Rui M. Almeida

Implications of oxidovanadium(IV) binding to actin.

Oxidovanadium(IV), a cationic species (VO(2+)) of vanadium(IV), binds to several proteins, including actin. Upon titration with oxidovanadium(IV), approximately 100% quenching of the intrinsic fluorescence of monomeric actin purified from rabbit skeletal muscle (G-actin) was observed, with a V(50) of 131 μM, whereas for the polymerized form of actin (F-actin) 75% of quenching was obtained and a V(50) value of 320 μM. Stern-Volmer plots were used to estimate an oxidovanadium(IV)-actin dissociation constant, with K(d) of 8.2 μM and 64.1 μM VOSO(4), for G-actin and F-actin, respectively. These studies reveal the presence of a high affinity binding site for oxidovanadium(IV) in actin, producing local conformational changes near the tryptophans most accessible to water in the three-dimensional structure of actin. The actin conformational changes, also confirmed by (1)H NMR, are accompanied by changes in G-actin hydrophobic surface, but not in F-actin. The (1)H NMR spectra of G-actin treated with oxidovanadium(IV) clearly indicates changes in the resonances ascribed to methyl group and aliphatic regions as well as to aromatics and peptide-bond amide region. In parallel, it was verified that oxidovanadium(IV) prevents the G-actin polymerization into F-actin. In the 0-200 μM range, VOSO(4) inhibits 40% of the extent of polymerization with an IC(50) of 15.1 μM, whereas 500 μM VOSO(4) totally suppresses actin polymerization. The data strongly suggest that oxidovanadium(IV) binds to actin at specific binding sites preventing actin polymerization. By affecting actin structure and function, oxidovanadium(IV) might be responsible for many cellular effects described for vanadium.

Rubredoxin as a paramagnetic relaxation-inducing probe.

The paramagnetic effect due to the presence of a metal center with unpaired electrons is no longer considered a hindrance in protein NMR spectroscopy. In the present work, the paramagnetic effect due to the presence of a metal center with unpaired electrons was used to map the interface of an electron transfer complex. Desulfovibrio gigas cytochrome c(3) was chosen as target to study the effect of the paramagnetic probe, Fe-rubredoxin, which produced specific line broadening in the heme IV methyl resonances M2(1) and M18(1). The rubredoxin binding surface in the complex with cytochrome c(3) was identified in a heteronuclear 2D NMR titration. The identified heme methyls on cytochrome c(3) are involved in the binding interface of the complex, a result that is in agreement with the predicted complexes obtained by restrained molecular docking, which shows a cluster of possible solutions near heme IV. The use of a paramagnetic probe in (1)HNMR titration and the mapping of the complex interface, in combination with a molecular simulation algorithm proved to be a valuable strategy to study electron transfer complexes involving non-heme iron proteins and cytochromes.

Gd(III) chelates as NMR probes of protein-protein interactions. Case study: rubredoxin and cytochrome c3.

Two cyclen-derived Gd probes, [Gd-DOTAM](3+) and [Gd-DOTP](5-) (DOTAM = 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetamide; DOTP = 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetrakis(methylenephosphonate)), were assessed as paramagnetic relaxation enhancement (PRE)-inducing probes for characterization of protein-protein interactions. Two proteins, Desulfovibrio gigas rubredoxin and Desulfovibrio gigas cytochrome c(3), were used as model partners. In a (1)H NMR titration it was shown that [Gd-DOTP](5-) binds to cytochrome c(3) near heme IV, causing pronounced PREs, characterized by line width broadenings of the heme methyl resonances at ratios as low as 0.08. A K(d) of 23 ± 1 μM was calculated based on chemical shift perturbation of selected heme methyl resonances belonging to three different heme groups, caused by allosteric effects upon [Gd-DOTP](5-) binding to cytochrome c(3) at a molar ratio of 2. The other probe, [Gd-DOTAM](3+), caused PREs on a well-defined patch near the metal center of rubredoxin (especially the patch constituted by residues D19-G23 and W37-S45, which broaden beyond detection). This effect was partially reversed for some resonances (C6-Y11, in particular) when cytochrome c(3) was added to this system. Both probes were successful in causing reversible PREs at the partner binding site, thus showing to be good probes to identify partners binding sites and since the interaction is reversible to structurally characterize protein complexes by better defining the complex interface.

Superoxide Reductase: Different Interaction Modes with its Two Redox Partners

Anaerobic organisms have molecular systems to detoxify reactive oxygen species when transiently exposed to oxygen. One of these systems is superoxide reductase, which reduces O2.− to H2O2 without production of molecular oxygen. In order to complete the reduction of superoxide anion, superoxide reductase requires an electron, delivered by its redox partners, which in Desulfovibrio gigas are rubredoxin and/or desulforedoxin. In this work, we characterized the interaction of Desulfovibrio gigas superoxide reductase with both electron donors by using steady‐state kinetics, 2D NMR titrations, and backbone relaxation measurements. The rubredoxin surface involved in the electron transfer complex with superoxide reductase comprises the solvent‐exposed hydrophobic residues in the vicinity of its metal center (Cys9, Gly10, Cys42, Gly43, and Ala44), and a Kd of 3 μM at 59 mM ionic strength was estimated by NMR. The ionic strength dependence of superoxide‐mediated rubredoxin oxidation by superoxide reductase has a maximum kapp of (37±12) min−1 at 157 mM. Relative to the electron donor desulforedoxin, its complex with superoxide reductase was not detected by chemical shift perturbation, though this protein is able to transfer electrons to superoxide reductase with a maximum kapp of (31±7) min−1 at an ionic strength of 57 mM. Competition experiments using steady‐state kinetics and NMR spectroscopy (backbone relaxation measurements and use of a paramagnetic relaxation enhancement probe) with Fe‐desulforedoxin in the presence of 15N‐Zn‐rubredoxin showed that these two electron donors compete for the same site on the enzyme surface, as shown in the model structure of the complex generated by using restrained molecular docking calculations. These combined strategies indicate that the two small electron donors bind in different manners, with the desulforedoxin complex being a short lived electron transfer complex or more dynamic, with many equivalent kinetically competent orientations.

Insights into the recognition and electron transfer steps in nitric oxide reductase from Marinobacter hydrocarbonoclasticus

Marinobacter hydrocarbonoclasticus nitric oxide reductase, cNOR, is an integral membrane protein composed of two subunits with different roles, NorC (electron transfer) and NorB (catalytic) that receives electrons from the soluble cytochrome c552 and reduces nitric oxide to nitrous oxide in the denitrification pathway. The solvent-exposed domain of NorC, harboring a c-type heme was heterologously produced, along with its physiological electron donor, cytochrome c552. These two proteins were spectroscopically characterized and shown to be similar to the native proteins, both being low-spin and Met-His coordinated, with the soluble domain of NorC presenting some additional features of a high-spin heme, which is consistent with the higher solvent accessibility of its heme and weaker coordination of the methionine axial ligand. The electron transfer complex between the two proteins has a 1:1 stoichiometry, and an upper limit for the dissociation constant was estimated by 1H NMR titration to be 1.2±0.4μM. Electrochemical techniques were used to characterize the interaction between the proteins, and a model structure of the complex was obtained by molecular docking. The electrochemical observations point to the modulation of the NorC reduction potential by the presence of NorB, tuning its ability to receive electrons from cytochrome c552.

Predicting Protein-Protein Interactions Using BiGGER: Case Studies

The importance of understanding interactomes makes preeminent the study of protein interactions and protein complexes. Traditionally, protein interactions have been elucidated by experimental methods or, with lower impact, by simulation with protein docking algorithms. This article describes features and applications of the BiGGER docking algorithm, which stands at the interface of these two approaches. BiGGER is a user-friendly docking algorithm that was specifically designed to incorporate experimental data at different stages of the simulation, to either guide the search for correct structures or help evaluate the results, in order to combine the reliability of hard data with the convenience of simulations. Herein, the applications of BiGGER are described by illustrative applications divided in three Case Studies: (Case Study A) in which no specific contact data is available; (Case Study B) when different experimental data (e.g., site-directed mutagenesis, properties of the complex, NMR chemical shift perturbation mapping, electron tunneling) on one of the partners is available; and (Case Study C) when experimental data are available for both interacting surfaces, which are used during the search and/or evaluation stage of the docking. This algorithm has been extensively used, evidencing its usefulness in a wide range of different biological research fields.

Electron transfer and docking between cytochrome cd1 nitrite reductase and different redox partners - A comparative study.

Cytochrome cd1 nitrite reductases (cd1NiRs) catalyze the reduction of nitrite to nitric oxide in denitrifying bacteria, such as Marinobacter hydrocarbonoclasticus. Previous work demonstrated that the enzymatic activity depends on a structural pre-activation triggered by the entry of electrons through the electron transfer (ET) domain, which houses a heme c center. The catalytic activity of M. hydrocarbonoclasticus cd1NiR (Mhcd1NiR) was tested by mediated electrochemistry, using small ET proteins and chemical redox mediators. The rate of enzymatic reaction depends on the nature of the redox partner, with cytochrome (cyt) c552 providing the highest value. In situations where cyt c552 is replaced by either a biological (cyt c from horse heart) or a chemical mediator the catalytic response was only observed at very low scan rates, suggesting that the intermolecular ET rate is much slower. Molecular docking simulations with the 3D model structure of Mhcd1NiR and cyt c552 or cyt c showed that hydrophobic interactions favor the formation of complexes where the heme c domain of the enzyme is the principal docking site. However, only in the case of cyt c552 the preferential areas of contact and Fe-Fe distances between heme c groups of the redox partners allow establishing competent ET pathways. The coupling of the enzyme with chemical redox mediators was also found not to be energetically favorable. These results indicate that although low activity functional complexes can be formed between Mhcd1NiR and different types of redox mediators, efficient ET is only observed with the putative physiological electron donor cyt c552.

Optimization of sitting posture classification based on user identification

In a precursory work, an intelligent sensing chair prototype was developed to classify 12 standardized sitting postures using 8 pneumatic bladders (4 in the chairs seat and 4 in the backrest) connected to piezoelectric sensors to measure inner pressure. A Classification of around 80% was obtained using Neural Networks. This work aims to demonstrate how algorithmic optimization can be applied to a newly developed prototype to improve posture classification performance. The aforementioned optimization is based on the split of users by sex and use two different previously trained Neural Networks (one for Male and the other for Female). Results showed that the best neural network parameters had an overall classification 89.0% (from the 92.1% for Female Classification and 85.8% for Male, which translates into an overall optimization of around 8%). Automatic separation of these sets was achieved with Decision Trees with an overall classification optimization of 87.1%.

Optimization of Sitting Posture Classification based on Anthropometric Data

An intelligent chair prototype was developed in order to detect and correct the adoption of bad sitting postures during long periods of time. A pneumatic system was enclosed in the chair (4 air bladders inside the seat pad and 4 in the backrest) to classify 12 standardized sitting postures, with a classification score of 80.9%. Recently we used algorithmic optimization applied to the existing classification algorithm (based on Neural Networks) to split users (using Classification Trees) by their sex and used two different previously trained Neural Networks (Male and Female) to get an improved classification of 89.0% when the user was identified and 87.1% for unidentified users. In this work we aim to investigate the usage of the anthropometric information (height and weight) to further optimize our classification process. Here we use four Machine Learning Techniques (Neural Networks, Support Vector Machines, Classification Trees and Naive Bayes) to automatically split the users in 2 classes (above and below the specific anthropometric median value). Results showed that Classification Trees worked best on automatically separating the body characteristics (i.e. Height) with a global optimization of 88.3%. During the classification process, if the user is identified, we skip the splitting step, and this optimization increases to 90.2%.

Real-Time Fuzzy Monitoring of Sitting Posture: Development of a New Prototype and a New Posture Classification Algorithm to Detect Postural Transitions

In a previous work, a chair prototype was used to detect 11 standardized siting postures of users, using just 8 air bladders (4 in the chair’s seat and 4 in the backrest) and one pressure sensor for each bladder. In this paper we describe the development of a new prototype, which is able to classify 12 standard postures with an overall score of 80.9 % (using a Neural Network Algorithm). We tested how this Algorithm worked during postural transitions (frontal and lateral flexion) and in intermediate postures, identifying some limitation of this Algorithm. This prompted the development of a Posture Classification Algorithm based on Fuzzy Logic and is able to determine if the user is adopting a good or a bad posture for specific time periods, using as input the Centre of Pressure, the Posture Adoption Time and the Posture Output from the existing Neural Network Algorithm. This newly developed Classification Algorithms is advancing the development of new Posture Correction Algorithms based on Fuzzy Actuators.