D3R grand challenge 2015: Evaluation of protein-ligand pose and affinity predictions.

The Drug Design Data Resource (D3R) ran Grand Challenge 2015 between September 2015 and February 2016. Two targets served as the framework to test community docking and scoring methods: (1) HSP90, donated by AbbVie and the Community Structure Activity Resource (CSAR), and (2) MAP4K4, donated by Genentech. The challenges for both target datasets were conducted in two stages, with the first stage testing pose predictions and the capacity to rank compounds by affinity with minimal structural data; and the second stage testing methods for ranking compounds with knowledge of at least a subset of the ligand-protein poses. An additional sub-challenge provided small groups of chemically similar HSP90 compounds amenable to alchemical calculations of relative binding free energy. Unlike previous blinded Challenges, we did not provide cognate receptors or receptors prepared with hydrogens and likewise did not require a specified crystal structure to be used for pose or affinity prediction in Stage 1. Given the freedom to select from over 200 crystal structures of HSP90 in the PDB, participants employed workflows that tested not only core docking and scoring technologies, but also methods for addressing water-mediated ligand-protein interactions, binding pocket flexibility, and the optimal selection of protein structures for use in docking calculations. Nearly 40 participating groups submitted over 350 prediction sets for Grand Challenge 2015. This overview describes the datasets and the organization of the challenge components, summarizes the results across all submitted predictions, and considers broad conclusions that may be drawn from this collaborative community endeavor.

CSAR data set release 2012: ligands, affinities, complexes, and docking decoys.

A major goal in drug design is the improvement of computational methods for docking and scoring. The Community Structure Activity Resource (CSAR) has collected several data sets from industry and added in-house data sets that may be used for this purpose ( www.csardock.org). CSAR has currently obtained data from Abbott, GlaxoSmithKline, and Vertex and is working on obtaining data from several others. Combined with our in-house projects, we are providing a data set consisting of 6 protein targets, 647 compounds with biological affinities, and 82 crystal structures. Multiple congeneric series are available for several targets with a few representative crystal structures of each of the series. These series generally contain a few inactive compounds, usually not available in the literature, to provide an upper bound to the affinity range. The affinity ranges are typically 3-4 orders of magnitude per series. For our in-house projects, we have had compounds synthesized for biological testing. Affinities were measured by Thermofluor, Octet RED, and isothermal titration calorimetry for the most soluble. This allows the direct comparison of the biological affinities for those compounds, providing a measure of the variance in the experimental affinity. It appears that there can be considerable variance in the absolute value of the affinity, making the prediction of the absolute value ill-defined. However, the relative rankings within the methods are much better, and this fits with the observation that predicting relative ranking is a more tractable problem computationally. For those in-house compounds, we also have measured the following physical properties: logD, logP, thermodynamic solubility, and pK(a). This data set also provides a substantial decoy set for each target consisting of diverse conformations covering the entire active site for all of the 58 CSAR-quality crystal structures. The CSAR data sets (CSAR-NRC HiQ and the 2012 release) provide substantial, publically available, curated data sets for use in parametrizing and validating docking and scoring methods.

Sixty years from discovery to solution: crystal structure of bovine liver catalase form III.

The crystallization and structural characterization of bovine liver catalase (BLC) has been intensively studied for decades. Forms I and II of BLC have previously been fully characterized using single-crystal X-ray diffraction. Form III has previously been analyzed by electron microscopy, but owing to the thinness of this crystal form an X-ray crystal structure had not been determined. Here, the crystal structure of form III of BLC is presented in space group P2(1)2(1)2(1), with unit-cell parameters a = 68.7, b = 173.7, c = 186.3 Å. The asymmetric unit is composed of the biological tetramer, which is packed in a tetrahedron motif with three other BLC tetramers. This higher resolution structure has allowed an assessment of the previously published electron-microscopy studies.

Structure of grouper iridovirus purine nucleoside phosphorylase.

Purine nucleoside phosphorylase (PNP) catalyzes the reversible phosphorolysis of purine ribonucleosides to the corresponding free bases and ribose 1-phosphate. The crystal structure of grouper iridovirus PNP (givPNP), corresponding to the first PNP gene to be found in a virus, was determined at 2.4 A resolution. The crystals belonged to space group R3, with unit-cell parameters a = 193.0, c = 105.6 A, and contained four protomers per asymmetric unit. The overall structure of givPNP shows high similarity to mammalian PNPs, having an alpha/beta structure with a nine-stranded mixed beta-barrel flanked by a total of nine alpha-helices. The predicted phosphate-binding and ribose-binding sites are occupied by a phosphate ion and a Tris molecule, respectively. The geometrical arrangement and hydrogen-bonding patterns of the phosphate-binding site are similar to those found in the human and bovine PNP structures. The enzymatic activity assay of givPNP on various substrates revealed that givPNP can only accept 6-oxopurine nucleosides as substrates, which is also suggested by its amino-acid composition and active-site architecture. All these results suggest that givPNP is a homologue of mammalian PNPs in terms of amino-acid sequence, molecular mass, substrate specificity and overall structure, as well as in the composition of the active site.

A novel function for the N-terminal nucleophile hydrolase fold demonstrated by the structure of an archaeal inosine monophosphate cyclohydrolase.

Inosine 5'-monophosphate (IMP) cyclohydrolase catalyzes the cyclization of 5-formaminoimidazole-4-carboxamide ribonucleotide (FAICAR) to IMP in the final step of de novo purine biosynthesis. Two major types of this enzyme have been discovered to date: PurH in Bacteria and Eukarya and PurO in Archaea. The structure of the MTH1020 gene product from Methanothermobacter thermoautotrophicus was previously solved without functional annotation but shows high amino acid sequence similarity to other PurOs. We determined the crystal structure of the MTH1020 gene product in complex with either IMP or 5-aminoimidazole-4-carboxamide ribonucleotide (AICAR) at 2.0 and 2.6 A resolution, respectively. On the basis of the sequence analysis, ligand-bound structures, and biochemical data, MTH1020 is confirmed as an archaeal IMP cyclohydrolase, thus designated as MthPurO. MthPurO has a four-layered alphabeta betaalpha core structure, showing an N-terminal nucleophile (NTN) hydrolase fold. The active site is located at the deep pocket between two central beta-sheets and contains residues strictly conserved within PurOs. Comparisons of the two types of IMP cyclohydrolase, PurO and PurH, revealed that there are no similarities in sequence, structure, or the active site architecture, suggesting that they are evolutionarily not related to each other. The MjR31K mutant of PurO from Methanocaldococcus jannaschii showed 76% decreased activity and the MjE102Q mutation completely abolished enzymatic activity, suggesting that these highly conserved residues play critical roles in catalysis. Interestingly, green fluorescent protein (GFP), which has no structural homology to either PurO or PurH but catalyzes a similar intramolecular cyclohydrolase reaction required for chromophore maturation, utilizes Arg96 and Glu222 in a mechanism analogous to that of PurO.

Structural insights into the mechanism of the PLP synthase holoenzyme from Thermotoga maritima.

Pyridoxal 5'-phosphate (PLP) is the biologically active form of vitamin B6 and is an important cofactor for several of the enzymes involved in the metabolism of amine-containing natural products such as amino acids and amino sugars. The PLP synthase holoenzyme consists of two subunits: YaaD catalyzes the condensation of ribulose 5-phosphate, glyceraldehyde-3-phosphate, and ammonia, and YaaE catalyzes the production of ammonia from glutamine. Here we describe the structure of the PLP synthase complex (YaaD-YaaE) from Thermotoga maritima at 2.9 A resolution. This complex consists of a core of 12 YaaD monomers with 12 noninteracting YaaE monomers attached to the core. Compared with the previously published structure of PdxS (a YaaD ortholog in Geobacillus stearothermophilus), the N-terminus (1-18), which includes helix alpha0, the beta2-alpha2 loop (46-56), which includes new helix alpha2a, and the C-terminus (270-280) of YaaD are ordered in the complex but disordered in PdxS. A ribulose 5-phosphate is bound to YaaD via an imine with Lys82. Previous studies have demonstrated a similar imine at Lys149 and not at Lys81 (equivalent to Lys150 and Lys82 in T. maritima) for the Bacillus subtilis enzyme suggesting the possibility that two separate sites on YaaD are involved in PLP formation. A phosphate from the crystallization solution is found bound to YaaD and also serves as a marker for a possible second active site. An ammonia channel that connects the active site of YaaE with the ribulose 5-phosphate binding site was identified. This channel is similar to one found in imidazole glycerol phosphate synthase; however, when the beta-barrels of the two complexes are superimposed, the glutaminase domains are rotated by about 180 degrees with respect to each other.

Structural analysis of threonine 342 mutants of soybean beta-amylase: role of a conformational change of the inner loop in the catalytic mechanism.

Two different conformations of the inner loop (residues 340-346) have been found in the soybean beta-amylase structures. In the "product form", the Thr 342 residue creates hydrogen bonds with Glu 186 (catalytic acid) and with the glucose residues at subsites -1 and +1, whereas most of those interactions are lost in the "apo form". To elucidate the relationship between the structural states of the inner loop and the catalytic mechanism, Thr 342 was mutated to Val, Ser, and Ala, respectively, and their crystal structures complexed with maltose were determined together with that of the apo enzyme at 1.27-1.64 A resolutions. The k(cat) values of the T342V, T342S, and T342A mutants decreased by 13-, 360-, and 1700-fold, respectively, compared to that of the wild-type enzyme. Whereas the inner loops in the wild-type/maltose and T342V/maltose complexes adopted the product form, those of the T342S/maltose and T342A/maltose complexes showed the apo form. Structural analyses suggested that the side chain of Thr 342 in product form plays an important role in distorting the sugar ring at subsite -1, stabilizing the deprotonated form of Glu 186, and grasping the glucose residue of the remaining substrate at subsite +1. The third hypothesis was proved by the fact that T342V hydrolyzes maltoheptaose following only multichain attack in contrast to multiple attack of the wild-type enzyme.

Crystallization and preliminary crystallographic analysis of endo-1,3-beta-glucanase from Arthrobacter sp.

Endo-1,3-beta-glucanases hydrolyze internal 1,3-beta-glucosyl linkages. The endo-1,3-beta-glucanase from Arthrobacter sp. was crystallized by the hanging-drop vapour-diffusion method. The crystals belonged to space group P4(1), with unit-cell parameters a = 71.31, c = 60.07 A, and contained one molecule per asymmetric unit. The Matthews coefficient (VM) and the solvent content were 2.35 A3 Da(-1) and 47.63%, respectively. Diffraction data were collected to a resolution of 1.66 A at SPring-8 using a MAR CCD area detector and gave a data set with an overall Rmerge of 5.4% and a completeness of 99.4%.

Crystallization and preliminary crystallographic analysis of monodehydroascorbate radical reductase from cucumber.

Monodehydroascorbate (MDA) radical reductase (EC 1.6.5.4) is an FAD enzyme that catalyzes the univalent reduction of MDA radical to ascorbate using NAD(P)H as an electron donor. The recombinant MDA reductase from cucumber was crystallized using polyethylene glycol 6000 as a precipitant. The crystals belong to space group P2(1), with unit-cell parameters a = 60.8, b = 138.6, c = 61.7 A, beta = 114.5 degrees, and contained two molecules per asymmetric unit. The Matthews coefficient (VM) and the solvent content are 2.46 A3 Da(-1) and 50.0%, respectively. Diffraction data were collected to a resolution of 2.4 A at 100 K using Cu Kalpha radiation with a multi-wire area detector and gave a data set with an overall Rsym of 10.0% and a completeness of 92.5%.

The roles of Glu186 and Glu380 in the catalytic reaction of soybean beta-amylase.

It has previously been suggested that the glutamic acid residues Glu186 and Glu380 of soybean beta-amylase play critical roles as a general acid and a general base catalyst, respectively. In order to confirm the roles of Glu186 and Glu380, each residue was mutated to a glutamine residue and the crystal structures of the substrate (E186Q/maltopentaose) and product (E380Q/maltose) complexes were determined at resolutions of 1.6 Angstrom and 1.9 Angstrom, respectively. Both mutant enzymes exhibited 16,000- and 37,000-fold decreased activity relative to that of the wild-type enzyme. The crystal structure of the E186Q/maltopentaose complex revealed an unambiguous five-glucose unit at subsites -2 to +3. Two maltose molecules bind on subsites -2 to -1 and +2 to +3 in the E380Q/maltose complex, whereas they bind in tandem to -2 to -1 and +1 to +2 in the wild-type/maltose complex. The conformation of the glucose residue at subsite -1 was identified as a stable (4)C(1) alpha-anomer in the E380Q/maltose complex, whereas a distorted ring conformation was observed in the wild-type/maltose complex. The side-chain movement of Gln380 to the position of a putative attacking water molecule seen in the wild-type enzyme caused the inactivation of the E380Q mutant and an altered binding pattern of maltose molecules. These results confirm the critical roles played by Glu186 in the donation of a proton to the glycosidic oxygen of the substrate, and by Glu380 in the activation of an attacking water molecule. The observed difference between the backbones of E186Q/maltopentaose and E380Q/maltose in terms of Thr342 suggests that the side-chain of Thr342 may stabilize the deprotonated form of Glu186 after the cleavage of the glycosidic bond.

Structural and enzymatic analysis of soybean beta-amylase mutants with increased pH optimum.

Comparison of the architecture around the active site of soybean beta-amylase and Bacillus cereus beta-amylase showed that the hydrogen bond networks (Glu380-(Lys295-Met51) and Glu380-Asn340-Glu178) in soybean beta-amylase around the base catalytic residue, Glu380, seem to contribute to the lower pH optimum of soybean beta-amylase. To convert the pH optimum of soybean beta-amylase (pH 5.4) to that of the bacterial type enzyme (pH 6.7), three mutants of soybean beta-amylase, M51T, E178Y, and N340T, were constructed such that the hydrogen bond networks were removed by site-directed mutagenesis. The kinetic analysis showed that the pH optimum of all mutants shifted dramatically to a neutral pH (range, from 5.4 to 6.0-6.6). The Km values of the mutants were almost the same as that of soybean beta-amylase except in the case of M51T, while the Vmax values of all mutants were low compared with that of soybean beta-amylase. The crystal structure analysis of the wild type-maltose and mutant-maltose complexes showed that the direct hydrogen bond between Glu380 and Asn340 was completely disrupted in the mutants M51T, E178Y, and N340T. In the case of M51T, the hydrogen bond between Glu380 and Lys295 was also disrupted. These results indicated that the reduced pKa value of Glu380 is stabilized by the hydrogen bond network and is responsible for the lower pH optimum of soybean beta-amylase compared with that of the bacterial beta-amylase.

Change in the crystal packing of soybean beta-amylase mutants substituted at a few surface amino acid residues.

In spite of the high similarity of amino acid sequence and three-dimensional structure between soybean beta-amylase (SBA) and sweet potato beta-amylase (SPB), their quaternary structure is quite different, being a monomer in SBA and a tetramer in SPB. Because most of the differences in amino acid sequences are found in the surface region, we tested the tetramerization of SBA by examining mutations of residues located at the surface. We designed the SBA tetramer using the SPB tetramer structure as a model and calculating the change of accessible surface area (DeltaASA) for each residue in order to select sites for the mutation. Two different mutant genes encoding SB3 (D374Y/L481R/P487D) and SB4 (K462S added to SB3), were constructed for expression in Escherichia coli and the recombinant proteins were purified. They existed as a monomer in solution, but gave completely different crystals from the native SBA. The asymmetric unit of the mutants contains four molecules, while that of native SBA contains one. The interactions of the created interfaces revealed that there were more intermolecular interactions in the SB3 than in the SB4 tetramer. The substituted charged residues on the surface are involved in interactions with adjacent molecules in a different way, forming a new crystal packing pattern.