Crystal structures of Fms1 and its complex with spermine reveal substrate specificity

Fms1 is a rate-limiting enzyme for the biosynthesis of pantothenic acid in yeast. Fms1 has polyamine oxidase (PAO) activity, which converts spermine into spermidine and 3-aminopropanal. The 3-aminopropanal is further oxidized to produce beta-alanine, which is necessary for the biosynthesis of pantothenic acid. The crystal structures of Fms1 and its complex with the substrate spermine have been determined using the single-wavelength anomalous diffraction (SAD) phasing method. Fms1 consists of an FAD-binding domain, with Rossmann fold topology, and a substrate-binding domain. The active site is a tunnel located at the interface of the two domains. The substrate spermine binds to the active site mainly via hydrogen bonds and hydrophobic interactions. In the complex, C11 but not C9 of spermine is close enough to the catalytic site (N5 of FAD) to be oxidized. Therefore, the products are spermidine and 3-aminopropanal, rather than 3-(aminopropyl) 4-aminobutyraldehyde and 1,3-diaminoprone.     

A structural account of substrate and inhibitor specificity differences between two naphthol reductases

Two short chain dehydrogenase/reductases mediate naphthol reduction reactions in fungal melanin biosynthesis. An X-ray structure of 1,3,6,8-tetrahydroxynaphthalene reductase (4HNR) complexed with NADPH and pyroquilon was determined for examining substrate and inhibitor specificities that differ from those of 1,3,8-trihydroxynaphthalene reductase (3HNR). The 1.5 A resolution structure allows for comparisons with the 1.7 A resolution structure of 3HNR complexed with the same ligands. The sequences of the two proteins are 46% identical, and they have the same fold. The 30-fold lower affinity of the 4HNR-NADPH complex for pyroquilon (a commercial fungicide that targets 3HNR) in comparison to that of the 3HNR-NADPH complex can be explained by unfavorable interactions between the anionic carboxyl group of the C-terminal Ile282 of 4HNR and CH and CH(2) groups of the inhibitor that are countered by favorable inhibitor interactions with 3HNR. 1,3,8-Trihydroxynaphthalene (3HN) and 1,3,6,8-tetrahydroxynaphthalene (4HN) were modeled onto the cyclic structure of pyroquilon in the 4HNR-NADPH-pyroquilon complex to examine the 300-fold preference of the enzyme for 4HN over 3HN. The models suggest that the C-terminal carboxyl group of Ile282 has a favorable hydrogen bonding interaction with the C6 hydroxyl group of 4HN and an unfavorable interaction with the C6 CH group of 3HN. Models of 3HN and 4HN in the 3HNR active site suggest a favorable interaction of the sulfur atom of the C-terminal Met283 with the C6 CH group of 3HN and an unfavorable one with the C6 hydroxyl group of 4HN, accounting for the 4-fold difference in substrate specificities. Thus, the C-terminal residues of the two naphthol reductase are determinants of inhibitor and substrate specificities.     

High resolution crystal structures and molecular dynamics studies reveal substrate binding in the porin Omp32

The porin Omp32 is the major outer membrane protein of the bacterium Delftia acidovorans. The crystal structures of the strongly anion-selective porin alone and in complex with the substrate malate were solved at 1.5 and 1.45 A resolution, respectively, and revealed a malate-binding motif adjacent to the channel constriction zone. Binding is mediated by interaction with a cluster of two arginine residues and two threonines. This binding site is specific for Omp32 and reflects the physiological adaptation of the organism to organic acids. Structural studies are combined with a 7-ns unbiased molecular dynamics simulation of the trimeric channel in a model membrane. Molecular dynamics trajectories show how malate ions are efficiently captured from the surrounding bulk solution by the electrostatic potential of the channel, translocated to the binding site region, and immobilized in the constriction zone. In accordance with these results, conductance measurements with Omp32 inserted in planar lipid membranes revealed binding of malate. The anion-selective channel Omp32 is the first reported example of a porin with a 16-stranded beta-barrel and proven substrate specificity. This finding suggests a new view on the correlation of porin structure with substrate binding in specific channels.     

Crystal structures of mouse class II alcohol dehydrogenase reveal determinants of substrate specificity and catalytic efficiency

The structure of mouse class II alcohol dehydrogenase (ADH2) has been determined in a binary complex with the coenzyme NADH and in a ternary complex with both NADH and the inhibitor N-cyclohexylformamide to 2.2 A and 2.1 A resolution, respectively. The ADH2 dimer is asymmetric in the crystal with different orientations of the catalytic domains relative to the coenzyme-binding domains in the two subunits, resulting in a slightly different closure of the active-site cleft. Both conformations are about half way between the open apo structure and the closed holo structure of horse ADH1, thus resembling that of ADH3. The semi-open conformation and structural differences around the active-site cleft contribute to a substantially different substrate-binding pocket architecture as compared to other classes of alcohol dehydrogenase, and provide the structural basis for recognition and selectivity of alcohols and quinones. The active-site cleft is more voluminous than that of ADH1 but not as open and funnel-shaped as that of ADH3. The loop with residues 296-301 from the coenzyme-binding domain is short, thus opening up the pocket towards the coenzyme. On the opposite side, the loop with residues 114-121 stretches out over the inter-domain cleft. A cavity is formed below this loop and adds an appendix to the substrate-binding pocket. Asp301 is positioned at the entrance of the pocket and may control the binding of omega-hydroxy fatty acids, which act as inhibitors rather than substrates. Mouse ADH2 is known as an inefficient ADH with a slow hydrogen-transfer step. By replacing Pro47 with His, the alcohol dehydrogenase activity is restored. Here, the structure of this P47H mutant was determined in complex with NADH to 2.5 A resolution. His47 is suitably positioned to act as a catalytic base in the deprotonation of the substrate. Moreover, in the more closed subunit, the coenzyme is allowed a position closer to the catalytic zinc. This is consistent with hydrogen transfer from an alcoholate intermediate where the Pro/His replacement focuses on the function of the enzyme.     

The structures of human dihydroorotate dehydrogenase with and without inhibitor reveal conformational flexibility in the inhibitor and substrate binding sites

Inhibitors of dihydroorotate dehydrogenase (DHODH) have been suggested for the treatment of rheumatoid arthritis, psoriasis, autoimmune diseases, Plasmodium, and bacterial and fungal infections. Here we present the structures of N-terminally truncated (residues Met30-Arg396) DHODH in complex with two inhibitors: a brequinar analogue (6) and a novel inhibitor (a fenamic acid derivative) (7), as well as the first structure of the enzyme to be characterized without any bound inhibitor. It is shown that 7 uses the "standard" brequinar binding mode and, in addition, interacts with Tyr356, a residue conserved in most class 2 DHODH proteins. Compared to the inhibitor-free structure, some of the amino acid side chains in the tunnel in which brequinar binds and which was suggested to be the binding site of ubiquinone undergo changes in conformation upon inhibitor binding. Using our data, the loop regions of residues Leu68-Arg72 and Asn212-Leu224, which were disordered in previously studied human DHODH structures, could be built into the electron density. The first of these loops, which is located at the entrance to the inhibitor-binding pocket, shows different conformations in the three structures, suggesting that it may interfere with inhibitor/cofactor binding. The second loop has been suggested to control the access of dihydroorotate to the active site of the enzyme and may be an important player in the enzymatic reaction. These observations provide new insights into the dynamic features of the DHODH reaction and suggest new approaches to the design of inhibitors against DHODH.     

Crystal structures of human NUDT5 reveal insights into the structural basis of the substrate specificity

Human NUDT5 (hNUDT5) is an ADP-ribose pyrophosphatase (ADPRase) belonging to the Nudix hydrolase superfamily. It presumably plays important roles in controlling the intracellular level of ADP-ribose (ADPR) to prevent non-enzymatic ADP-ribosylation by hydrolyzing ADPR to AMP and ribose 5'-phosphate. We report here the crystal structures of hNUDT5 in apo form, in complex with ADPR, and in complex with AMP with bound Mg2+. hNUDT5 forms a homodimer with substantial domain swapping and assumes a structure more similar to Escherichia coli ADPRase ORF209 than human ADPRase NUDT9. The adenine moiety of the substrates is specifically recognized by the enzyme via hydrogen-bonding interactions between N1 and N6 of the base and Glu47 of one subunit, and between N7 of the base and Arg51 of the other subunit, providing the molecular basis for the high selectivity of hNUDT5 for ADP-sugars over other sugar nucleotides. Structural comparisons with E. coli ADPRase ORF209 and ADPXase ORF186 indicate that the existence of an aromatic residue on loop L8 in ORF186 seems to be positively correlated with its enzymatic activity on APnA, whereas hNUDT5 and ORF209 contain no such residue and thus have low or no activities on APnA.     

Infliximab crystal structures reveal insights into self-association

Aggregation and self-association in protein-based biotherapeutics are critical quality attributes that are tightly controlled by the manufacturing process. Aggregates have the potential to elicit immune reactions, including neutralizing anti-drug antibodies, which can diminish the drug's efficacy upon subsequent dosing. The structural basis of reversible self-association, a form of non-covalent aggregation in the native state, is only beginning to emerge for many biologics and is often unique to a given molecule. In the present study, crystal structures of the infliximab (Remicade) Fc and Fab domains were determined. The Fab domain structures are the first to be reported in the absence of the antigen (i.e., tumor necrosis factor), and are consistent with a mostly rigid complementarity-determining region loop structure and rotational flexibility between variable and constant regions. A potential self-association interface is conserved in two distinct crystal forms of the Fab domain, and solution studies further demonstrate that reversible self-association of infliximab is mediated by the Fab domain. The crystal structures and corresponding solution studies help rationalize the propensity for infliximab to self-associate and provide insights for the design of improved control strategies in biotherapeutics development.     

Bioinformatic analysis of functional differences between the immunoproteasome and the constitutive proteasome

Intracellular proteins are degraded largely by proteasomes. In cells stimulated with gamma interferon , the active proteasome subunits are replaced by "immuno" subunits that form immunoproteasomes. Phylogenetic analysis of the immunosubunits has revealed that they evolve faster than their constitutive counterparts. This suggests that the immunoproteasome has evolved a function that differs from that of the constitutive proteasome. Accumulating experimental degradation data demonstrate, indeed, that the specificity of the immunoproteasome and the constitutive proteasome differs. However, it has not yet been quantified how different the specificity of two forms of the proteasome are. The main question, which still lacks direct evidence, is whether the immunoproteasome generates more MHC ligands. Here we use bioinformatics tools to quantify these differences and show that the immunoproteasome is a more specific enzyme than the constitutive proteasome. Additionally, we predict the degradation of pathogen proteomes and find that the immunoproteasome generates peptides that are better ligands for MHC binding than peptides generated by the constitutive proteasome. Thus, our analysis provides evidence that the immunoproteasome has co-evolved with the major histocompatibility complex to optimize antigen presentation in vertebrate cells.     

Crystal structures of human CtBP in complex with substrate MTOB reveal active site features useful for inhibitor design

The oncogenic corepressors C-terminal Binding Protein (CtBP) 1 and 2 harbor regulatory d-isomer specific 2-hydroxyacid dehydrogenase (d2-HDH) domains. 4-Methylthio 2-oxobutyric acid (MTOB) exhibits substrate inhibition and can interfere with CtBP oncogenic activity in cell culture and mice. Crystal structures of human CtBP1 and CtBP2 in complex with MTOB and NAD⁺ revealed two key features: a conserved tryptophan that likely contributes to substrate specificity and a hydrophilic cavity that links MTOB with an NAD⁺ phosphate. Neither feature is present in other d2-HDH enzymes. These structures thus offer key opportunities for the development of highly selective anti-neoplastic CtBP inhibitors.     

The crystal structures of dihydropyrimidinases reaffirm the close relationship between cyclic amidohydrolases and explain their substrate specificity

In eukaryotes, dihydropyrimidinase catalyzes the second step of the reductive pyrimidine degradation, the reversible hydrolytic ring opening of dihydropyrimidines. Here we describe the three-dimensional structures of dihydropyrimidinase from two eukaryotes, the yeast Saccharomyces kluyveri and the slime mold Dictyostelium discoideum, determined and refined to 2.4 and 2.05 angstroms, respectively. Both enzymes have a (beta/alpha)8-barrel structural core embedding the catalytic di-zinc center, which is accompanied by a smaller beta-sandwich domain. Despite loop-forming insertions in the sequence of the yeast enzyme, the overall structures and architectures of the active sites of the dihydropyrimidinases are strikingly similar to each other, as well as to those of hydantoinases, dihydroorotases, and other members of the amidohydrolase superfamily of enzymes. However, formation of the physiologically relevant tetramer shows subtle but nonetheless significant differences. The extension of one of the sheets of the beta-sandwich domain across a subunit-subunit interface in yeast dihydropyrimidinase underlines its closer evolutionary relationship to hydantoinases, whereas the slime mold enzyme shows higher similarity to the noncatalytic collapsin-response mediator proteins involved in neuron development. Catalysis is expected to follow a dihydroorotase-like mechanism but in the opposite direction and with a different substrate. Complexes with dihydrouracil and N-carbamyl-beta-alanine obtained for the yeast dihydropyrimidinase reveal the mode of substrate and product binding and allow conclusions about what determines substrate specificity, stereoselectivity, and the reaction direction among cyclic amidohydrolases.     

Bradykinin analogues with beta-amino acid substitutions reveal subtle differences in substrate specificity between the endopeptidases EC 3.4.24.15 and EC 3.4.24.16.

The closely related zinc metalloendopeptidases EC 3.4.24.15 (EP24.15) and EC 3.4.24.16 (EP24.16) cleave many common substrates, including bradykinin (BK). As such, there are few substrate-based inhibitors which are sufficiently selective to distinguish their activities. We have used BK analogues with either alanine or beta-amino acid (containing an additional carbon within the peptide backbone) substitutions to elucidate subtle differences in substrate specificity between the enzymes. The cleavage of the analogues by recombinant EP24.15 and EP24.16 was assessed, as well as their ability to inhibit the two enzymes. Alanine-substituted analogues were generally better substrates than BK itself, although differences between the peptidases were observed. Similarly, substitution of the four N-terminal residues with beta-glycine enhanced cleavage in some cases, but not others. beta-Glycine substitution at or near the scissile bond (Phe5-Ser6) completely prevented cleavage by either enzyme: interestingly, these analogues still acted as inhibitors, although with very different affinities for the two enzymes. Also of interest, beta-Gly8-BK was neither a substrate nor an inhibitor of EP24.15, yet could still interact with EP24.16. Finally, while both enzymes could be similarly inhibited by the D-stereoisomer of beta-C3-Phe5-BK (IC50 approximately 20 microM, compared to 8 microM for BK), EP24.16 was relatively insensitive to the L-isomer (IC50 12 approximately microM for EP24.15, >40 microM for EP24.16). These studies indicate subtle differences in substrate specificity between EP24.15 and EP24.16, and suggest that beta-amino acid analogues may be useful as templates for the design of selective inhibitors.     

X-ray crystal structures reveal two activated states for RhoC

RhoC is a member of the Rho family of Ras-related (small) GTPases and shares significant sequence similarity with the founding member of the family, RhoA. However, despite their similarity, RhoA and RhoC exhibit different binding preferences for effector proteins and give rise to distinct cellular outcomes, with RhoC being directly implicated in the invasiveness of cancer cells and the development of metastasis. While the structural analyses of the signaling-active and -inactive states of RhoA have been performed, thus far, the work on RhoC has been limited to an X-ray structure for its complex with the effector protein, mDia1 (for mammalian Diaphanous 1). Therefore, in order to gain insights into the molecular basis for RhoC activation, as well as clues regarding how it mediates distinct cellular responses relative to those induced by RhoA, we have undertaken a structural comparison of RhoC in its GDP-bound (signaling-inactive) state versus its GTP-bound (signaling-active) state as induced by the nonhydrolyzable GTP analogues, guanosine 5'-(beta,gamma-iminotriphosphate) (GppNHp) and guanosine 5'-(3-O-thiotriphosphate) (GTPgammaS). Interestingly, we find that GppNHp-bound RhoC only shows differences in its switch II domain, relative to GDP-bound RhoC, whereas GTPgammaS-bound RhoC exhibits differences in both its switch I and switch II domains. Given that each of the nonhydrolyzable GTP analogues is able to promote the binding of RhoC to effector proteins, these results suggest that RhoC can undergo at least two conformational transitions during its conversion from a signaling-inactive to a signaling-active state, similar to what has recently been proposed for the H-Ras and M-Ras proteins. In contrast, the available X-ray structures for RhoA suggest that it undergoes only a single conformational transition to a signaling-active state. These and other differences regarding the changes in the switch domains accompanying the activation of RhoA and RhoC provide plausible explanations for the functional specificity exhibited by the two GTPases.     

Insights into channel architecture and substrate specificity from crystal structures of two macrocycle-forming thioesterases of modular polyketide synthases

Modular polyketide synthases (PKSs) synthesize the polyketide cores of pharmacologically important natural products such as erythromycin and picromycin. Understanding PKSs at high resolution could present new opportunities for chemoenzymatic synthesis of complex molecules. The crystal structures of macrocycle-forming thioesterase (TE) domains from the picromycin synthase (PICS) and 6-deoxyerythronolide B synthase (DEBS) were determined to 1.8-3.0 A with an R(crys) of 19.2-24.4%, including three structures of PICS TE (crystallized at pH 7.6, 8.0, and 8.4) and a second crystal form of DEBS TE. As predicted by the previous work on DEBS TE [Tsai, S. C., et al. (2001) Proc. Natl. Acad. Sci. U.S.A. 98, 14808-14813], PICS TE contains an open substrate channel and a hydrophobic dimer interface. Notwithstanding their similarity, the dimer interfaces and substrate channels of DEBS TE and PICS TE reveal key differences. The structural basis for the divergent substrate specificities of DEBS TE and PICS TE is analyzed. The size of the substrate channel increases with increasing pH, presumably due to electrostatic repulsion in the channel at elevated pH. Together, these structures support previous predictions that macrocycle-forming thioesterases from PKSs share the same protein fold, an open substrate channel, a similar catalytic mechanism, and a hydrophobic dimer interface. They also provide a basis for the design of enzymes capable of catalyzing regioselective macrocyclization of natural or synthetic substrates. A series of high-resolution snapshots of a protein channel at different pHs is presented alongside analysis of channel residues, which could help in the redesign of the protein channel architecture.     

ADP-ribosyl cyclase; crystal structures reveal a covalent intermediate

ADP-ribosyl cyclase catalyzes the elimination of nicotinamide from NAD and cyclization to cADPR, a known second messenger in cellular calcium signaling pathways. We have determined to 2.0 A resolution the structure of Aplysia cyclase with ribose-5-phosphate bound covalently at C3' and with the base exchange substrate (BES), pyridylcarbinol, bound to the active site. In addition, further refinement at 2.4 A resolution of the structure of nicotinamide-bound cyclase, which was previously reported, reveals that ribose-5-phosphate is also covalently bound in this structure, and a second nicotinamide site was identified. The structures of native and mutant Glu179Ala cyclase were also solved to 1.7 and 2.0 A respectively. It is proposed that the second nicotinamide site serves to promote cyclization by clearing the active site of the nicotinamide byproduct. Moreover, a ribosylation mechanism can be proposed in which the cyclization reaction proceeds through a covalently bound intermediate.     

Activity probe for in vivo profiling of the specificity of proteasome inhibitor bortezomib

Proteasome inhibitors, such as the dipeptide boronic acid bortezomib, are emerging as important tools in the treatment of the fatal hematologic malignancy multiple myeloma. Despite the recent US Food and Drug Administration approval of bortezomib (PS341, Velcade) for the treatment of refractory multiple myeloma, many of the basic pharmacologic parameters of bortezomib and its mode of action on myeloma cells remain to be determined. We describe the synthesis and use of a cell-permeant active site-directed probe, which allows profiling of proteasomal activities in living cells. When we compared proteasome activity patterns in cultured cells and crude cell extracts with this probe, we observed substantial differences, stressing the importance for bioassays compatible with live cells to ensure accuracy of such measurements. Using this probe, we investigated the in vivo subunit specificities of bortezomib and another inhibitor, MG132.     

The crystal and solution structures of glyceraldehyde-3-phosphate dehydrogenase reveal different quaternary structures

The presence of an isoform of glyceraldehyde-3-phosphate dehydrogenase (kmGAPDH1p) associated with the cell wall of a flocculent strain of Kluyveromyces marxianus was the first report of a non-cytosolic localization of a glycolytic enzyme, but the mechanism by which the protein is transported to the cell surface is not known. To identify structural features that could account for the multiple localizations of the protein, the three-dimensional structure of kmGAPDH1p was determined by x-ray crystallography and small angle x-ray scattering. The x-ray crystallographic structure of kmGAPDH1p revealed a dimer, although all GAPDH homologs studied thus far have a tetrameric structure with 222 symmetry. Interestingly, the structure of kmGAPDH1p in solution revealed a tetramer with a 70 degrees tilt angle between the dimers. Moreover, the separation between the centers of the dimers composing the kmGAPDH1p tetramer diminished from 34 to 30 A upon NAD(+) binding, this latter value being similar to the observed in the crystallographic models of GAPDH homologs. The less compact structure of apo-kmGAPDH1p could already be the first image of the transition intermediate between the tetramer observed in solution and the dimeric form found in the crystal structure, which we postulate to exist in vivo because of the protein's multiple subcellular localizations in this yeast species.     

A comparison between the binding modes of a substrate and inhibitor to papain as observed in complex crystal structures

On the basis of the crystal structures of papain complexed with the substrate analogue benzyloxycarbonyl-L-phenylalanyl-L-alanine chloromethyl-ketone (Drenth, J., Kalk, K.H., and Swen, H.M. (1976) Biochemistry 15, 3731-3738) and with the inhibitor E-64-c, the binding modes were compared at the atomic level to clarify the functional difference between the substrate and inhibitor. Irrespective of the reverse chemical bonding in the peptide bonds, both the molecules are located at the S subsites of papain with similar interactions. However, the inhibitory activity of E-64-c is characterized by the stereochemical function of a carboxyoxirane ring and the tight binding of the isopentylaminoleucyl side chain to the S subsites.     

Structural basis for differences in substrate specificity of aspartate aminotransferase isoenzymes

X-Ray structural data concerning the substrate binding site of cytosolic chicken aspartate aminotransferase (AspAT) are reported. The structure of the complex of AspAT with the substrate-like inhibitor maleate has been refined at 2.2 A resolution. The lengths of hydrogen bonds between a bound molecule of maleate and side chains of amino acid residues in the active site are presented as well as other interatomic distances in the substrate binding site. The data obtained for the cytosolic AspAT have been compared with those for the mitochondrial chicken AspAT. It has been inferred that differences in substrate specificity of the AspAT isoenzymes are determined by interactions involving amino acid residues which are situated in the immediate vicinity of the active site and influence ionization or orientation of functional groups interacting with substrate. An explanation is suggested for different rates of transamination of aromatic amino acids in the active sites of the cytosolic and mitochondrial isoenzymes.