Biomedical informatics training at the University of Wisconsin-Madison.

OBJECTIVES: The purpose of this paper is to describe biomedical informatics training at the University of Wisconsin-Madison (UW-Madison). METHODS: We reviewed biomedical informatics training, research, and faculty/trainee participation at UW-Madison. RESULTS: There are three primary approaches to training 1) The Computation & Informatics in Biology & Medicine Training Program, 2) formal biomedical informatics offered by various campus departments, and 3) individualized programs. Training at UW-Madison embodies the features of effective biomedical informatics training recommended by the American College of Medical Informatics that were delineated as: 1) curricula that integrate experiences among computational sciences and application domains, 2) individualized and interdisciplinary cross-training among a diverse cadre of trainees to develop key competencies that he or she does not initially possess, 3) participation in research and development activities, and 4) exposure to a range of basic informational and computational sciences. CONCLUSIONS: The three biomedical informatics training approaches immerse students in multidisciplinary training and education that is supported by faculty trainers who participate in collaborative research across departments. Training is provided across a range of disciplines and available at different training stages. Biomedical informatics training at UW-Madison illustrates how a large research University, with multiple departments across biological, computational and health fields, can provide effective and productive biomedical informatics training via multiple bioinformatics training approaches.

Ligand pathways in myoglobin: a review of Trp cavity mutations.

The pathways for ligand entry and exit in myoglobin have now been well established by a wide variety of experimental results, including pico- to nano- to microsecond transient absorbance measurements and time-resolved X-ray crystallographic measurements. Trp insertions have been used to block, one at a time, the three major cavities occupied by photodissociated ligands. In this work, we review the effects of the L29(B10)W mutation, which places a large indole ring in the initial 'docking site' for photodissociated ligands. Then, the effects of blocking the Xe4 site with I28W, V68W, and I107W mutations and the Xe1 cavity with L89W, L104W, and F138W mutations are described. The structures of four of these mutants are shown for the first time (Trp28, Trp68, Trp107, and Trp 138 sperm whale metMb). All available results support a 'side path' mechanism in which ligands move into and out of myoglobin by outward rotation of the HisE7 side chain, but after entry can migrate into internal cavities, including the distal Xe4 and proximal Xe1 binding sites. The distal cavities act like the pocket of a baseball glove, catching the ligand and holding it long enough for the histidine gate to close and facilitate internal coordination with the heme iron atom. The physiological role of the proximal Xe1 site is less clear because changes in the size of this cavity have minimal effects on overall O(2) binding parameters.

Role of CO2 in the formation of gold deposits.

Much of global gold production has come from deposits with uneconomic concentrations of base metals, such as copper, lead and zinc. These 'gold-only' deposits are thought to have formed from hot, aqueous fluids rich in carbon dioxide, but only minor significance has been attached to the role of the CO2 in the process of gold transport. This is because chemical bonding between gold ions and CO2 species is not strong, and so it is unlikely that CO2 has a direct role in gold transport. An alternative indirect role for CO2 as a weak acid that buffers pH has also appeared unlikely, because previously inferred pH values for such gold-bearing fluids are variable. Here we show that such calculated pH values are unlikely to record conditions of gold transport, and propose that CO2 may play a critical role during gold transport by buffering the fluid in a pH range where elevated gold concentration can be maintained by complexation with reduced sulphur. Our conclusions, which are supported by geochemical modelling, may provide a platform for new gold exploration methods.

A fast Newton method for entropy maximization in statistical phase estimation.

A fast Newton method is presented for solving the entropy maximization problem in the Bayesian statistical approach to phase estimation. The method requires only O(n log n) instead of standard O(n3) floating point operations per iteration, while converging in the same rate as the standard Newton method. The method is described and related computational issues are discussed. Numerical results on simple test cases are also presented to demonstrate the behavior of the method.

Molecular structure of dihydroorotase: a paradigm for catalysis through the use of a binuclear metal center.

Dihydroorotase plays a key role in pyrimidine biosynthesis by catalyzing the reversible interconversion of carbamoyl aspartate to dihydroorotate. Here we describe the three-dimensional structure of dihydroorotase from Escherichia coli determined and refined to 1.7 A resolution. Each subunit of the homodimeric enzyme folds into a "TIM" barrel motif with eight strands of parallel beta-sheet flanked on the outer surface by alpha-helices. Unexpectedly, each subunit contains a binuclear zinc center with the metal ions separated by approximately 3.6 A. Lys 102, which is carboxylated, serves as a bridging ligand between the two cations. The more buried or alpha-metal ion in subunit I is surrounded by His 16, His 18, Lys 102, Asp 250, and a solvent molecule (most likely a hydroxide ion) in a trigonal bipyramidal arrangement. The beta-metal ion, which is closer to the solvent, is tetrahedrally ligated by Lys 102, His 139, His 177, and the bridging hydroxide. L-Dihydroorotate is observed bound to subunit I, with its carbonyl oxygen, O4, lying 2.9 A from the beta-metal ion. Important interactions for positioning dihydroorotate into the active site include a salt bridge with the guanidinium group of Arg 20 and various additional electrostatic interactions with both protein backbone and side chain atoms. Strikingly, in subunit II, carbamoyl L-aspartate is observed binding near the binuclear metal center with its carboxylate side chain ligating the two metals and thus displacing the bridging hydroxide ion. From the three-dimensional structures of the enzyme-bound substrate and product, it has been possible to propose a unique catalytic mechanism for dihydroorotase. In the direction of dihydroorotate hydrolysis, the bridging hydroxide attacks the re-face of dihydroorotate with general base assistance by Asp 250. The carbonyl group is polarized for nucleophilic attack by the bridging hydroxide through a direct interaction with the beta-metal ion. During the cyclization of carbamoyl aspartate, Asp 250 initiates the reaction by abstracting a proton from N3 of the substrate. The side chain carboxylate of carbamoyl aspartate is polarized through a direct electrostatic interaction with the binuclear metal center. The ensuing tetrahedral intermediate collapses with C-O bond cleavage and expulsion of the hydroxide which then bridges the binuclear metal center.

How the CO in myoglobin acquired its bend: lessons in interpretation of crystallographic data.

Contrary to the expectation of chemists, the first X-ray structures of carbon monoxide bound to myoglobin (Mb) showed a highly distorted Fe-C-O bond system. These results appeared to support the idea of a largely steric mechanism for discrimination by the protein against CO binding, a lethal act for the protein in terms of its physiological function. The most recent independently determined high-resolution structures of Mb-CO have allowed the 25 year old controversy concerning the mode of CO binding to be resolved. The CO is now seen to bind in a roughly linear fashion without substantial bending, consistent with chemical expectations and spectroscopic measurements. Access to deposited diffraction data prompted a reevaluation of the sources of the original misinterpretation. A series of careful refinements of models against the data at high (1.1 A) and modest resolutions (1.5 A) have been performed in anisotropic versus isotropic modes. The results suggest that the original artifact was a result of lower quality crystals combined with anisotropic motion and limited resolution of the diffraction data sets. This retrospective analysis should serve as a caution for all researchers using structural tools to draw far-reaching biochemical conclusions.

Waterproofing the heme pocket. Role of proximal amino acid side chains in preventing hemin loss from myoglobin.

The ability of myoglobin to bind oxygen reversibly depends critically on retention of the heme prosthetic group. Globin side chains at the Leu(89)(F4), His(97)(FG3), Ile(99)(FG5), and Leu(104)(G5) positions on the proximal side of the heme pocket strongly influence heme affinity. The roles of these amino acids in preventing heme loss have been examined by determining high resolution structures of 14 different mutants at these positions using x-ray crystallography. Leu(89) and His(97) are important surface amino acids that interact either sterically or electrostatically with the edges of the porphyrin ring. Ile(99) and Leu(104) are located in the interior region of the proximal pocket beneath ring C of the heme prosthetic group. The apolar amino acids Leu(89), Ile(99), and Leu(104) "waterproof" the heme pocket by forming a barrier to solvent penetration, minimizing the size of the proximal cavity, and maintaining a hydrophobic environment. Substitutions with smaller or polar side chains at these positions result in exposure of the heme to solvent, the appearance of crystallographically defined water molecules in or near the proximal pocket, and large increases in the rate of hemin loss. Thus, the naturally occurring amino acid side chains at these positions serve to prevent hydration of the His(93)-Fe(III) bond and are highly conserved in all known myoglobins and hemoglobins.

Crystal structure of a nonsymbiotic plant hemoglobin.

BACKGROUND: Nonsymbiotic hemoglobins (nsHbs) form a new class of plant proteins that is distinct genetically and structurally from leghemoglobins. They are found ubiquitously in plants and are expressed in low concentrations in a variety of tissues including roots and leaves. Their function involves a biochemical response to growth under limited O(2) conditions. RESULTS: The first X-ray crystal structure of a member of this class of proteins, riceHb1, has been determined to 2.4 A resolution using a combination of phasing techniques. The active site of ferric riceHb1 differs significantly from those of traditional hemoglobins and myoglobins. The proximal and distal histidine sidechains coordinate directly to the heme iron, forming a hemichrome with spectral properties similar to those of cytochrome b(5). The crystal structure also shows that riceHb1 is a dimer with a novel interface formed by close contacts between the G helix and the region between the B and C helices of the partner subunit. CONCLUSIONS: The bis-histidyl heme coordination found in riceHb1 is unusual for a protein that binds O(2) reversibly. However, the distal His73 is rapidly displaced by ferrous ligands, and the overall O(2) affinity is ultra-high (K(D) approximately 1 nM). Our crystallographic model suggests that ligand binding occurs by an upward and outward movement of the E helix, concomitant dissociation of the distal histidine, possible repacking of the CD corner and folding of the D helix. Although the functional relevance of quaternary structure in nsHbs is unclear, the role of two conserved residues in stabilizing the dimer interface has been identified.

Crystal structure of tropomyosin at 7 Angstroms resolution.

Tropomyosin is a 400A-long coiled coil that polymerizes to form a continuous filament that associates with actin in muscle and numerous non-muscle cells. Tropomyosin and troponin together form a calcium-sensitive switch that is responsible for thin-filament regulation of striated muscle. Subtle structural features of the molecule, including non-canonical aspects of its coiled-coil motif, undoubtedly influence its association with f-actin and its role in thin filament regulation. Previously, careful inspection of native diffraction intensities was sufficient to construct a model of tropomyosin at 9A resolution in a spermine-induced crystal form that diffracts anisotropically to 4A resolution. Single isomorphous replacement (SIR) phasing has now provided an empirical determination of the structure at 7A resolution. A novel method of heavy-atom analysis was used to overcome difficulties in interpretation of extremely anisotropic diffraction. The packing arrangement of the molecules in the crystal, and important aspects of the tropomyosin geometry such as non-uniformities of the pitch and variable bending and radius of the coiled coil are evident.

Protein-assisted pericyclic reactions: an alternate hypothesis for the action of quantal receptors.

The rules for allowable pericyclic reactions indicate that the photoisomerizations of retinals in rhodopsins can be formally analogous to thermally promoted Diels-Alder condensations of monoenes with retinols. With little change in the seven-transmembrane helical environment these latter reactions could mimic the retinal isomerization while providing highly sensitive chemical reception. In this way archaic progenitors of G-protein-coupled chemical quantal receptors such as those for pheromones might have been evolutionarily plagiarized from the photon quantal receptor, rhodopsin, or vice versa. We investigated whether the known structure of bacteriorhodopsin exhibited any similarity in its active site with those of the two known antibody catalysts of Diels-Alder reactions and that of the photoactive yellow protein. A remarkable three-dimensional motif of aromatic side chains emerged in all four proteins despite the drastic differences in backbone structure. Molecular orbital calculations supported the possibility of transient pericyclic reactions as part of the isomerization-signal transduction mechanisms in both bacteriorhodopsin and the photoactive yellow protein. It appears that reactions in all four of the proteins investigated may be biological analogs of the organic chemists' chiral auxiliary-aided Diels-Alder reactions. Thus the light receptor and the chemical receptor subfamilies of the heptahelical receptor family may have been unified at one time by underlying pericyclic chemistry.

Metal-ion affinity and specificity in EF-hand proteins: coordination geometry and domain plasticity in parvalbumin.

BACKGROUND: The EF-hand family is a large set of Ca(2+)-binding proteins that contain characteristic helix-loop-helix binding motifs that are highly conserved in sequence. Members of this family include parvalbumin and many prominent regulatory proteins such as calmodulin and troponin C. EF-hand proteins are involved in a variety of physiological processes including cell-cycle regulation, second messenger production, muscle contraction, microtubule organization and vision. RESULTS: We have determined the structures of parvalbumin mutants designed to explore the role of the last coordinating residue of the Ca(2+)-binding loop. An E101D substitution has been made in the parvalbumin EF site. The substitution decreases the Ca(2+)-binding affinity 100-fold and increases the Mg(2+)-binding affinity 10-fold. Both the Ca(2+)- and Mg(2+)-bound structures have been determined, and a structural basis has been proposed for the metal-ion-binding properties. CONCLUSIONS: The E101D mutation does not affect the Mg(2+) coordination geometry of the binding loop, but it does pull the F helix 1.1 A towards the loop. The E101D-Ca(2+) structure reveals that this mutant cannot obtain the sevenfold coordination preferred by Ca(2+), presumably because of strain limits imposed by tertiary structure. Analysis of these results relative to previously reported structural information supports a model wherein the characteristics of the last coordinating residue and the plasticity of the Ca(2+)-binding loop delimit the allowable geometries for the coordinating sphere.

Effects of the location of distal histidine in the reaction of myoglobin with hydrogen peroxide.

To clarify how the location of distal histidine affects the activation process of H2O2 by heme proteins, we have characterized reactions with H2O2 for the L29H/H64L and F43H/H64L mutants of sperm whale myoglobin (Mb), designed to locate the histidine farther from the heme iron. Whereas the L29H/H64L double substitution retarded the reaction with H2O2, an 11-fold rate increase versus wild-type Mb was observed for the F43H/H64L mutant. The Vmax values for 1-electron oxidations by the myoglobins correlate well with the varied reactivities with H2O2. The functions of the distal histidine as a general acid-base catalyst were examined based on the reactions with cumene hydroperoxide and cyanide, and only the histidine in F43H/H64L Mb was suggested to facilitate heterolysis of the peroxide bond. The x-ray crystal structures of the mutants confirmed that the distal histidines in F43H/H64L Mb and peroxidase are similar in distance from the heme iron, whereas the distal histidine in L29H/H64L Mb is located too far to enhance heterolysis. Our results indicate that the proper positioning of the distal histidine is essential for the activation of H2O2 by heme enzymes.

Protein structure determination using a database of interatomic distance probabilities.

The accelerated pace of genomic sequencing has increased the demand for structural models of gene products. Improved quantitative methods are needed to study the many systems (e.g., macromolecular assemblies) for which data are scarce. Here, we describe a new molecular dynamics method for protein structure determination and molecular modeling. An energy function, or database potential, is derived from distributions of interatomic distances obtained from a database of known structures. X-ray crystal structures are refined by molecular dynamics with the new energy function replacing the Van der Waals potential. Compared to standard methods, this method improved the atomic positions, interatomic distances, and side-chain dihedral angles of structures randomized to mimic the early stages of refinement. The greatest enhancement in side-chain placement was observed for groups that are characteristically buried. More accurate calculated model phases will follow from improved interatomic distances. Details usually seen only in high-resolution refinements were improved, as is shown by an R-factor analysis. The improvements were greatest when refinements were carried out using X-ray data truncated at 3.5 A. The database potential should therefore be a valuable tool for determining X-ray structures, especially when only low-resolution data are available.

Distribution of calcitonin gene-related peptide, atrial natriuretic peptide and neuropeptide Y in the rat heart.

Calcitonin gene-related peptide (CGRP) is a novel neuropeptide, predicted on the basis of structural analysis of the rat calcitonin gene. It is a neurotransmitter which has been suggested to take part in sensory transmission. In this study, we have examined the distribution of this peptide, alpha-atrial natriuretic peptide immunoreactivity (irANP) and neuropeptide Y (NPY) within different regions of the rat heart. Attempts were also made to compare the distributions of these peptides in the regions examined, through different methods of immunocytochemistry and further comparing these results with those obtained through radioimmunoassay. The distributions of the peptides in the atria were similar to results obtained with radioimmunoassay, but there were no myocytes containing irANP in the ventricles with immunocytochemistry as opposed to radioimmunoassay. While the staining obtained for irANP in the atrium was more intense in the right, CGRP and NPY nerve fibres were two to three times more abundant in the left atrium. The high local concentration of a vasoconstrictor peptide in the region of coronary vessels may suggest that it is involved in the control of vascular smooth muscle tone. The method of choice with the immunocytochemical studies was that of the susa wax technique for irANP. Caution should therefore be observed when interpreting results based only on a single staining technique.

Crystal structures of Bacillus stearothermophilus adenylate kinase with bound Ap5A, Mg2+ Ap5A, and Mn2+ Ap5A reveal an intermediate lid position and six coordinate octahedral geometry for bound Mg2+ and Mn2+.

Crystal structures of Bacillus stearothermophilus adenylate kinase with bound Ap5A, Mn2+ Ap5A, and Mg2+ Ap5A have been determined by X-ray crystallography to resolutions of 1.6 A, 1.85 A, and 1.96 A, respectively. The protein's lid domain is partially open, being both rotated and translated away from bound Ap5A. The flexibility of the lid domain in the ternary state and its ability to transfer force directly to the the active site is discussed in light of our proposed entropic mechanism for catalytic turnover. The bound Zn2+ atom is demonstrably structural in nature, with no contacts other than its ligating cysteine residues within 5 A. The B. stearothermophilus adenylate kinase lid appears to be a truncated zinc finger domain, lacking the DNA binding finger, which we have termed a zinc knuckle domain. In the Mg2+ Ap5A and Mn2+ Ap5A structures, Mg2+ and Mn2+ demonstrate six coordinate octahedral geometry. The interactions of the Mg2+-coordinated water molecules with the protein and Ap5A phosphate chain demonstrate their involvement in catalyzing phosphate transfer. The protein selects for beta-y (preferred by Mg2+) rather than alpha-gamma (preferred by Mn2+) metal ion coordination by forcing the ATP phosphate chain to have an extended conformation.

Characterizing global substates of myoglobin.

BACKGROUND: The massive amount of information generated from current molecular dynamics simulations makes the data difficult to analyze efficiently. Principal component analysis has been used for almost a century to detect and characterize data relationships and to reduce the dimensionality for problems in many fields. Here, we present an adaptation of principal component analysis using a partial singular value decomposition (SVD) for investigating both the localized and global motions of macromolecules. RESULTS: Configuration space projections from the SVD analysis of a variety of myoglobin simulations are used to characterize the dynamics of the protein. This technique reveals new dynamical motifs, which quantify proposed hierarchical structures of conformational substates for proteins and provide a means by which configuration space sampling efficiency may be probed. The SVD clearly shows that solvent effects facilitate transitions between global conformational substates for myoglobin molecular dynamics simulations. Lyapunov exponents calculated from the configuration space divergence of 15 trajectories agree with previous predictions for the chaotic behavior of complex protein systems. CONCLUSIONS: Configuration space projections provide invaluable information about protein motions that would be extremely difficult to obtain otherwise. While the configuration space for myoglobin is quite large, it does have structure. Our analysis of this structure shows that the protein hops between a number of distinct global conformational states, much like the local behavior observed for an individual residue.

Solution and crystal structures of a sperm whale myoglobin triple mutant that mimics the sulfide-binding hemoglobin from Lucina pectinata.

The bivalve mollusc Lucina pectinata harbors sulfide-oxidizing chemoautotrophic bacteria and expresses a monomeric hemoglobin I, HbI, with normal O2, but extraordinarily high sulfide affinity. The crystal structure of aquomet Lucina HbI has revealed an active site with three residues not commonly found in vertebrate globins: Phe(B10), Gln(E7), and Phe(E11) (Rizzi, M., Wittenberg, J. B., Coda, A., Fasano, M., Ascenzi, P., and Bolognesi, M. (1994) J. Mol. Biol. 244, 86-89). Engineering these three residues into sperm whale myoglobin results in a triple mutant with approximately 700-fold higher sulfide affinity than for wild-type. The single crystal x-ray structure of the aquomet derivative of the myoglobin triple mutant and the solution 1H NMR active site structures of the cyanomet derivatives of both the myoglobin mutant and Lucina HbI have been determined to examine further the structural origin of their unusually high sulfide affinities. The major differences in the distal pocket is that in the aquomet form the carbonyl of Gln64(E7) serves as a H-bond acceptor, whereas in the cyanomet form the amido group acts as H-bond donor to the bound ligand. Phe68(E11) is rotated approximately 90 degrees about chi2 and located approximately 1-2 A closer to the iron atom in the myoglobin triple mutant relative to its conformation in Lucina HbI. The change in orientation potentially eliminates the stabilizing interaction with sulfide and, together with the decrease in size of the distal pocket, accounts for the 7-fold lower sulfide affinity of the myoglobin mutant compared with that of Lucina HbI.

Nitric oxide myoglobin: crystal structure and analysis of ligand geometry.

The structure of the ferrous nitric oxide form of native sperm whale myoglobin has been determined by X-ray crystallography to 1.7 angstroms resolution. The nitric oxide ligand is bent with respect to the heme plane: the Fe-N-O angle is 112 degrees. This angle is smaller than those observed in model compounds and in lupin leghemoglobin. The exact angle appears to be influenced by the strength of the proximal bond and hydrogen bonding interactions between the distal histidine and the bound ligand. Specifically, the N(epsilon) atom of histidine64 is located 2.8 angstroms away from the nitrogen atom of the bound ligand, implying electrostatic stabilization of the FeNO complex. This interpretation is supported by mutagenesis studies. When histidine64 is replaced with apolar amino acids, the rate of nitric oxide dissociation from myoglobin increases tenfold.