RESUMO
ReaxFF is a computationally efficient force field to simulate complex reactive dynamics in extended molecular models with diverse chemistries, if reliable force-field parameters are available for the chemistry of interest. If not, they must be optimized by minimizing the error ReaxFF makes on a relevant training set. Because this optimization is far from trivial, many methods, in particular, genetic algorithms (GAs), have been developed to search for the global optimum in parameter space. Recently, two alternative parameter calibration techniques were proposed, that is, Monte-Carlo force field optimizer (MCFF) and covariance matrix adaptation evolutionary strategy (CMA-ES). In this work, CMA-ES, MCFF, and a GA method (OGOLEM) are systematically compared using three training sets from the literature. By repeating optimizations with different random seeds and initial parameter guesses, it is shown that a single optimization run with any of these methods should not be trusted blindly: nonreproducible, poor or premature convergence is a common deficiency. GA shows the smallest risk of getting trapped into a local minimum, whereas CMA-ES is capable of reaching the lowest errors for two-third of the cases, although not systematically. For each method, we provide reasonable default settings, and our analysis offers useful guidelines for their usage in future work. An important side effect impairing parameter optimization is numerical noise. A detailed analysis reveals that it can be reduced, for example, by using exclusively unambiguous geometry optimization in the training set. Even without this noise, many distinct near-optimal parameter vectors can be found, which opens new avenues for improving the training set and detecting overfitting artifacts.
RESUMO
Tularemia caused by Francisella tularensis is a zoonotic infection of the Northern Hemisphere that mainly affects the skin, lymph nodes, bloodstream, and lungs. Other manifestations of tularemia are very rare, especially those with musculoskeletal involvement. Presenting in 2016, we diagnosed two cases of periprosthetic knee joint infections (PJI) caused by Francisella tularensis in Europe (one in Switzerland and one in the Czech Republic). We found only two other PJI cases in the literature, another knee PJI diagnosed 1999 in Ontario, Canada, and one hip PJI in Illinois, USA, in 2017. Diagnosis was made in all cases by positive microbiological cultures after 3, 4, 7, and 12 days. All were successfully treated, two cases by exchange of the prosthesis, one with debridement and retention, and one with repeated aspiration of the synovial fluid only. Antibiotic treatment was given between 3 weeks and 12 months with either ciprofloxacin-rifampin or with doxycycline alone or doxycycline in combination with gentamicin. Zoonotic infections should be considered in periprosthetic infections in particular in culture-negative PJIs with a positive histology or highly elevated leukocyte levels in synovial aspiration. Here, we recommend prolonging cultivation time up to 14 days, performing specific PCR tests, and/or conducting epidemiologically appropriate serological tests for zoonotic infections, including that for F. tularensis.
Assuntos
Antibacterianos/uso terapêutico , Articulação do Joelho/microbiologia , Infecções Relacionadas à Prótese/diagnóstico , Infecções Relacionadas à Prótese/microbiologia , Tularemia/diagnóstico , Idoso , Idoso de 80 Anos ou mais , Animais , Anticorpos Antibacterianos/sangue , Técnicas Bacteriológicas , Feminino , Francisella tularensis , Humanos , Articulação do Joelho/efeitos dos fármacos , Masculino , Infecções Relacionadas à Prótese/tratamento farmacológico , Estudos Retrospectivos , Líquido Sinovial/microbiologia , Resultado do Tratamento , Tularemia/complicações , Tularemia/tratamento farmacológico , Zoonoses/diagnóstico , Zoonoses/tratamento farmacológico , Zoonoses/microbiologiaRESUMO
Computational studies of the reaction mechanisms of various enzymes are nowadays based almost exclusively on hybrid QM/MM models. Unfortunately, the success of this approach strongly depends on the selection of the QM region, and computational cost is a crucial limiting factor. An interesting alternative is offered by empirical reactive molecular force fields, especially the ReaxFF potential developed by van Duin and co-workers. However, even though an initial parametrization of ReaxFF for biomolecules already exists, it does not provide the desired level of accuracy. We have conducted a thorough refitting of the ReaxFF force field to improve the description of reaction energetics. To minimize the human effort required, we propose a fully automated approach to generate an extensive training set comprised of thousands of different geometries and molecular fragments starting from a few model molecules. Electrostatic parameters were optimized with QM electrostatic potentials as the main target quantity, avoiding excessive dependence on the choice of reference atomic charges and improving robustness and transferability. The remaining force field parameters were optimized using the VD-CMA-ES variant of the CMA-ES optimization algorithm. This method is able to optimize hundreds of parameters simultaneously with unprecedented speed and reliability. The resulting force field was validated on a real enzymatic system, ppGalNAcT2 glycosyltransferase. The new force field offers excellent qualitative agreement with the reference QM/MM reaction energy profile, matches the relative energies of intermediate and product minima almost exactly, and reduces the overestimation of transition state energies by 27-48% compared with the previous parametrization.
Assuntos
Automação , Glicosiltransferases/metabolismo , Simulação de Dinâmica Molecular , Teoria Quântica , Algoritmos , Glicosiltransferases/química , Humanos , Eletricidade EstáticaRESUMO
Hybrid QM/MM computational studies can provide invaluable insight into the mechanisms of enzymatic reactions that can be exploited for rational drug design. Various approaches are available for such studies. However, their strengths and weaknesses may not be immediately apparent. Using the retaining glycosyltransferase ppGalNAcT2 as a case study, we compare different methodologies used to obtain reaction paths and transition state information. Ab Initio MD using CPMD coupled with the String Method is used to derive the minimum free energy reaction path. The geometrical features of the free energy path, especially around the transition state, agree with the minimum potential energy path obtained by the much less computationally expensive Nudged Elastic Band method. The barrier energy, however, differs by 8 kcal/mol. The free energy surface generated by metadynamics provides a rough overview of the reaction and can confirm the physical relevance of optimized paths or provide an initial guess for path optimization methods. Calculations of enzymatic reactions are usually performed at best at the DFT level of theory. A comparison of widely used functionals with high-level DLPNO-CCSD(T)/CBS data on the potential energy profile serves as a validation of the usability of DFT for this type of enzymatic reaction. The M06-2X meta-hybrid functional in particular matches the DLPNO-CCSD(T)/CBS reference extremely well with errors within 1 kcal/mol. However, even pure-GGA functional OPBE provides sufficient accuracy for this type of study.
Assuntos
Glicosiltransferases/química , Modelos Moleculares , Teoria Quântica , Glicosiltransferases/metabolismo , Magnésio/química , Estrutura Terciária de Proteína , TermodinâmicaRESUMO
The glycosylation of cell surface proteins plays a crucial role in a multitude of biological processes, such as cell adhesion and recognition. To understand the process of protein glycosylation, the reaction mechanisms of the participating enzymes need to be known. However, the reaction mechanism of retaining glycosyltransferases has not yet been sufficiently explained. Here we investigated the catalytic mechanism of human isoform 2 of the retaining glycosyltransferase polypeptide UDP-GalNAc transferase by coupling two different QM/MM-based approaches, namely a potential energy surface scan in two distance difference dimensions and a minimum energy reaction path optimisation using the Nudged Elastic Band method. Potential energy scan studies often suffer from inadequate sampling of reactive processes due to a predefined scan coordinate system. At the same time, path optimisation methods enable the sampling of a virtually unlimited number of dimensions, but their results cannot be unambiguously interpreted without knowledge of the potential energy surface. By combining these methods, we have been able to eliminate the most significant sources of potential errors inherent to each of these approaches. The structural model is based on the crystal structure of human isoform 2. In the QM/MM method, the QM region consists of 275 atoms, the remaining 5776 atoms were in the MM region. We found that ppGalNAcT2 catalyzes a same-face nucleophilic substitution with internal return (SNi). The optimized transition state for the reaction is 13.8 kcal/mol higher in energy than the reactant while the energy of the product complex is 6.7 kcal/mol lower. During the process of nucleophilic attack, a proton is synchronously transferred to the leaving phosphate. The presence of a short-lived metastable oxocarbenium intermediate is likely, as indicated by the reaction energy profiles obtained using high-level density functionals.
Assuntos
Glicosiltransferases/química , Glicosiltransferases/ultraestrutura , Modelos Químicos , Modelos Moleculares , Polissacarídeos/química , Polissacarídeos/ultraestrutura , Algoritmos , Sítios de Ligação , Catálise , Simulação por Computador , Glicosilação , Cinética , Ligação Proteica , Conformação ProteicaRESUMO
Solvent-driven aggregation of a series of porphyrin derivatives was studied by UV/Vis and circular dichroism spectroscopy. The porphyrins are characterised by the presence in the meso positions of steroidal moieties further conjugated with glucosyl groups. The presence of these groups makes the investigated macrocycles amphiphilic and soluble in aqueous solvent, namely, dimethyl acetamide/water. Aggregation of the macrocycles is triggered by a change in bulk solvent composition leading to formation of large architectures that express supramolecular chirality, steered by the presence of the stereogenic centres on the periphery of the macrocycles. The aggregation behaviour and chiroptical features of the aggregates are strongly dependent on the number of moieties decorating the periphery of the porphyrin framework. In particular, experimental evidence indicates that the structure of the steroid linker dictates the overall chirality of the supramolecular architectures. Moreover, the porphyrin concentration strongly affects the aggregation mechanism and the CD intensities of the spectra. Notably, AFM investigations reveal strong differences in aggregate morphology that are dependent on the nature of the appended functional groups, and closely in line with the changes in aggregation mechanism. The suprastructures formed at lower concentration show a network of long fibrous structures spanning over tens of micrometres, whereas the aggregates formed at higher concentration have smaller rod-shaped structures that can be recognised as the result of coalescence of smaller globular structures. The fully steroid substituted derivative forms globular structures over the whole concentration range explored. Finally, a rationale for the aggregation phenomena was given by semiempirical calculations at the PM6 level.
Assuntos
Glicosídeos/química , Porfirinas/química , Esteroides/química , Dicroísmo Circular , Cinética , Microscopia de Força Atômica , Estrutura Molecular , Solventes , Espectrofotometria Ultravioleta , Água/químicaRESUMO
o-Phthaldialdehyde was used in combination with a series of six thiosugars for the pre-column chiral derivatization of selected primary amino acids. The diastereomers were resolved on a conventional reversed-phase column and with fluorescence detection. A surprisingly effective resolution of 1-isoindolyl-(1-thioglycosides) derived from 1-thio-beta-L-fucose was observed.
Assuntos
Leucina/análogos & derivados , Tioaçúcares/química , Leucina/química , EstereoisomerismoRESUMO
The reaction system containing Cu(II), hydrogen peroxide and D-arabinono-1,4-lactone was found to be effective in the decolorization and reduction of toxicity of azo, thiazine-, triphenylmethane- and anthraquinone-based synthetic dyes. More than 85% decolorization was obtained with 100ppm Acridine Orange, Azure B, Chicago Sky Blue 6B, Crystal Violet, Evans Blue, Poly B-411, Reactive Blue 2, Reactive Blue 5, and Remazol Brilliant Blue R incubated for 24h in the presence of 10mM CuSO(4), 20mM D-arabinono-1,4-lactone and 80 mM H(2)O(2). The rate of decolorization was not affected by pH in the range of 3-9. The rapid decolorization was accompanied by a fast decomposition of H(2)O(2) in the reaction mixture and by a fast production of hydroxyl radicals.
Assuntos
Corantes/química , Cobre/química , Poluição Ambiental/prevenção & controle , Peróxido de Hidrogênio/química , Concentração de Íons de Hidrogênio , Medições Luminescentes , Espectrofotometria UltravioletaRESUMO
Selected azo, acridine, triphenyl methane, anthraquinone and thiazine-based dyes were decolorized using a catalytic system consisting of Cu(II)/glucaric acid/H(2)O(2). More than 90% decolorization was obtained with 100 ppm Acridine Orange, Azure B, Chicago Sky Blue, Crystal Violet, Methyl Orange, Poly B-411, Reactive Black 5, Reactive Blue 2, and Remazol Brilliant Blue R within 24 h. Seventy to eighty percent decolorization was achieved within the first 6 h. The decolorizaton was not affected by pH. The involvement of hydroxyl radicals produced in the system in the decolorization of the dye molecules was confirmed by electron spin resonance study.