Interpretation of the temperature-dependent color of blue copper protein mutants.

The electronic absorption spectrum of the mutant of the blue copper protein amicyanin with a pseudoazurin loop (AmiPse) shows a remarkable temperature dependence. The absorption band at approximately 460 nm increases at low temperature while the transition at approximately 600 nm is not much affected by a variation of the temperature. An approximate density functional theory (DFT) study of the active site model [Cu(II)(imidazole)(2)(SCH(3))(S(CH(3))(2))](+) (protein backbone and solvation neglected) leads to two local minimum structures (axial and rhomb) which both have a geometry close to that typical for blue copper proteins. One (rhomb) has two structurally different histidine donors, and this geometry is also found in most experimental type 1 structures. The two forms axial and rhomb are distortional isomers and are energetically almost degenerate. The temperature dependence of the spectrum of AmiPse is interpreted with a temperature-dependent change of the relative population of the two local minimum structures with slightly different energy. The 460 nm transition is believed to be due to preferential population of the structure rhomb; this is in agreement with the published assignment of the high energy transition, based on thorough spectroscopic and computational studies. Consequences of a perturbation of the "gas phase" structures axial and rhomb by the protein and solvation are also discussed on the basis of published, experimentally observed structures and spectroscopic data.

A new molecular mechanics force field for the oxidized form of blue copper proteins.

A molecular mechanics force field for blue copper proteins has been developed, based on a rigid potential energy surface scan of the Cu(II)/His/His/Cys/Met chromophore, using DFT (B3LYP) calculations and the AMBER force field for the protein backbone. The strain-energy-minimized structures of the model chromophore alone are in excellent agreement with the DFT-optimized structure, and those of the entire set of cupredoxins (five structures are considered) are, within the experimental error limits, in good agreement with the single crystal structural data. However, the structural variation in the computed structures is much smaller than those in the experimental structures. It is shown that, due to the large error limits in the experimental data, a validation of the force field with experimental structural data is impossible because, within the error limits, all experimental structures considered are virtually identical. A validation on the basis of spectroscopic data and their correlation with experimental and computed structural data is proposed, and, as a first example, the correlation of intensity ratios of the charge transfer transitions with a specific distortion mode is presented. The quality of the correlation, using the computed structures, is higher than that with the X-ray structures, and this indicates that the computed structures are meaningful.

Structural variation in transition-metal bispidine compounds.

The experimentally determined molecular structures of 40 transition metal complexes with the tetradentate bispyridine-substituted bispidone ligand, 2,4-bis(2-pyridine)-3,7-diazabicyclo[3.3.1]nonane-9-one [M(bisp)XYZ]n+; M = CrIII, MnII, FeII, CoII, CuII, CuI, ZnII; X, Y, Z = mono- or bidentate co-ligands; penta-, hexa- or heptacoordinate complexes) are characterized in detail, supported by force-field and DFT calculations. While the bispidine ligand is very rigid (N3...N7 distance = 2.933 +/- 0.025 A), it tolerates a large range of metal-donor bond lengths (2.07 A < sigma(M-N)/4 < 2.35 A). Of particular interest is the ratio of the bond lengths between the metal center and the two tertiary amine donors (0.84 A < M-N3/M-N7 < 1.05 A) and the fact that, in terms of this ratio there seem to be two clusters with M-N3 < M-N7 and M-N3 > or = M-N7. Calculations indicate that the two structural types are close to degenerate, and the structural form therefore depends on the metal ion, the number and type of co-ligands, as well as structural variations of the bispidine ligand backbone. Tuning of the structures is of importance since the structurally differing complexes have very different stabilities and reactivities.

Computation of cavity shapes, sizes, and plasticities.

A new molecular mechanics approach has been developed and used to scan the optimum geometry (size and shape) of a host molecule and the energy cost for the deformation of the bonding cavity, based on a general, unspecific guest with given docking sites and a variable size. Lagrange multipliers are used to constrain the sum of internal coordinates (host-guest docking-site distances), and no assumptions with respect to the type and strength of the host-guest bonding have to be made. This new approach has been fully implemented in a molecular mechanics program, and it is used to compute the size, shape, and plasticity of a rigid, asymmetrical, tetradentate (Namine )2 (Npyridine )2 ligand. It is shown that all other methods for the computation of ligand hole sizes that have been reported so far are not able to compute the ligand cavities independently of the metal ion, and they lead to strikingly different shapes, sizes, and plasticities. ©1999 John Wiley & Sons, Inc. J Comput Chem 20: 781-785, 1999.