No abstract

Thiamin-dependent enzymes play key roles in sugar metabolism, typically catalyzing the decarboxylation of ␣-keto acids and the transfer of an aldehyde or an acyl group (1-5). Examples include the E1 components in pyruvate dehydrogenase complexes (PDHc), 2 pyruvate decarboxylase, transketolase, etc. Crystallographic studies (6 -12) have elucidated many of the structural, stereochemical, and biochemical details in the mechanism of action of these enzymes and in the catalytic role of the cofactor ThDP (thiamin diphosphate, vitamin B1 diphosphate, Fig. 1, top left). Despite the enormous contributions made by these and other studies to our understanding of how such enzymes function, important details still remain obscure. There are, for example, no detailed structural data on the first ThDP-bound intermediate in the presence of any enzyme (for example, ␣-lactylthiamin diphosphate (␣-LThDP) in PDHc E1 and pyruvate decarboxylase), which is postulated to form in the currently accepted mechanism of thiamin catalysis (Fig. 1, top, third object from the right). In an effort to obtain structural information pertaining to this key intermediate, we have determined the crystal structure of PDHc E1 from Escherichia coli in complex with ␣-phosphonolactylthiamin diphosphate (PLThDP).PLThDP is the product of the reaction between ThDP and methylacetylphosphonate, with the latter being an analogue of the true substrate pyruvate and a potent inhibitor of PDHc. The complex formed with PLThDP instead of ThDP therefore mimics the structure of the enzyme-bound, reactive tetrahedral intermediate ␣-LThDP (13) in the decarboxylation step of the PDHc E1 reaction. It differs from the complex formed with the true substrate only in the replacement of the carboxylate group by a methyl phosphonate (PO 3 Me) group. However, unlike the C2␣-CO 2 bond normally cleaved in the reaction with pyruvate, the C2␣-PO 3 Me bond remains intact. The reaction is therefore trapped in a pre-CO 2 release-like state, and the structure represents a covalently bound, pre-decarboxylation reaction intermediate analogue.There have been three covalently bound reaction intermediate structures reported for ThDP-dependent enzymes (11,12,14), but they all represented the planar enamine intermediate (Fig. 1, top right object) that exists only after decarboxylation. The E1-PLThDP structure is thus the first structural example of a covalently bound, pre-decarboxylation reaction intermediate analogue in any ThDP-dependent enzyme.* This work was supported by a grant from the Veterans Affairs Merit Review Program and National Institutes of Health Grant GM-61791 (to W. F.) and by National Institutes of Health Grant GM-62330 (to F. J.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The atomic coordinates and structure factors (codes 2G25 and 2G28)

The crystal structure of the recombinant thiamin diphosphate-dependent E1 component from the Escherichia coli pyruvate dehydrogenase multienzyme complex (PDHc) has been determined at a resolution of 1.85 A. The E. coli PDHc E1 component E1p is a homodimeric enzyme and crystallizes with an intact dimer in an asymmetric unit. Each E1p subunit consists of three domains: N-terminal, middle, and C-terminal, with all having alpha/beta folds. The functional dimer contains two catalytic centers located at the interface between subunits. The ThDP cofactors are bound in the "V" conformation in clefts between the two subunits (binding involves the N-terminal and middle domains), and there is a common ThDP binding fold. The cofactors are completely buried, as only the C2 atoms are accessible from solution through the active site clefts. Significant structural differences are observed between individual domains of E1p relative to heterotetrameric multienzyme complex E1 components operating on branched chain substrates. These differences may be responsible for reported alternative E1p binding modes to E2 components within the respective complexes. This paper represents the first structural example of a functional pyruvate dehydrogenase E1p component from any species. It also provides the first representative example for the entire family of homodimeric (alpha2) E1 multienzyme complex components, and should serve as a model for this class of enzymes.

The catalytic stereoselective synthesis of compounds with chiral phosphorus centers remains an unsolved problem. State-of-the-art methods rely on resolution or stoichiometric chiral auxiliaries. Phosphoramidate prodrugs are a critical component of pronucleotide (ProTide) therapies used in the treatment of viral disease and cancer. Here we describe the development of a catalytic stereoselective method for the installation of phosphorus-stereogenic phosphoramidates to nucleosides through a dynamic stereoselective process. Detailed mechanistic studies and computational modeling led to the rational design of a multifunctional catalyst that enables stereoselectivity as high as 99:1.

Verubecestat 3 (MK-8931), a diaryl amide-substituted 3-imino-1,2,4-thiadiazinane 1,1-dioxide derivative, is a high-affinity β-site amyloid precursor protein cleaving enzyme 1 (BACE1) inhibitor currently undergoing Phase 3 clinical evaluation for the treatment of mild to moderate and prodromal Alzheimer's disease. Although not selective over the closely related aspartyl protease BACE2, verubecestat has high selectivity for BACE1 over other key aspartyl proteases, notably cathepsin D, and profoundly lowers CSF and brain Aβ levels in rats and nonhuman primates and CSF Aβ levels in humans. In this annotation, we describe the discovery of 3, including design, validation, and selected SAR around the novel iminothiadiazinane dioxide core as well as aspects of its preclinical and Phase 1 clinical characterization.

Thiamin thiazolone diphosphate (ThTDP), a potent inhibitor of the E1 component from the Escherichia coli pyruvate dehydrogenase multienzyme complex (PDHc), binds to the enzyme with greater affinity than does the cofactor thiamin diphosphate (ThDP). To identify what determines this difference, the crystal structure of the apo PDHc E1 component complex with ThTDP and Mg(2+) has been determined at 2.1 A and compared to the known structure of the native holoenzyme, PDHc E1-ThDP-Mg(2+) complex. When ThTDP replaces ThDP, reorganization occurs in the protein structure in the vicinity of the active site involving positional and conformational changes in some amino acid residues, a change in the V coenzyme conformation, addition of new hydration sites, and elimination of others. These changes culminate in an increase in the number of hydrogen bonds to the protein, explaining the greater affinity of the apoenzyme for ThTDP. The observed hydrogen bonding pattern is not an invariant feature of ThDP-dependent enzymes but rather specific to this enzyme since the extra hydrogen bonds are made with nonconserved residues. Accordingly, these sequence-related hydrogen bonding differences likewise explain the wide variation in the affinities of different thiamin-dependent enzymes for ThTDP and ThDP. The sequence of each enzyme determines its ability to form hydrogen bonds to the inhibitor or cofactor. Mechanistic roles are suggested for the aforementioned reorganization and its reversal in PDHc E1 catalysis: to promote substrate binding and product release. This study also provides additional insight into the role of water in enzyme inhibition and catalysis.

Least squares alignment of the E. coli pyruvate dehydrogenase multienzyme complex E1 subunit and yeast transketolase crystal structures indicates a general structural similarity between the two enzymes and provides a plausible location for a short-loop region in the E1 structure that was unobserved due to disorder. The residue H407, located in this region, is shown to be able to penetrate the active site. Suggested by this comparison, the H407A E1 variant was created, and H407 was shown to participate in the reductive acetylation of both an independently expressed lipoyl domain and the intact 1-lipoyl E2 subunit. While the H407A substitution only modestly affected the reaction through pyruvate decarboxylation (ca. 14% activity compared to parental E1), the overall complex has a much impaired activity, at most 0.15% compared to parental E1. Isothermal titration calorimetry measurements show that the binding of the lipoyl domain to the H407A E1 variant is much weaker than that to parental E1. At the same time, mass spectrometric measurements clearly demonstrate much impaired reductive acetylation of the independently expressed lipoyl domain and of the intact 1-lipoyl E2 by the H407A variant compared to the parental E1. A proposal is presented to explain the remarkable conservation of the three-dimensional structure at the active centers of the E. coli E1 subunit and transketolase on the basis of the parallels in the ligation-type reactions carried out and the need to protonate a very weak acid, a dithiolane sulfur atom in the former, and a carbonyl oxygen atom in the latter.

The cocrystal of celecoxib–nicotinamide has been revisited to reveal its crystal structure and unusual formation properties. Gibbs free energy and enthalpy of formation at 25 °C were determined by solubility and thermal analysis. The formation of celecoxib–nicotinamide was found to be an endothermic process and driven by entropy, a mechanism different from many reported cocrystals but in agreement with previously reported lattice energy calculation. The cocrystal is stable only above a transition temperature as predicted from thermodynamic data and confirmed experimentally. This formation mechanism provides new opportunities for cocrystal research from structural prediction to process development and enriches our understanding of the nature of cocrystallization.