Data parallel large-scale molecular dynamics for liquids

An efficient data parallel computational scheme is presented for large-scale molecular dynamics (MD ) simulations of liquids with short-range interactions. The method is based on decomposition of the simulation cell into equally sized subcells, with the shortest side length equal to the cutoff radius. Inter- and intracell interactions are calculated in a coarse-grained manner. A geometric sorting procedure, based on particle distances to subcell boundaries, is used to minimize the overall computations and the nonproductive communications. Using only nearest-neighbor communications, an efficient scheme is developed for periodic updates of the contents of subcells due to the migration of particles. Special “null-particles” are introduced, which act as buffers during the periodic updates and allow for a globally uniform algorithm during the calculations. Communication cost is about 7% of the total CPU time. The method is found to be linearly scalable with the number of particles, performing better as the ratio of virtual to physical processors increases. The MD code is written in Fortran 90 and implemented on a CM-200. The overall speed is approximately 5.9 μs. per MD step and per particle for 1 million particles and 5.5 μs for 5 million particles. The method should be easily transferred to other massively parallel computers of SIMD and MIMD type. © 1993 John Wiley & Sons, Inc.     

X-ray spectroscopy of enzyme active site analogues and related molecules: bis(dithiolene)molybdenum(IV) and -tungsten(IV,VI) complexes with variant terminal ligands

The X-ray absorption spectra at the molybdenum and selenium K-edges and the tungsten L2,3-edges are acquired for a set of 14 Mo(IV) and W(IV,VI) bis(dithiolene) complexes related to the active sites of molybdo- and tungstoenzymes. The set includes square pyramidal [MoIVL(S2C2Me2)2]- (L = O2-, R3SiO-, RO-, RS-, RSe-) and [WIV(OR)(S2C2Me2)2]-, distorted trigonal prismatic [MoIV(CO)(SeR)(S2C2Me2)2]- and [WIV(CO)L(S2C2Me2)2]- (L = RS-, RSe-), and distorted octahedral [WVIO(OR)(S2C2Me2)2]-. The dithiolene simulates the pterin-dithiolene cofactor ligand, and L represents a protein ligand. Bond lengths are determined by EXAFS analysis using the GNXAS protocol. Normalized edge spectra, non-phase-shift-corrected Fourier transforms, and EXAFS data and fits are presented. Bond lengths determined by EXAFS and X-ray crystallography agree to < or = 0.02 A as do the M-Se distances determined by both metal and selenium EXAFS. The complexes [MoIV(QR)(S2C2Me2)2]- simulate protein ligation by the DMSO reductase family of enzymes, including DMSO reductase itself (Q = O), dissimilatory nitrate reductase (Q = S), and formate dehydrogenase (Q = Se). Edge shifts of these complexes correlate with the ligand electronegativities. Terminal ligand binding is clearly distinguished in the presence of four Mo-S(dithiolene) interactions. Similarly, five-coordinate [ML(S2C2Me2)2]- and six-coordinate [M(CO)L(S2C2Me2)2]- are distinguishable by edge and EXAFS spectra. This study expands a previous XAS investigation of bis(dithiolene)metal(IV,V,VI) complexes (Musgrave, K. B.; Donahue, J. P.; Lorber, C.; Holm, R. H.; Hedman, B.; Hodgson, K. O. J. Am. Chem. Soc. 1999, 121, 10297) by including a larger inventory of molecules with variant physiologically relevant terminal ligation. The previous and present XAS results should prove useful in characterizing and refining metric features and structures of enzyme sites.     

Synthesis and spectroscopic studies of non-heme diiron(III) species with a terminal hydroperoxide ligand: models for hemerythrin

Two compounds, [Fe2(mu-OH)(mu-Ph4DBA)(TMEDA)2(OTf)] (4) and [Fe2(mu-OH)(mu-Ph4DBA)(DPE)2(OTf)] (7), where Ph4DBA(2-) is the dinucleating bis(carboxylate) ligand dibenzofuran-4,6-bis(diphenylacetate), have been prepared as synthetic models for the dioxygen-binding non-heme diiron protein hemerythrin (Hr). X-ray crystallography reveals that, in the solid state, these compounds contain the asymmetric coordination environment found at the diiron center in the reduced form of the protein, deoxyHr. M?ssbauer spectra of the models (4, delta = 1.21(2), DeltaE(Q) = 2.87(2) mm s(-1); 7, delta(av) = 1.23(1), DeltaE(Qav) = 2.79(1) mm s(-1)) and deoxyHr (delta = 1.19, DeltaE(Q) = 2.81 mm s(-1)) are also in good agreement. Oxygenation of the diiron(II) complexes dissolved in CH2Cl2 containing 3 equiv of N-MeIm (4) or neat EtCN (7) at -78 degrees C affords a red-orange solution with optical bands at 336 nm (7300 M(-1) cm(-1)) and 470 nm (2600 M(-1) cm(-1)) for 4 and at 334 nm (6400 M(-1) cm(-1)) and 484 nm (2350 M(-1) cm(-1)) for 7. These spectra are remarkably similar to that of oxyHr, 330 nm (6800 M(-1) cm(-1)) and 500 nm (2200 M(-1) cm(-1)). The electron paramagnetic resonance (EPR) spectrum of the cryoreduced, mixed-valence dioxygen adduct of 7 displays properties consistent with a (mu-oxo)diiron(II,III) core. An investigation of 7 and its dioxygen-bound adduct by extended X-ray absorption fine structure (EXAFS) spectroscopy indicates that the oxidized species contains a (mu-oxo)diiron(III) core with iron-ligand distances in agreement with those expected for oxide, carboxylate, and amine/hydroperoxide donor atoms. The analogous cobalt complex [Co2(mu-OH)(mu-Ph4DBA)(TMEDA)2(OTf)] (6) was synthesized and structurally characterized, but it was unreactive toward dioxygen.     

Determination by X-ray absorption spectroscopy of the Fe-Fe separation in the oxidized form of the hydroxylase of methane monooxygenase alone and in the presence of MMOD

The diiron active site in the hydroxylase of Methylococcus capsulatus (Bath) methane monooxygenase (MMOH) has been studied in the oxidized form by X-ray absorption spectroscopy (XAS). Previous investigations by XAS and X-ray crystallography have identified two different distances (3.0 and 3.4 angstroms) between the two Fe atoms in the dinuclear site. The present study has employed a systematic extended X-ray absorption fine structure (EXAFS) fitting methodology, utilizing known and simulated active site and relevant model structures, to determine unambiguously the Fe-Fe separation in the oxidized form of MMOH. Consistent and unique fits were only possible for an Fe-Fe distance of 3.0 angstroms. This methodology was then applied to study potential changes in the active site local structure in the presence of MMOD, a protein of unknown function in multicomponent MMO. Fe K-edge and EXAFS analyses revealed negligible changes in the diiron site electronic and geometric structure upon addition of MMOD to oxidized MMOH.     

Sulfur K-edge XAS and DFT calculations on [Fe4S4]2+ clusters: effects of H-bonding and structural distortion on covalency and spin topology

Sulfur K-edge X-ray absorption spectroscopy of a hydrogen-bonded elongated [Fe4S4]2+ cube is reported. The data show that this synthetic cube is less covalent than a normal compressed cube with no hydrogen bonding. DFT calculations reveal that the observed difference in electronic structure has significant contributions from both the cluster distortion and from hydrogen bonding. The elongated and compressed Fe4S4 structures are found to have different spin topologies (i.e., orientation of the delocalized Fe2S2 subclusters which are antiferromagnetically coupled to each other). It is suggested that the H-bonding interaction with the counterion does not contribute to the cluster elongation. A magneto-structural correlation is developed for the Fe4S4 cube that is used to identify the redox-active Fe2S2 subclusters in active sites of HiPIP and ferredoxin proteins involving these clusters.     

Microanalysis of catecholamines in human plasma by high-performance liquid chromatography with amperometric detection as compared with a radioenzymatic method

A high-performance liquid chromatographic procedure is described for the quantitative determination of epinephrine, norepinephrine, and dopamine in human plasma. The method, which is based on adsorption of the catecholamines to alumina and, after liberation, separation on a microparticulate bonded strong cation-exchange resin and amperometric detection, has been optimized to give complete baseline separation of the substances of interest. Dihydroxybenzylamine, a nonendogenous catecholamine, is used as the internal standard. The detection limit is about 0.1 pmol for dopamine. Analysis of data obtained for norepinephrine and epinephrine from a total of 59 plasma samples showed a good correlation to the corresponding values obtained with a radioenzymatic method. Some results from normal and pathological conditions are compared.     

An yttrium Mξ source for ESCA

A new source for producing YMξ radiation in an ESCA spectrometer is described. The system makes use of continuous evaporation of yttrium on a rotating anode. It is demonstrated that this scheme allows YMξ-excited electron spectra to be recorded without gradual loss of resolution and intensity due to anode oxidation. Electron spectra of Ar, Hg and N₂ excited by YMξ radiation are studied. Relative photoelectric cross-sections of the 5d_{$\text{5}\text{2}$}, 5d_{$\text{5}\text{2}$} and 6s orbitals in Hg and the four valence orbitals in N₂ are measured. The valence electron shake-up spectrum of N₂ is discussed.     

Fluorine Kα radiation for the excitation of electron spectra

The applicability of the F Kα X-ray line (hv = 676 eV) from potassium fluoride for the excitation of electron spectra is demonstrated. Relative photoionization cross-sections in Au and gaseous Hg have been measured. The F Kα radiation from KF, RbF and CsF has been studied. The Ba Mζ line (hv = 602 eV) is found to be too broad for use in ESCA, even when emitted from metallic barium.