key cysteine residues within the proximal 31 amino acids of the splice insert required for CO sensitivity. CO-mediated inhibition was independent of nitric oxide and protein kinase G but was prevented by antioxidants and the reducing agent, dithiothreitol. Inhibition of NADPH oxidase and xanthine oxidase did not affect the inhibitory actions of CO. Instead, inhibitors of complex III (but not complex I) of the mitochondrial electron transport chain and a mitochondrially targeted antioxidant (Mito Q) fully prevented the effects of CO. Our data indicate that the cardioprotective effects of HO-1 activity may be attributable to an inhibitory action of CO on cardiac L-type Ca2+ channels. Inhibition arises from the ability of CO to promote generation of reactive oxygen species from complex III of mitochondria. This in turn leads to redox modulation of any or all of three critical cysteine residues in the channel's cytoplasmic C-terminal tail, resulting in channel inhibition.

Horseradish peroxidase compound I as a tool to investigate reactive protein-cysteine residues: from quantification to kinetics

Proteins containing reactive cysteine residues (protein-Cys) are receiving increased attention as mediators of hydrogen peroxide signaling. These proteins are mainly identified by mining the thiol proteomes of oxidized protein-Cys in cells and tissues. However, it is difficult to determine if oxidation occurs through a direct reaction with hydrogen peroxide or by thiol-disulfide exchange reactions. Kinetic studies with purified proteins provide invaluable information about the reactivity of protein-Cys residues with hydrogen peroxide. Previously, we showed that the characteristic UV-Vis spectrum of horseradish peroxidase compound I, produced from the oxidation of horseradish peroxidase by hydrogen peroxide, is a simple, reliable, and useful tool to determine the second-order rate constant of the reaction of reactive protein-Cys with hydrogen peroxide and peroxynitrite. Here, the method is fully described and extended to quantify reactive protein-Cys residues and micromolar concentrations of hydrogen peroxide. Members of the peroxiredoxin family were selected for the demonstration and validation of this methodology. In particular, we determined the pK(a) of the peroxidatic thiol of rPrx6 (5.2) and the second-order rate constant of its reactions with hydrogen peroxide ((3.4 +/- 0.2) x 10(7) M(-1) s(-1)) and peroxynitrite ((3.7 +/- 0.4) x 10(5) M(-1) s(-1)) at pH 7.4 and 25 degrees C. (C) 2011 Elsevier Inc. All rights reserved.

Recycling of the high valence states of heme proteins by cysteine residues of thimet-oligopeptidase

The peptidolytic enzyme THIMET-oligopeptidase (TOP) is able to act as a reducing agent in the peroxidase cycle of myoglobin (Mb) and horseradish peroxidase (HRP). The TOP-promoted recycling of the high valence states of the peroxidases to the respective resting form was accompanied by a significant decrease in the thiol content of the peptidolytic enzyme. EPR (electron paramagnetic resonance) analysis using DBNBS spin trapping revealed that TOP also prevented the formation of tryptophanyl radical in Mb challenged by H2O2. The oxidation of TOP thiol groups by peroxidases did not promote the inactivating oligomerization observed in the oxidation promoted by the enzyme aging. These findings are discussed towards a possible occurrence of these reactions in cells.

m5C RNA and m5C DNA methyl transferases use different cysteine residues as catalysts

A family of RNA m5C methyl transferases (MTases) containing over 55 members in eight subfamilies has been identified recently by an iterative search of the genomic sequence databases by using the known 16S rRNA m5C 967 MTase, Fmu, as an initial probe. The RNA m5C MTase family contained sequence motifs that were highly homologous to motifs in the DNA m5C MTases, including the ProCys sequence that contains the essential Cys catalyst of the functionally similar DNA-modifying enzymes; it was reasonable to assign the Cys nucleophile to be that in the conserved ProCys. The family also contained an additional conserved Cys residue that aligns with the nucleophilic catalyst in m5U54 tRNA MTase. Surprisingly, the mutant of the putative Cys catalyst in the ProCys sequence was active and formed a covalent complex with 5-fluorocytosine-containing RNA, whereas the mutant at the other conserved Cys was inactive and unable to form the complex. Thus, notwithstanding the highly homologous sequences and similar functions, the RNA m5C MTase uses a different Cys as a catalytic nucleophile than the DNA m5C MTases. The catalytic Cys seems to be determined, not by the target base that is modified, but by whether the substrate is DNA or RNA. The function of the conserved ProCys sequence in the RNA m5C MTases remains unknown.

Mutations affecting conserved cysteine residues within the extracellular domain of Neu promote receptor dimerization and activation.

Overexpression of the Neu/ErbB-2 receptor tyrosine kinase has been implicated in the genesis of human breast cancer. Indeed, expression of either activated or wild-type neu in the mammary epithelium of transgenic mice results in the induction of mammary tumors. Previously, we have shown that many of the mammary tumors arising in transgenic mice expressing wild-type neu occur through somatic activating mutations within the neu transgene itself. Here we demonstrate that these mutations promote dimerization of the Neu receptor through the formation of disulfide bonds, resulting in its constitutive activation. To explore the role of conserved cysteine residues within the region deleted in these altered Neu proteins, we examined the transforming potential of a series of Neu receptors in which the individual cysteine residues were mutated. These analyses indicated that mutation of certain cysteine residues resulted in the oncogenic activation of Neu. The increased transforming activity displayed by the altered receptors correlated with constitutive dimerization that occurred in a disulfide bond-dependent manner. We further demonstrate that addition of 2-mercaptoethanol to the culture medium interfered with the specific transforming activity of the mutant Neu receptors. These observations suggest that oncogenic activation of Neu results from constitutive disulfide bond-dependent dimerization.

Crystal structures of reduced and oxidized DsbA: investigation of domain motion and thiolate stabilization

Background: The redox proteins that incorporate a thioredoxin fold have diverse properties and functions. The bacterial protein-folding factor DsbA is the most oxidizing of the thioredoxin family. DsbA catalyzes disulfide-bond formation during the folding of secreted proteins, The extremely oxidizing nature of DsbA has been proposed to result from either domain motion or stabilizing active-site interactions in the reduced form. In the domain motion model, hinge bending between the two domains of DsbA occurs as a result of redox-related conformational changes. Results: We have determined the crystal structures of reduced and oxidized DsbA in the same crystal form and at the same pH (5.6). The crystal structure of a lower pH form of oxidized DsbA has also been determined (pH 5.0). These new crystal structures of DsbA, and the previously determined structure of oxidized DsbA at pH 6.5, provide the foundation for analysis of structural changes that occur upon reduction of the active-site disulfide bond. Conclusions: The structures of reduced and oxidized DsbA reveal that hinge bending motions do occur between the two domains. These motions are independent of redox state, however, and therefore do not contribute to the energetic differences between the two redox states, instead, the observed domain motion is proposed to be a consequence of substrate binding. Furthermore, DsbA's highly oxidizing nature is a result of hydrogen bond, electrostatic and helix-dipole interactions that favour the thiolate over the disulfide at the active site.

A common inhibitory binding site for zinc and odorants at the voltage-gated K+ channel of rat olfactory receptor neurons

This study compared the effects of zinc and odorants on the voltage-gated K+ channel of rat olfactory neurons. Zinc reduced current magnitude, depolarized the voltage activation curve and slowed activation kinetics without affecting inactivation or deactivation kinetics. Zinc inhibition was potentiated by the NO compound, S-nitroso-cysteine. The pH- and diethylpyrocarbonate-dependence of zinc inhibition suggested that zinc acted by binding to histidine residues. Cysteine residues were eliminated as contributing to the zinc-binding site. The odorants, acetophenone and amyl acetate, also reduced current magnitude, depolarized the voltage activation curve and selectively slowed activation kinetics. Furthermore, the diethylpyrocarbonate- and pH-dependence of odorant inhibition implied that the odorants also bind to histidine residues. Zinc inhibitory potency was dramatically diminished in the presence of odorants, implying competition for a common binding site. These observations indicate that the odorants and zinc share a common inhibitory binding site on the external surface of the voltage-gated K+ channel.

Mutational analysis of cysteine-rich domains of the epithelium sodium channel (ENaC). Identification of cysteines essential for channel expression at the cell surface.

One of the characteristic features of the structure of the epithelial sodium channel family (ENaC) is the presence of two highly conserved cysteine-rich domains (CRD1 and CRD2) in the large extracellular loops of the proteins. We have studied the role of CRDs in the functional expression of rat alphabetagamma ENaC subunits by systematically mutating cysteine residues (singly or in combinations) into either serine or alanine. In the Xenopus oocyte expression system, mutations of two cysteines in CRD1 of alpha, beta, or gamma ENaC subunits led to a temperature-dependent inactivation of the channel. In CRD1, one of the cysteines of the rat alphaENaC subunit (Cys158) is homologous to Cys133 of the corresponding human subunit causing, when mutated to tyrosine (C133Y), pseudohypoaldosteronism type 1, a severe salt-loosing syndrome in neonates. In CRD2, mutation of two cysteines in alpha and beta but not in the gamma subunit also produced a temperature-dependent inactivation of the channel. The main features of the mutant cysteine channels are: (i) a decrease in cell surface expression of channel molecules that parallels the decrease in channel activity and (ii) a normal assembly or rate of degradation as assessed by nondenaturing co-immunoprecipitation of [35S]methionine-labeled channel protein. These data indicate that the two cysteines in CRD1 and CRD2 are not a prerequisite for subunit assembly and/or intrinsic channel activity. We propose that they play an essential role in the efficient transport of assembled channels to the plasma membrane.

Methionine residues as endogenous antioxidants in proteins

Cysteine and methionine are the two sulfur-containing residues normally found in proteins. Cysteine residues function in the catalytic cycle of many enzymes, and they can form disulfide bonds that contribute to protein structure. In contrast, the specific functions of methionine residues are not known. We propose that methionine residues constitute an important antioxidant defense mechanism. A variety of oxidants react readily with methionine to form methionine sulfoxide, and surface exposed methionine residues create an extremely high concentration of reactant, available as an efficient oxidant scavenger. Reduction back to methionine by methionine sulfoxide reductases would allow the antioxidant system to function catalytically. The effect of hydrogen peroxide exposure upon glutamine synthetase from Escherichia coli was studied as an in vitro model system. Eight of the 16 methionine residues could be oxidized with little effect on catalytic activity of the enzyme. The oxidizable methionine residues were found to be relatively surface exposed, whereas the intact residues were generally buried within the core of the protein. Furthermore, the susceptible residues were physically arranged in an array that guarded the entrance to the active site.

Distance determination in proteins using designed metal ion binding sites and site-directed spin labeling: application to the lactose permease of Escherichia coli.

As shown in the accompanying paper, the magnetic dipolar interaction between site-directed metal-nitroxide pairs can be exploited to measure distances in T4 lysozyme, a protein of known structure. To evaluate this potentially powerful method for general use, particularly with membrane proteins that are difficult to crystallize, both a paramagnetic metal ion binding site and a nitroxide side chain were introduced at selected positions in the lactose permease of Escherichia coli, a paradigm for polytopic membrane proteins. Thus, three individual cysteine residues were introduced into putative helix IV of a lactose permease mutant devoid of native cysteine residues containing a high-affinity divalent metal ion binding site in the form of six contiguous histidine residues in the periplasmic loop between helices III and IV. In addition, the construct contained a biotin acceptor domain in the middle cytoplasmic loop to facilitate purification. After purification and spin labeling, electron paramagnetic resonance spectra were obtained with the purified proteins in the absence and presence of Cu(II). The results demonstrate that positions 103, 111, and 121 are 8, 14, and > 23 A from the metal binding site. These data are consistent with an alpha-helical conformation of transmembrane domain IV of the permease. Application of the technique to determine helix packing in lactose permease is discussed.

Resolution and characterisation of multiple isoforms of bovine kappa-casein by 2-DE following a reversible cysteine-tagging enrichment strategy

Visualisation of multiple isoforms of kappa-casein on 2-D gels is restricted by the abundant alpha- and beta-caseins that not only limit gel loading but also migrate to similar regions as the more acidic kappa-casein isoforms. To overcome this problem, we took advantage of the absence of cysteine residues in alpha(S1)- and beta-casein by devising an affinity enrichment procedure based on reversible biotinylation of cysteine residues. Affinity capture of cysteine-containing proteins on avidin allowed the removal of the vast majority of alpha(S1)- and beta-casein, and on subsequent 2-D gel analysis 16 gel spots were identified as kappa-casein by PMF. Further analysis of the C-terminal tryptic peptide along with structural predictions based on mobility on the 2-D gel allowed us to assign identities to each spot in terms of genetic variant (A or B), phosphorylation status (1, 2 or 3) and glycosylation status (from 0 to 6). Eight isoforms of the A and B variants with the same PTMs were observed. When the casein fraction of milk from a single cow, homozygous for the B variant of kappa-casein, was used as the starting material, 17 isoforms from 13 gel spots were characterised. Analysis of isoforms of low abundance proved challenging due to the low amount of material that could be extracted from the gels as well as the lability of the PTMs during MS analysis. However, we were able to identify a previously unrecognised site, T-166, that could be phosphorylated or glycosylated. Despite many decades of analysis of milk proteins, the reasons for this high level of heterogeneity are still not clear.