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.

DdLIM Is a Cytoskeleton-associated Protein Involved in the Protrusion of Lamellipodia in Dictyostelium

DdLim, a multi-domain member of the cysteine-rich family of LIM domain proteins, was isolated from Dictyostelium cells where it localizes in lamellipodia and at sites of membrane ruffling. The transcription and expression of DdLim are developmentally regulated, and the timing of its increased association with the actin cytoskeleton coincides with the acquisition in starved cells of a motile, chemotactic behavior. Vegetative cells that overexpress DdLim contain large lamella and exhibit ruffling at the cortex. The high frequency of large, multinucleated mutant cells found in suspension culture suggests that excess DdLim interferes with cytokinesis. DdLim was also identified as a protein in a Dictyostelium cell lysate that associated indirectly, but in a guanosine triphosphate-dependent manner, with a GST-rac1 fusion protein. The data presented suggest that DdLim acts as an adapter protein at the cytoskeleton-membrane interface where it is involved in a receptor-mediated rac1-signaling pathway that leads to actin polymerization in lamellipodia and ultimately cell motility.

Role of metallothionein in nitric oxide signaling as revealed by a green fluorescent fusion protein

Although the function of metallothionein (MT), a 6- to 7-kDa cysteine-rich metal binding protein, remains unclear, it has been suggested from in vitro studies that MT is an important component of intracellular redox signaling, including being a target for nitric oxide (NO). To directly study the interaction between MT and NO in live cells, we generated a fusion protein consisting of MT sandwiched between two mutant green fluorescent proteins (GFPs). In vitro studies with this chimera (FRET-MT) demonstrate that fluorescent resonance energy transfer (FRET) can be used to follow conformational changes indicative of metal release from MT. Imaging experiments with live endothelial cells show that agents that increase cytoplasmic Ca2+ act via endogenously generated NO to rapidly and persistently release metal from MT. A role for this interaction in intact tissue is supported by the finding that the myogenic reflex of mesenteric arteries is absent in MT knockout mice (MT−/−) unless endogenous NO synthesis is blocked. These results are the first application of intramolecular green fluorescent protein (GFP)-based FRET in a native protein and demonstrate the utility of FRET-MT as an intracellular surrogate indicator of NO production. In addition, an important role of metal thiolate clusters of MT in NO signaling in vascular tissue is revealed.

Cloning of rat MEK kinase 1 cDNA reveals an endogenous membrane-associated 195-kDa protein with a large regulatory domain.

The coding sequence of rat MEK kinase 1 (MEKK1) has been determined from multiple, independent cDNA clones. The cDNA is full-length based on the presence of stop codons in all three reading frames of the 5' untranslated region. Probes from the 5' and the 3' coding sequences both hybridize to a 7-kb mRNA. The open reading frame is 4.5 kb and predicts a protein with molecular mass of 161,225 Da, which is twice the size of the previously published MEKK1 sequence and reveals 801 amino acids of novel coding sequence. The novel sequence contains two putative pH domains, two proline-rich regions, and a cysteine-rich region. Antisera to peptides derived from this new sequence recognize an endogenous protein in human and rodent cells of 195 kDa, consistent with the size of the expressed rat MEKK1 clone. Endogenous and recombinant rat MEKK1 are enriched in membranes; little of either is found in soluble fractions. Expression of recombinant rat MEKK1 leads to activation of three mitogen-activated protein kinase modules in the order c-Jun N-terminal kinase/stress-activated protein kinase > p38 mitogen-activated protein kinase = extracellular signal-regulated kinase 2.

Vascular endothelial growth factor-related protein: a ligand and specific activator of the tyrosine kinase receptor Flt4.

The tyrosine kinases Flt4, Flt1, and Flk1 (or KDR) constitute a family of endothelial cell-specific receptors with seven immunoglobulin-like domains and a split kinase domain. Flt1 and Flk1 have been shown to play key roles in vascular development; these two receptors bind and are activated by vascular endothelial growth factor (VEGF). No ligand has been identified for Flt4, whose expression becomes restricted during development to the lymphatic endothelium. We have identified cDNA clones from a human glioma cell line that encode a secreted protein with 32% amino acid identity to VEGF. This protein, designated VEGF-related protein (VRP), specifically binds to the extracellular domain of Flt4, stimulates the tyrosine phosphorylation of Flt4 expressed in mammalian cells, and promotes the mitogenesis of human lung endothelial cells. VRP fails to bind appreciably to the extracellular domain of Flt1 or Flk1. The protein contains a C-terminal, cysteine-rich region of about 180 amino acids that is not found in VEGF. A 2.4-kb VRP mRNA is found in several human tissues including adult heart, placenta, ovary, and small intestine and in fetal lung and kidney.

Isolation of cDNA clones encoding RICH: a protein induced during goldfish optic nerve regeneration with homology to mammalian 2',3'-cyclic-nucleotide 3'-phosphodiesterases.

Using data derived from peptide sequencing of p68/70, a protein doublet induced during optic nerve regeneration in goldfish, we have isolated cDNAs that encode RICH (regeneration-induced CNPase homolog) from a goldfish regenerating retina cDNA library. The predicted RICH protein comprises 411 amino acids, possesses a pI of 4.48, and shows significant homology to the mammalian myelin marker enzyme 2',3'-cyclic-nucleotide 3'-phosphodiesterase (CNPase; EC 3.1.4.37). The mRNA encoding RICH was demonstrated, by both Northern blot analysis and RNase protection assays, to be induced as much as 8-fold in regenerating goldfish retinas at 20 days after nerve crush. Analysis of total RNA samples from various tissues showed a broad distribution of RICH mRNA, with the highest levels observed in gravid ovary. The data obtained strongly suggest that RICH is identical or very similar to p68/70. The molecular cloning of RICH provides the means for a more detailed analysis of its function in nerve regeneration. Additionally, the homology of RICH and CNPase suggests that further investigation may provide additional insight into the role of these proteins in the nervous system.

Molecular characterisation of a predominant antigenic region of Giardia lamblia variant surface protein H7.

During infection, the intestinal protozoan parasite Giardia lamblia undergoes continuous antigenic variation which is determined by diversification of the parasite's major surface antigen, named VSP (variant surface protein). One member from this protein family, VSP H7, is expressed by G. lamblia clone GS/M-83-H7. In the present study, we characterised a highly antigenic portion of VSP H7 which is positioned inside a 130 amino acid C-terminal region of the protein. This region overlaps with a cysteine-rich motif that is rather conserved within the VSP family. Detailed molecular dissection of the antigenic portion monitored a 12 amino acid peptidyl structure which constitutes a non-conformational epitope of VSP H7. In the murine host, this epitope is recognised relatively early (before day 10 p.i.) during infection and stimulates a strong intestinal immunoglobulin A response. At late infective stages (after day 10 p.i.) this immune reaction is progressively complemented by reactions against 'late' antigenic epitopes which are also located inside the 130 amino acid antigenic portion but in closer proximity to the C-terminal end of VSP H7 than the 12 amino acid epitope. Both the high antigenicity and the conserved character suggest that the 12 amino acid epitope is a key factor within the immunological interplay between G. lamblia and the experimental murine host.

The Role of the Stroma and CYR61 in Chemoresistance in Pancreatic Cancer

Pancreatic ductal adenocarcinoma (PDAC) is a lethal cancer in part due to inherent resistance to chemotherapy, including the first-line drug gemcitabine. Gemcitabine is a nucleoside pyrimidine analog that has long been the backbone of chemotherapy for PDAC, both as a single agent, and more recently, in combination with nab-paclitaxel. Since gemcitabine is hydrophilic, it must be transported through the hydrophobic cell membrane by transmembrane nucleoside transporters. Human equilibrative nucleoside transporter-1 (hENT1) and human concentrative nucleoside transporter-3 (hCNT3) both have important roles in the cellular uptake of the nucleoside analog gemcitabine. While low expression of hENT1 and hCNT3 has been linked to gemcitabine resistance clinically, mechanisms regulating their expression in the PDAC tumor microenvironment are largely unknown. We identified that the matricellular protein Cysteine-Rich Angiogenic Inducer 61 (CYR61) negatively regulates expression of hENT1 and hCNT3. CRISPR/Cas9-mediated knockout of CYR61 significantly increased expression of hENT1 and hCNT3 and cellular uptake of gemcitabine. CRSIPR-mediated knockout of CYR61 sensitized PDAC cells to gemcitabine-induced apoptosis. Conversely, adenovirus-mediated overexpression of CYR61 decreased hENT1 expression and reduced gemcitabine-induced apoptosis. We demonstrate that CYR61 is expressed primarily by stromal pancreatic stellate cells (PSCs) within the PDAC tumor microenvironment, with Transforming Growth Factor- β (TGF-β) inducing the expression of CYR61 in PSCs through canonical TGF-β-ALK5-Smad signaling. Activation of TGF-β signaling or expression of CYR61 in PSCs promotes resistance to gemcitabine in an in vitro co-culture assay with PDAC cells. Our results identify CYR61 as a TGF-β induced stromal-derived factor that regulates gemcitabine sensitivity in PDAC and suggest that targeting CYR61 may improve chemotherapy response in PDAC patients.

Biochemical studies of postsynaptic density signaling proteins with a focus on synaptic GTPase activating protein and PDZ domains

Memory storage in the brain involves adjustment of the strength of existing synapses and formation of new neural networks. A key process underlying memory formation is synaptic plasticity, the ability of excitatory synapses to strengthen or weaken their connections in response to patterns of activity between their connected neurons. Synaptic plasticity is governed by the precise pattern of Ca²⁺ influx through postsynaptic N-methyl-D-aspartate-type glutamate receptors (NMDARs), which can lead to the activation of the small GTPases Ras and Rap. Differential activation of Ras and Rap acts to modulate synaptic strength by promoting the insertion or removal of 2-amino-3-(3-hydroxy-5-methyl-isoxazol-4-yl)propanoic acid receptors (AMPARs) from the synapse. Synaptic GTPase activating protein (synGAP) regulates AMPAR levels by catalyzing the inactivation of GTP-bound (active) Ras or Rap. synGAP is positioned in close proximity to the cytoplasmic tail regions of the NMDAR through its association with the PDZ domains of PSD-95. SynGAP’s activity is regulated by the prominent postsynaptic protein kinase, Ca²⁺/calmodulin-dependent protein kinase II (CaMKII) and cyclin-dependent kinase 5 (CDK5), a known binding partner of CaMKII. Modulation of synGAP’s activity by phosphorylation may alter the ratio of active Ras to Rap in spines, thus pushing the spine towards the insertion or removal of AMPARs, subsequently strengthening or weakening the synapse. To date, all biochemical studies of the regulation of synGAP activity by protein kinases have utilized impure preparations of membrane bound synGAP. Here we have clarified the effects of phosphorylation of synGAP on its Ras and Rap GAP activities by preparing and utilizing purified, soluble recombinant synGAP, Ras, Rap, CaMKII, CDK5, PLK2, and CaM. Using mass spectrometry, we have confirmed the presence of previously identified CaMKII and CDK5 sites in synGAP, and have identified novel sites of phosphorylation by CaMKII, CDK5, and PLK2. We have shown that the net effect of phosphorylation of synGAP by CaMKII, CDK5, and PLK2 is an increase in its GAP activity toward HRas and Rap1. In contrast, there is no effect on its GAP activity toward Rap2. Additionally, by assaying the GAP activity of phosphomimetic synGAP mutants, we have been able to hypothesize the effects of CDK5 phosphorylation at specific sites in synGAP. In the course of this work, we also found, unexpectedly, that synGAP is itself a Ca²⁺/CaM binding protein. While Ca²⁺/CaM binding does not directly affect synGAP activity, it causes a conformational change in synGAP that increases the rate of its phosphorylation and exposes additional phosphorylation sites that are inaccessible in the absence of Ca²⁺/CaM.

The postsynaptic density (PSD) is an electron-dense region in excitatory postsynaptic neurons that contains a high concentration of glutamate receptors, cytoskeletal proteins, and associated signaling enzymes. Within the PSD, three major classes of scaffolding molecules function to organize signaling enzymes and glutamate receptors. PDZ domains present in the Shank and PSD-95 scaffolds families serve to physically link AMPARs and NMDARs to signaling molecules in the PSD. Because of the specificity and high affinity of PDZ domains for their ligands, I reasoned that these interacting pairs could provide the core components of an affinity chromatography system, including affinity resins, affinity tags, and elution agents. I show that affinity columns containing the PDZ domains of PSD-95 can be used to purify active PDZ domain-binding proteins to very high purity in a single step. Five heterologously expressed neuronal proteins containing endogenous PDZ domain ligands (NMDAR GluN2B subunit Tail, synGAP, neuronal nitric oxide synthase PDZ domain, cysteine rich interactor of PDZ three and cypin) were purified using PDZ domain resin, with synthetic peptides having the sequences of cognate PDZ domain ligands used as elution agents. I also show that conjugation of PDZ domain-related affinity tags to Proteins Of Interest (POIs) that do not contain endogenous PDZ domains or ligands does not alter protein activity and enables purification of the POIs on PDZ domain-related affinity resins.

Bioactivity of nitrolinoleate: effects on adhesion molecules and CD40-CD40L system

The vascular effects of nitrolinoleate (LNO(2)), an endogenous product of linoleic acid (LA) nitration by nitric oxide-derived species and a potential nitrosating agent, were investigated on rat endothelial-leukocyte interactions. Confocal microscopy analysis demonstrated that LNO(2) was capable to deliver free radical nitric oxide ((center dot)NO) into cells, 5 min after its administration to cultured cells, with a peak of liberation at 30 min. THP-1 monocytes incubated with LNO(2) for 5 min presented nitrosation of CD40, leading to its inactivation. Other anti-inflammatory actions of LNO(2) were observed in vivo by intravital microscopy assays. LNO(2) decreased the number of adhered leukocytes in postcapillary venules of the mesentery network. In addition to this, LNO(2) reduced mRNA and protein expression of 2-integrin in circulating leukocytes, as well as VCAM-1 in endothelial cells isolated from postcapillary venules, confirming its antiadhesive effects on both cell types. Moreover, 2-(4-carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide, a nitric oxide scavenger, partially abolished the inhibitory action of LNO(2) on leukocyte-endothelium interaction, suggesting that the antiadhesion effects of LNO(2) involve a dual role in leukocyte adhesion, acting as a nitric oxide donor as well as through nitric oxide-independent mechanisms. In conclusion, LNO(2) inhibited adhesion molecules expression and promoted (center dot)NO inactivation of the CD40-CD40L system, both important processes of the inflammatory response. (C) 2010 Elsevier Inc. All rights reserved.

Snake venomics and antivenomics of Crotalus durissus subspecies from Brazil: Assessment of geographic variation and its implication on snakebite management

We report the comparative proteomic and antivenomic characterization of the venoms of subspecies cascavella and collilineatus of the Brazilian tropical rattlesnake Crotalus durissus. The venom proteomes of C. d. collilineatus and C. d. cascavella comprise proteins in the range of 4-115 kDa belonging to 9 and 8 toxin families, respectively. Collilineatus and cascavella venoms contain 20-25 main toxins belonging to the following protein families: disintegrin, PLA(2), serine proteinase, cysteine-rich secretory protein (CRISP), vascular endothelial growth factor-like (VEGF), L-amino acid oxidase, C-type lectin-like, and snake venom metalloproteinase (SVMP). As judged by reverse-phase HPLC and mass spectrometry, cascavella and collilineatus share about 90% of their venom proteome. However, the relative occurrence of the toxin families departs among the two C. durissus subspecies venoms. The most notable difference is the presence of the myotoxin crotamine in some C. d. collilineatus specimens (averaging 20.8% of the total proteins of pooled venom), which is absent in the venom of C. d. cascavella. On the other hand, the neurotoxic PLA2 crotoxin represents the most abundant protein in both C. durissus venoms, comprising 67.4% of the toxin proteome in C. d. collilineatus and 72.5% in C. d. cascavella. Myotoxic PLA(2)s are also present in the two venoms albeit in different relative concentrations (18.1% in C. d. cascavella vs. 4.6% in C. d. collilineatus). The venom composition accounts for the clinical manifestations caused by C. durissus envenomations: systemic neurotoxicity and myalgic symptoms and coagulation disturbances, frequently accompanied by myoglobinuria and acute renal failure. The overall compositions of C. d. subspecies cascavella and collilineatus venoms closely resemble that of C. d. terrificus, supporting the view that these taxa can be considered geographical variations of the same species. Pooled venom from adult C.d. cascavella and neonate C.d. terrificus lack crotamine, whereas this skeletal muscle cell membrane depolarizing inducing myotoxin accounts for similar to 20% of the total toxins of venom pooled from C.d. collilineatus and C.d. terrificus from Southern Brazil. The possible relevance of the observed venom variability among the tropical rattlesnake subspecies was assessed by antivenomics using anti-crotalic antivenoms produced at Instituto Butantan and Instituto Vital Brazil. The results revealed that both antivenoms exhibit impaired immunoreactivity towards crotamine and display restricted (similar to 60%) recognition of PLA(2) molecules (crotoxin and D49-myotoxins) from C. d. cascavella and C. d. terrificus venoms. This poor reactivity of the antivenoms may be due to a combination of factors: on the one hand, an inappropriate choice of the mixture of venoms for immunization and, on the other hand, the documented low immunogenicity of PLA(2) molecules. C. durissus causes most of the lethal snakebite accidents in Brazil. The implication of the geographic variation of venom composition for the treatment of bites by different C. durissus subspecies populations is discussed. (C) 2010 Elsevier B.V. All rights reserved.

Crystal structure at 1.1 angstrom resolution of alpha-conotoxin PnIB: Comparison with alpha-conotoxins PnIA and GI

Conotoxins are small, cysteine-rich peptides isolated from the venom of Conus spp. of predatory marine snails, which selectively target specific receptors and ion channels critical to the functioning of the neuromuscular system. alpha-Conotoxins PnIA and PnIB are both 16-residue peptides (differing in sequence at only two positions) isolated from the molluscivorous snail Conus pennaceus. In contrast to the muscle-selective alpha-conotoxin GI from Conus geographus, PnIA and PnIB block the neuronal nicotinic acetylcholine receptor (nAChR). Here, we describe the crystal structure of PnIB, solved at a resolution of 1.1 Angstrom and phased using the Shake-and-Bake direct methods program. PnIB crystals are orthorhombic and belong to the space group P2(1)2(1)2(1) with the following unit cell dimensions: a = 14.6 Angstrom, b = 26.1 Angstrom, and c = 29.2 Angstrom. The final refined structure of alpha-conotoxin PnIB includes all 16 residues plus 23 solvent molecules and has an overall R-factor of 14.7% (R-free of 15.9%). The crystal structures of the alpha-conotoxins PnIB and PnIA are solved from different crystal forms, with different solvent contents. Comparison of the structures reveals them to be very similar, showing that the unique backbone and disulfide architecture is not strongly influenced by crystal lattice constraints or solvent interactions. This finding supports the notion that this structural scaffold is a rigid support for the presentation of important functional groups. The structures of PnIB and PnIA differ in their shape and surface charge distribution from that of GI.

Novel gene containing multiple epidermal growth factor-like motifs transiently expressed in the papillae of the ascidian tadpole larvae

We have investigated molecular mechanisms of the embryonic development of an ascidian, a primitive chordate which shares features of both invertebrates and vertebrates, with a view to identifying genes involved in development and metamorphosis, We isolated 12 partial cDNA sequences which were expressed in a stage-specific manner using differential display, We report here the isolation of a full-length cDNA sequence for one of these genes which was specifically expressed during the tailbud and larval stages of ascidian development, This cDNA, 1213 bp in length, is predicted to encode a protein of 337 amino acids containing four epidermal growth factor (EGF)-like repeats and three novel cysteine-rich repeats, Characterization of its spatial expression pattern by in situ hybridisation in late tailbud and larval embryos demonstrated strong expression localised throughout the papillae and anteriormost trunk and weaker expression in the epidermis of the remainder of the embryo, As recent evidence indicates that the signal for metamorphosis originates in the anterior trunk region, these results suggest that this gene may have a role in signalling the initiation of metamorphosis. (C) 1997 Wiley-Liss, Inc.

Antidepressant-like effects of N-acetyl-L-cysteine in rats

Oxidative stress disturbances have been reported in depressed patients and in animals submitted to stress. Recent evidence suggests that antidepressants may have antioxidant properties. However, the therapeutic potential of antioxidants as antidepressant drugs has not been systematically investigated. Therefore, this study tested the hypothesis that N-acetyl-L-cysteine (NAC), a cysteine prodrug with powerful antioxidant activity, would possess anti depressant-like properties in the forced swimming test. Male Wistar rats were subjected to 15 min of forced swimming and immediately afterward, 5, and 23 h later received intraperitoneal injections of NAC (5, 15, 50, 150, and 250 mg/kg), imipramine, 0 5 mg/kg) or vehicle. One hour later they were submitted to the 5 min test swimming session, where immobility time was recorded. Independent groups of animals received the same treatments and their exploratory activity was measured in an open arena for 5 min. NAC (at the doses of 15, 50, and 150 mg/kg) and imipramine induced a significant decrease in immobility time without changing exploratory behavior measured in an open arena. These results suggest that antioxidants such as NAC may have antidepressant effects. Behavioural Pharmacology (C) 19:747-750 2008 Wolters Kluwer Health vertical bar Lippincott Williams & Wilkins.

NMR structure and backbone dynamics of a concatemer of epidermal growth factor homology modules of the human low-density lipoprotein receptor

The ligand-binding region of the low-density lipoprotein (LDL) receptor is formed by seven N-terminal, imperfect, cysteine-rich (LB) modules. This segment is followed by an epidermal growth factor precursor homology domain with two N-terminal, tandem, EGF-like modules that are thought to participate in LDL binding and recycling of the endocytosed receptor to the cell surface. EGF-A and the concatemer, EGF-AB, of these modules were expressed in Escherichia coli. Correct protein folding of EGF-A and the concatemer EGF-AB was achieved in the presence or absence of calcium ions, in contrast to the LB modules, which require them for correct folding. Homonuclear and heteronuclear H-1-N-15 NMR spectroscopy at 17.6 T was used to determine the three-dimensional structure of the concatemer. Both modules are formed by two pairs of short, anti-parallel beta -strands. In the concatemer, these modules have a fixed relative orientation, stabilized by calcium ion-binding and hydrophobic interactions at the interface. N-15 longitudinal and transverse relaxation rates, and {H-1}-N-15 heteronuclear NOEs were used to derive a model-free description of the backbone dynamics of the molecule. The concatemer appears relatively rigid, particularly near the calcium ion-binding site at the module interface, with an average generalized order parameter of 0.85 +/- 0.11. Some mutations causing familial hypercholesterolemia may now be rationalized. Mutations of D41, D43 and E44 in the EGF-B calcium ion-binding region may affect the stability of the linker and thus the orientation of the tandem modules. The diminutive core also provides little structural stabilization, necessitating the presence of disulfide bonds. The structure and dynamics of EGF-AB contrast with the N-terminal LB modules, which require calcium ions both for folding to form the correct disulfide connectivities and for maintenance of the folded structure, and are connected by highly mobile linking peptides. (C) 2001 Academic Press.