Crystal structure of dihydroorotate dehydrogenase from Leishmania major

Biochimie 94 (2012) 1739e1748

Contents lists available at SciVerse ScienceDirect

Biochimie journal homepage: www.elsevier.com/locate/biochi

Research paper

Crystal structure of dihydroorotate dehydrogenase from Leishmania major Artur T. Cordeiro 1, Patricia R. Feliciano, Matheus P. Pinheiro, M. Cristina Nonato* Laboratório de Cristalografia de Proteínas, Departamento de Física e Química, Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo, Av. Café S/N, Monte Alegre, Ribeirão Preto 14040-903, S.P, Brazil

a r t i c l e i n f o

a b s t r a c t

Article history: Received 16 February 2012 Accepted 3 April 2012 Available online 21 April 2012

Dihydroorotate dehydrogenase (DHODH) is the fourth enzyme in the de novo pyrimidine biosynthetic pathway and has been exploited as the target for therapy against proliferative and parasitic diseases. In this study, we report the crystal structures of DHODH from Leishmania major, the species of Leishmania associated with zoonotic cutaneous leishmaniasis, in its apo form and in complex with orotate and fumarate molecules. Both orotate and fumarate were found to bind to the same active site and exploit similar interactions, consistent with a ping-pong mechanism described for class 1A DHODHs. Analysis of LmDHODH structures reveals that rearrangements in the conformation of the catalytic loop have direct influence on the dimeric interface. This is the first structural evidence of a relationship between the dimeric form and the catalytic mechanism. According to our analysis, the high sequence and structural similarity observed among trypanosomatid DHODH suggest that a single strategy of structure-based inhibitor design can be used to validate DHODH as a druggable target against multiple neglected tropical diseases such as Leishmaniasis, Sleeping sickness and Chagas’ diseases. Ó 2012 Elsevier Masson SAS. All rights reserved.

Keywords: Leishmania major Dihydroorotate dehydrogenase Crystal structure Orotate binding Fumarate binding Dimeric interface

1. Introduction The term “Negleted Tropical Diseases (NTDs)” refers to a large and complex group of parasitic and bacterial infectious diseases. They represent one of the most serious burdens to public health, afflicting more than 1 billion people - around one sixth of the world population - most of which live in the poorest and the most marginalized areas of developing countries [1]. Despite the apparent similarities, the control initiatives and forms of treatment of NTDs are highly heterogeneous. There are effective drugs to treat many of these diseases, available at low cost

Abbreviations: NTDs, Neglected Tropical Diseases; DHODH, dihydroorotate dehydrogenase; LmDHODH, Leishmania major dihydroorotate dehydrogenase; LmDHODH-oro, Leishmania major dihydroorotate dehydrogenase in complex with orotate; LmDHODH-fum, Leishmania major dihydroorotate dehydrogenase in complex with fumarate; TrypDHODH, trypanosomatid dihydroorotate dehydrogenase. * Corresponding author. Tel.: þ55 16 3602 4432; fax: þ55 16 3602 4880. E-mail address: [email protected] (M.C. Nonato). 1 Present address: Brazilian Biosciences National Laboratory, Brazilian Center of Research in Energy and Materials, P.O. Box 6192, Zip code 13083-970, Campinas e SP, Brazil. 0300-9084/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.biochi.2012.04.003

or donated from pharmaceutical companies, and the control depends on effective eradication programs [1e4]. Nevertheless, there is still a large group of NTDs with no current treatment or depending on old medicines, which cause serious side-effects, have poor clinical effect, and display drug resistance [5,6]. One example of this scenario is found for the Leishmaniases, a complex of vector-borne diseases caused by approximately 20 species of obligate intramacrophage parasitic protozoa from the genus Leishmania. Classified as NTDs, the Leishmaniasis affect 12 million men, women and children in 88 countries around the world with approximately 2 million new cases per year [7,8]. They are manifested in a variety of clinical forms, ranging from cutaneous to visceral leishmaniases, the latter being the most severe and which can be fatal if left untreated. The current treatment for the leishmaniases is mainly based on antimonial derivates of low specificity that result in undesirable side-effects and shows susceptibility to the development of drug resistance [9]. Although there is an urgent need for the development of better tools to prevent, control and treat Leishmaniases, funding is a critical issue. With only 0.6% of total international development assistance for health allocated to all NTDs, few public resources are available in order to manage and develop new therapeutics to treat this important disease.

1740

A.T. Cordeiro et al. / Biochimie 94 (2012) 1739e1748

Fortunately, there is increased attention from the research community in the study of the mechanisms related to virulence and pathogenesis of Leishmania which is significantly contributing to the discovery of new leads for diagnostics and drug development [10e13]. Genome sequencing projects of Leishmania major [14], Leishmania brasiliensis and Leishmania infantum [15], as well as transcriptomics [16] and proteomics projects [17e19], are producing data on gene expression profile, protein localization, three-dimensional structure, post-translational modifications, therefore providing a rich source for the understanding the evolution and biology of these pathogens. Metabolic pathways, which have traditionally provided attractive targets for drug development, are of particular interest [20e22]. As an example, the purine and pyrimidine nucleotide biosynthesis, produced either by de novo biosynthetic or salvage pathways, have already been pointed out as a promising source of drug target candidates [23e25]. In particular, evidence garnered from a variety of parasitic protozoa suggest that the selective inhibition of dihydroorotate dehydrogenase, the fourth enzyme of the de novo pyrimidine biosynthetic pathway, should be considered as an attractive therapeutic target against Leishmaniasis [26e29]. The enzyme dihydroorotate dehydrogenase (DHODH; E.C, 1.3.3.1) is a flavoenzyme that catalyzes the oxidation of (S)-dihydroorotate to orotate, the only redox reaction in the de novo pyrimidine biosynthesis pathway. On the basis of sequence similarity, the DHODHs are divided in two major classes, i.e., class 1 (A and B) and class 2 [30]. This division correlates with subcellular location of the proteins as well as their preferences for electron acceptors. Enzymes of class 1 are dimeric, located in the cytosol and can be found in Gram-positive bacteria, in the anaerobic yeast Saccharomyces cerevisiae and in all trypanosomatids [26,31e33]. They share less than 20% sequence identity with the monomeric membrane-bound enzymes of class 2, found in the mitochondria of eukaryotes [34] and in some prokaryotes such as the Gramnegative bacteria related to Escherichia coli [35e37]. The cytoplasmic enzymes utilize fumarate or NADþ to oxidize FMNH2, whereas the membrane-bound enzymes require quinones as their physiological oxidizing agent [38]. It is our interest to evaluate the potential of the L. major DHODH (LmDHODH) as a drug target candidate by initially performing a detailed functional characterization of the enzyme. In our previous work, LmDHODH has been kinetically characterized to reduce fumarate to succinate during the oxidation of FMNH2 to FMN [33]. The reaction catalyzed by LmDHODH occurs by a Ping-Pong BieBi mechanism where the product of the first half reaction (orotate) leaves the active site to allow the entrance of the substrate (fumarate) for the second half reaction. In the present work, we have moved further by focusing the structural studies of LmDHODH. Here, we describe the crystallographic structures of the recombinant LmDHODH crystallized in its apo form (LmDHODH-apo) and complexed with orotate (LmDHODH-oro) and fumarate (LmDHODH-fum) molecules. Careful analysis of all three structures reveals that both orotate and fumarate molecules occupy the same active site and exploit similar interactions. Furthermore, analysis of LmDHODH structures reveals that rearrangements in the dimeric interface have direct influence on the conformation of the catalytic loop shedding light on the structural role of the dimeric form to regulate the catalytic mechanism of class 1A DHODHs. We further recognized that the high degree of sequence and structural similarity observed among trypanosomatid DHODH suggests that ligands designed to bind LmDHODH can be exploited to validate DHODH as a therapeutic target not only for Leishmaniasis, but also against other trypanosomatid releated diseases, such as Chagas’ disease caused by Trypanosoma cruzi and Sleeping Sickness (African trypanosomiasis) caused by Trypanosoma brucei.

Table 1 Crystallographic data. Datasets

LmDHODH-apoa LmDHODH-oro LmDHODH-fum

Wavelength (Å) Space group Unit cell parameters (Å)

1.5418 P61 a ¼ 143.7 c ¼ 69.8 Resolution range (Å) 28.5e2.0 Unique reflections 58 045 Redundancy 5.6 (5.5) Data completeness (%) 100 (100) 10.5 (2.5) I/s(I) 6.4 (30.2) Rsymb (%) Molecule per asymmetric unit 2 2.9 VM (Å3. Da1) Solvent content (%) 58.0

1.427 P61 a ¼ 141.2 c ¼ 68.4 46.2e2.5 26 995 4.2 (3.9) 99.6 (99.4) 18.8 (2.5) 8.1 (48.7) 2 2.9 57.3

1.427 P61 a ¼ 140.9 c ¼ 68.3 35.2e1.9 61 101 10.6(10.4) 100 (100) 25.1 (5.8) 7.9 (37.9) 2 2.9 57.3

16.9 20.0 23.05 0.012 2.282

17.7 20.4 25.67 0.011 2.273

16.5 18.2 19.26 0.010 2.308

97.6 2.4 0.0

97.5 2.5 0.0

97.3 2.7 0.0

Structure refinement Rwork c (%) Rfree d (%) Overall B (Å2) RMSD bonds (Å) RMSD angles ( ) Ramachandran analysise Favored (%) Allowed (%) Outliers (%) a

Dataset statistics reproduced from [39]. P P Rsym ¼ jIobs - j/ Iobs, where < I > is the mean intensity of multiple observations of symmetry-related reflections. P P c Rwork ¼ h k l jjFoj-jFcjj/ jFoj, where Fo is the observed structure factor and Fc is the calculated structure factor. d Rfree is the same as Rwork except calculated using 5% of the data that are not included in any refinement calculations. e Distribution of dihedral angles in Ramachandran diagram were calculated with MolProbity program [45]. b

2. Materials and methods 2.1. Protein crystallization and X-ray diffraction data collection Recombinant LmDHODH was expressed in E. coli BL21(DE3) strain and purified to homogeneity by nickel affinity chromatography as described elsewhere [33]. LmDHODH-apo crystals have been previously obtained by the hanging and sitting drop vapordiffusion methods using lithium sulfate as precipitant agent and ammonium sulfate as an additive [39]. Crystals of LmDHODH-oro and LmDHODH-fum complexes were obtained by employing cocrystallization strategies. For LmDHODH-oro complex, purified LmDHODH protein solution was initially dialyzed against 50 mM Hepes pH 7.2 and 150 mM NaCl in presence of 1 mM orotate, and concentrated up to 9.4 mg/mL. For LmDHODH-fum complex, the purified LmDHODH protein solution has been dialyzed against 50 mM Hepes pH 7.2 and 150 mM NaCl, concentrated up to 4 mg/ mL and incubated for 3 h with 5 mM fumarate. Cocrystallization experiments were performed at 293 K by using the sitting drop vapor-diffusion method. Drops were prepared by mixing 2.0 mL of protein solution and 2.0 mL of reservoir solution containing 1.2e1.3 M lithium sulfate, 0.3e0.35 M ammonium sulfate in 0.1 M sodium citrate tribasic dihydrate pH ¼ 5.3e5.6. Prior to X-ray data collection, the crystals were transferred to a cryoprotectant solution comprising reservoir solution supplemented with 20% glycerol, and flash-cooled in a nitrogen stream. Data collection for LmDHODH-apo was performed using an inhouse rotating-anode generator [39]. For both LmDHODH-oro and LmDHODH-fum complexes, the data collections were performed at the protein crystallography beamline D03B-MX1 of the Brazilian Synchrotron Light Laboratory (LNLS) using X-rays of wavelength 1.427 Å radiation and recorded on a marCCD165 detector. All three X-ray data sets were collected at 100 K. Diffraction data were

A.T. Cordeiro et al. / Biochimie 94 (2012) 1739e1748

indexed and integrated using MOSFLM [40], and scaled using SCALA [41]. Table 1 summarizes data collections statistics.

1741

3. Results and discussion 3.1. Structure determination

2.2. Phasing and refinement The structure of LmDHODH-apo was solved by molecular replacement using the coordinates (excluding the water molecules) of the dimeric structure of Lactococcus lactis DHODH as the search model (PDB code: 1DOR [42]). Crystals of the two complexes LmDHODH-oro and LmDHODH-fum proved to be isomorphous to LmDHODH-apo crystal previously obtained in the space group P61. Thus, the coordinates of the LmDHODH-apo with waters removed were used to obtain an initial set of phases for both LmDHODH-oro and LmDHODH-fum complexes. The coordinates for the orotate and fumarate molecules and appropriate chemical restraints and topology files were prepared using Hetero-compound Information Center - Uppsala (HIC-Up) server. Waters were built using Coot [43] and validated by hand. The models were initially refined by rigid body refinement, followed by several rounds of model rebuilding using Coot and refinement with REFMAC5 [44]. Table 1 summarizes refinement statistics. The stereochemistry for LmDHODH-apo, LmDHODH-oro and LmDHODH-fum models were validated with Molprobity [45]. All pictures were generated with the PYMOL software [46]. 2.2.1. Protein data bank accession codes Coordinates and structure factors have been deposited in the PDB with the accession codes 3GYE (LmDHODH-apo), 3GZ3 (LmDHODH-oro), and 3TQ0 (LmDHODH-fum).

Crystallization, X-ray diffraction data acquisition and initial phasing of LmDHODH in its apo form (LmDHODH-apo) was previously described [39]. LmDHODH-apo crystallized in P61 space group using lithium sulfate as the precipitant agent and its crystal structure was solved by molecular replacement using the crystallographic structure of L. lactis DHODH as the search model [42]. A complete dataset to 2.0 Å resolution was used to refine the LmDHODH-apo structure, which consisted of a dimer in the asymmetric unit. Both polypeptide chains (chains A and B) comprises residues from Met1 to Leu129 and Asp143 to Leu312, and bind non-covalently to one FMN molecule. Residues from Ser130 to Tyr142 and eight C-terminal residues (Asp313 to Arg320) were excluded from the final structure due to the lack of interpretable electron density. The final LmDHODH-apo model contains two polypeptide chains, two FMN and two glycerol molecules and 358 solvent sites treated as water oxygens. Crystals of LmDHODH-oro and LmDHODH-fum complexes were obtained by cocrystallization techniques. LmDHODH-oro and LmDHODH-fum crystals displayed a high degree of isomorphism related to the LmDHODH-apo, with the same hexagonal symmetry and cell parameter deviation lower than 1%. The refined coordinates of LmDHODH-apo (excluding the solvent molecules) were used for initial rigid body refinement of LmDHODH-oro and LmDHODH-fum structures at 2.5 and 1.9 Å resolution, respectively.

Fig. 1. Stereoscopic ribbon representation of Leishmania major DHODH. Top, LmDHODH monomer where a central barrel composed of eight parallel b strands (b1- b8) illustrated in magenta is surrounded by eight a-helices (a1- a8) illustrated in yellow. Two N-terminal antiparallel b strands (ba-bb) are found at the bottom of the barrel (red) and additional secondary structural elements and loops form a protuberant subdomain present at the top of the barrel (green, aquamarine and orange). Active site loop (b4- a4 loop) is shown in salmon. Stick representation of the prosthetic FMN group is illustrated in yellow. Bottom, the native functional dimer found in the asymmetric unit. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

1742

A.T. Cordeiro et al. / Biochimie 94 (2012) 1739e1748

The polypeptide chains of LmDHODH-oro (chains A and B) are composed by the same group of residues as found for LmDHODHapo, where the segments from Ser130 to Tyr142 and Asp313 to Arg320 with no interpretable electron density were excluded from LmDHODH-oro final coordinates. The final structure contains two polypeptide chains, two FMN, two orotate and two glycerol molecules, and 430 solvent sites treated as water oxygens. The polypeptide chains of LmDHODH-fum crystallographic model comprises residues ranging from Ser0 (a plasmid-encoded residue that precedes Met1) to Leu312. The C-terminal residues ranging from Asp313 to Arg320 were also not included in the final coordinates due to the lack of electron density. Contrary to the LmDHODH-apo and LmDHODH-oro models, residues Ser130 to Tyr142 of LmDHODH-fum were observed in an ordered state with mean Bfactor value equal to 18.3 (chain A) and 19.1 (chain B) Å2. The final structure contains two polypeptide chains, two FMN, two fumarate, two glycerol molecules, one nickel atom, four sulfate ions, and 388 solvent sites treated as water oxygens. Final refinement and geometry statistics for all three structures are detailed in Table 1. 3.2. Overall structures LmDHODH monomer shows the typical a/b barrel fold of class 1A DHODH, with a central barrel composed of eight parallel b strands (b1- b8) surrounded by eight a-helices (a1-a8) (Fig. 1). Two N-terminal antiparallel b strands (ba-bb) are found at the bottom of the barrel, whereas additional secondary structural elements and loops, including a conserved four-stranded antiparallel b-sheet (bc, bd, be and bf), form a protuberant subdomain present at the top of the barrel. The prosthetic FMN group is located between the top of the barrel and the subdomain formed by these insertions. The dimeric structure of LmDHODH is very similar to all available models for class 1A DHODHs (Fig. 1): L. lactis DHODH (LlDHODH, PDB code: 2DOR [47],), T. brucei DHODH (TbDHODH, PDB code: 2B4G [27],), Leishmania donovani DHODH (LdDHODH, PDB code: 3C61, results not published), T. cruzi (Y strain) DHODH (TcDHODH_Y, PDB code: 3C3N [48],) and T. cruzi (Tulahuen strain) DHODH (TcDHODH_T, PDB codes: 2E68, 2E6A, 2E6D, 2E6F, 2DJL and 2DJX [49];) and Streptococcus mutans (StmDHODH, PDB code: 3OIX [50];). The superposition of Ca atoms between LmDHODH-apo and the structurally equivalent atoms show on average (chain A) a rmsd of 0.20 Å for LmDHODH-oro, 0.37 Å for LmDHODH-fum, 0.28 Å for LdDHODH, 0.46 Å for TbDHODH, 0.61 Å for TcDHODH_Y, 0.59 Å for TcDHODH_T, 1.01 for StmDHODH and 0.91 Å for LlDHODH; values very similar to the Ca rmsd of 0.22 Å found when we superpose

chains A and B in the asymmetric unit in LmDHODH-apo structure. Analysis of rmsd distribution for each residue along the protein sequence suggest a conserved structural pattern for the a/b barrel found in class 1 DHODH structures. The largest differences are found at the N- and C-terminal regions, from Leu129 to Tyr142, from Phe144 to Val157 and from Val202 to Pro214. Our findings are in agreement with the sequence variability and intrinsic flexibility of those regions. In particular residues found from Leu129 to Tyr142 are part of the catalytic loop, where the mobility is related to its ability for controlling the access and release of substrate from the active site [51]. 3.3. Binding mode of orotate and fumarate molecules Orotate and fumarate molecules correspond to the product of catalysis and the oxidant agent for class 1A DHODHs, respectively. Both molecules were clearly present in the active site, as observed by the final sA-weighted 2Fo-Fc maps for the corresponding complexes (Fig. 2). Both orotate and fumarate molecules were found approximately parallel to the isoalloxazine ring. In detail, the pairs of orotate atoms (N1)/(O2) and (N3)/(O4) make bidentate hydrogen bonds to (Od1)/(Nd2) side chain atoms of Asn68 and Asn195, respectively. Orotate (N1), Asn68 (O) and (Od1) atoms are also hydrogen bonded to a solvent molecule (O2) and (O4) orotate atoms are involved in additional hydrogen bonds with (Og) from Ser196 and (Nd2) atoms from Asn128 and Asn195. One orotate carboxylate oxygen atom (O71) participates as an acceptor in hydrogen bonds with Lys44 (Nz) and Leu72 (N) atoms. The second orotate carboxylate oxygen (O72) is also the acceptor in hydrogen bonds with Gly71 (N) and Met70 (N) atoms. Additionally, O72 is also hydrogen bonded to a solvent molecule. Both waters involved in hydrogen bonds with orotate in the active site of chain A are matched by equivalent water molecules interacting with the orotate molecule associated to chain B. Similarly, the pair of atoms O4A/O4B from fumarate participates in a bidentate hydrogen bond to Nd2/Od1 from Asn195, additionally (O4A) makes a second hydrogen bond to Asn128 (Nd2) (Fig. 3). Fumarate (O1B) atom makes hydrogen bond to Leu72 (N) and Lys44 (Nz), while (O1A) makes hydrogen bound to Gly71 (N). Fumarate (O1A) establishes a hydrogen bond to Asn133 (Nd2) located in the active site loop. Additionally, Asn133 makes hydrogen bonds via its main chain (O) and side chain (Nd2) atoms to Asn54 (Nd2) and Asn68 (Od1). Interactions involving Asn133 are not observed in either the LmDHODH-apo or the LmDHODH-oro because of the disordered state of their active site loops. In LmDHODH-fum, a water molecule occupies the equivalent position of orotate (O2) atom in

Fig. 2. sA-weighted 2Fo-Fc electron density maps countered at 1.5 s level (where s is the root mean square value of the electron density) for orotate (A) and fumarate (B).

A.T. Cordeiro et al. / Biochimie 94 (2012) 1739e1748

1743

Fig. 3. Stereoscopic three-dimensional representation of the LmDHODH active site showing the superposition of orotate (pink) and fumarate (green) complexes. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

LmDHODH-oro, and interacts via hydrogen bonds with fumarate (O4B), Asn133 (Nd2), Asn68 (Nd2) and Ser196 (Og) (Fig. 4). The latter two interactions are equivalent to the hydrogen bonds between orotate (O2) and Asn68 (Nd2) and Ser196 (Og) from LmDHODH-oro. Cys131 which has been previously described to be the catalytic base for class 1A DHODH [30,49] and is responsible for abstracting a proton from the substrate dihydroorotate (C5) is not present in the LmDHODH-apo and LmDHODH-oro due to the lack of interpretable electron density. In the LmDHODH-fum structure, where the catalytic loop is stabilized, the Cys131 is located with its Sg at distances 3.6 Å, 4.2 Å and 4.3 Å from fumarate C3, C4 and O1A, respectively. Overall, the global structures of the protein molecules in the two cocrystal structures are almost identical. Nevertheless, a notable difference is observed in LmDHODH-fum structure where the activesite loop is found in a closed conformation. Different from

LmDHODH-apo and LmDHODH-oro structures where the same region is found disordered, this observation could initially suggest that only fumarate could stabilize the active-site loop in its closed conformation. However, inspection of several crystal structures available for class 1A DHODH in the apo form (1DOR [42], 2DJX [49], 3OIX [50], LmDHODHapo); in complex to orotate (2DOR [47], 2E6A [49], 3C61 [data not published], 2B4G [27], LmDHODH-oro); in complex to fumarate (2E6D [49], LmDHODH-fum) or even in complex to different ligands (2BSL, 2BX7 [52], 3C3N [48], 2E68, 2E6F, 2DJL [49], 3MJY and 3MHU [53]) supports the observation that the position of the catalytic loop described for class 1A crystal structures does not intrinsically depend on the occupancy of the active site. For example, in the apo structures the loop is found stabilized in the closed position for 1DOR and 2DJX, whereas for 3OIX the same region is found stabilized in an open conformation and for LmDHODH-apo is found disordered. For DHODH-orotate complexes the loop is commonly found stabilized in

Fig. 4. Stereoscopic drawing of the dimer interface in LmDHODH structure. In LmDHODH-fum structure a nickel ion is found at the dimer interface with double occupancy. Interactions involved in nickel binding are represented in dashed black lines A water molecule (orange) mediates the interaction of nickel ion (purple) and the protein residues: His174, Phe170 and Tyr142 (light blue). The presence of a nickel ion in this cavity induces a shift in both His174 and Asp171 conformations when compared to the LmDHODH-apo structure (green). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

1744

A.T. Cordeiro et al. / Biochimie 94 (2012) 1739e1748

the closed conformation except for LmDHODH-oro where the same region is found disordered. In general, we observe that the different conformations adopted by the active-site loop in LmDHODH structures is favored by the crystal form, where the active site loop is solventexposed allowing us to capture different conformations. This is very different, for example, from T. cruzi strain Y DHODH [48] and T. brucei DHODH [27] crystal structures where forces of crystal packing also play an important role in stabilizing the conformation of the active site loop. Noteworthy, analysis of class 1A DHODH-inhibitor complexes reveals that not only the catalytic loop can adopt both the closed conformation (2BSL and 2BX7) and a disordered state (3MJY and 3MHU), but also that the length of the disordered region may vary. For example, compared to LmDHODH-apo and LmDHODH-oro where the flexible region comprises residues from Ser130 to Tyr142, the disordered region for LmDHODH in complex with orotate analogs, 5-aminoorotic acid (3MJY) and 5-nitroorotic acid (3MHU), is shorter and comprises residues from Cys131 to Pro138 and from Pro132 to Pro138, respectively. The shortened disordered regions found in 3MJY and 3MHU structures are consequence of direct interactions between the ligands and the base of the catalytic loop, suggesting that class 1A inhibitors can explore fumarate and orotate interactions in the active site in order to enhance selectivity and also interfere with loop flexibility to increase potency. In summary, the three-dimensional structures of LmDHODH-oro and LmDHODH-fum, where orotate and fumarate are found to bind to the same active site, are consistent with the previous kinetic characterization which showed that catalysis of LmDHODH is performed by a Ping-Pong BieBi mechanism [32]. 3.4. Dimer interface LmDHODH exists as homodimer both in solution [33] and in the asymmetric unit. The two monomers of LmDHODH are related by a non-crystallographic 2-fold axis with forty-six amino acid residues from each subunit stabilizing the dimer interface. Hydrophobic, hydrophilic and, in particular, stacking interactions play an important role in dimer stability. P-stacking interactions are found between Phe170 (chain A) and Phe170 (chain B), Phe218 (chain A) and Phe172 (chain B), Phe218 (chain B) and Phe172 (chain A) and a cationp interaction is found between Lys215 and Phe172.

The great majority of residues found at the interface: Pro57, Leu64, Pro138, Gln139, Phe170, Asp171, Phe175, Gly198, Asn199, Gly200, Leu201,Ile203, Glu208, Val210, Val211, Lys213,Lys215, Phe218, Tyr225, Pro228, Thr229, Leu231, Ala232, Leu263 Lys297,Tyr299 are completely conserved within class 1A DHODHs. In addition, conservative substitutions are found at positions Ile197, Glu206, Ser209, Ile212, Leu221, Asn235 and Arg310. This high degree of sequence conservation sheds light on a potential functional role for the dimeric form in class 1A DHODHs. Actually, the relevance of the dimeric form for class 1A DHODH has been previously reported [54]. In their studies, Ottosen and collaborators showed that a dimer dissociation induced by site directed mutagenesis led to the loss of enzymatic activity in L. lactis DHODH. Furthermore, single-molecule studies previously reported have provided evidences for half-sites reactivity in class 1A DHODH, in which only one subunit reacts at a time [55]. In this present model, the dimeric interface could also play an important role in the exchange of information between the two subunits during catalysis. Interestingly, when the catalytic loop is found stabilized, as only observed in LmDHODH-fum structure, a cavity (200 Å3) surrounded by the fully conserved residues Phe170, His174, Tyr169 and Tyr142 is formed at the dimer interface (Video 1). During crystallographic refinement of LmDHODH-fum complex, analysis of sA-weighted 2Fo-Fc and Fo-Fc maps indicated the existence of a strong electron density in this cavity. A careful inspection of the 1.9 Å resolution map revealed that a specific atom tightly interacts with neighbor residues and water molecules within this cavity. Supplementary video related to this article can be found at doi: 10.1016/j.biochi.2012.04.003 To further investigate the character of this atom, we estimated the atomic occupancies and temperature factor for different atoms present in the buffers used for protein purification and crystallization. Water molecules, alkali metal cations: Naþ, Kþ, Liþ, divalent cations: Ca2þ, Ni2þ and the monovalent ions Cl and NHþ 4 , were evaluated to fit and satisfy both chemical and electron density requirements. Our strategy shows that the binding species are most likely to be Ni2þ (present during protein purification), although we cannot completely rule out the possibility of coexistence of other species with lower occupancies. As a consequence of the 2-fold axis of the dimer, the Ni2þ ion has been found to occupy this cavity with double occupancy (Fig. 4) and

Fig. 5. Stereoscopic drawing of the molecular surface in LmDHODH structure. The sites S1, S2, S3, S4 and S5 are indicated by pink, blue, orange, green and red, respectively (A) and (B) indicates chain A (white) and B (light gray). A ribbon representation (in black) indicates the position of the catalytic loop in both subunits and which involves residues from S2, S3 and S5 sites. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

A.T. Cordeiro et al. / Biochimie 94 (2012) 1739e1748

both Ni2þ positions display similar interactions with water molecules and protein residues from both subunits. A water molecule is found to mediate an interaction between each Ni2þ ion (OeNi distance of 2.5 Å) and both Phe170 (OeN distance of 2.8 Å) and His174 (O-Nd1 distance of 2.6 Å) residues (Fig. 4). In addition, the position of the divalent cation is stabilized by the aromatic p-system of Tyr142, as a further evidence of the chemical nature of our ligand. A comparison among LmDHODH-apo, LmDHODH-oro and LmDHODH-fum structures indicates that a conformational change in both His174 and Asp171 side chains and a consequent disruption of a hydrogen bond between those two residues can be observed when the divalent cation is found in the dimer interface (Fig. 4). The range of interactions observed in the LmDHODH-fum complex, are present neither in LmDHODH-apo nor LmDHODH-oro crystal structures, as a consequence of the high flexibility observed for the catalytic loop. Actually, it can be observed that the movement of the region comprising from Leu129 to Tyr142 and required for regulating catalysis has a direct influence on the dimeric interface, resulting in dynamic changes in the volume and charge distribution of the cavity where the cation ion has been found. Furthermore, our findings support the observation that the

1745

stabilization of the catalytic loop does not depend on the occupancy of the active site. Analysis of 19 class 1A DHODH structures available at the Protein Data Bank reveals that this binding pocket at the dimer interface is always occupied by ligands of different volume and chemical properties. Water molecules are usually found, as observed in the structures of LmDHODH-apo and LmDHODH-oro complexes. Glycerol molecules in different orientations and a DTT molecule have also been modeled in the electron density found this cavity (PDB IDs 3OIX, 1JUE, 2BSL, 2BX7, 3C3N and 3C61). In one particular case, analysis of the high resolution data available for the crystal structure of T. cruzi DHODH in complex to oxonate (PDB ID 2E6F [49],) allowed us to unambiguously assign a PEG molecule for remaining electron density found at the dimer interface (supplemental Fig. S1). Furthermore, the final sA-weighted 2Fo-Fc and Fo-Fc maps for the available structures frequently show remaining spurious density in this region accounting for either low occupancy or high flexible molecules. This promiscuous nature of this binding pocket could be interpreted as a result of extensive hydrophobic contacts as well as a network of polar interactions available on the periphery of the protein surface.

Fig. 6. Sequence alignment of trypanosomatid DHODHs: LmDHODH (Leishmania major), LiDHODH (Leishmania infantum) LdDHODH (Leishmania donovani), LmxDHODH (Leishmania mexicana), LbrDHODH (Leishmania braziliensis), TbDHODH (Trypanosoma brucei), TcDHODH-Y (Trypanosoma cruzi strain Y), TcDHODH-CLBrener1 (T. cruzi strain CLBrener) and TcDHODH-Tulahuen1 (T. cruzi strain Tulahuen). The sites S1, S2, S3, S4 and S5 are indicated by pink, blue, orange, green and red stars, respectively. The alignment was performed using MULTALIN [57] and graphically displayed using ESPript [58]. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

1746

A.T. Cordeiro et al. / Biochimie 94 (2012) 1739e1748

To date there is no evidence supporting the need of a cation ion or any ligand for the stabilization the dimeric form or for regulating the catalytic activity in class 1A DHODHs. However, as observed in this work, there is a structural relationship between the dimer interface and the conformation of the catalytic loop, suggesting that the binding of specific molecules at the oligomer interface may alter the catalytic properties of class 1A DHODH. Thus, considering the relevance of the oligomerization state for class 1A DHODH activity [54,55], together with the absence of the dimeric interface in class 2 DHODHs, it is of our interest to exploit this binding site for the design of class 1A DHODH selective inhibitors. Site-directed mutagenesis studies are now in progress in order to provide the molecular basis that correlates the dimer interface and protein function.

structures. At the S2 site a highly conserved Leu102 is replaced by Phe102 in L. brasiliensis DHODH and Asn107 is replaced by Ser107 in T. brucei DHODH. Despite those few substitutions, the profile of interactions for molecular recognition is kept conserved at S2 and S4 binding cavities within TrypDHODHs. The only point mutation that deserves special attention is at Tyr150 which is found replaced by Cys150 in the majority of Leishmania species with exception of Leishmania braziliensis (Fig. 6). The differences in volume, geometry of its side chain at the S2 site and the highly reactive character of this free cysteine residue could drive an alternative route for the design of selective inhibitors covering a large spectrum of Leishmania DHODHs.

3.5. Trypanosomatid DHODHs and implications for inhibitor design

We report the determination of the crystal structure of LmDHODH in its apo form and in complex with both orotate and fumarate molecules. Our findings provide a new contribution to the understanding of the mechanism of action of class 1A DHODHs and for the structure-assisted design of inhibitors against LmDHODH. Furthermore, the high sequence and structural similarities observed among trypanosomal DHODHs associated to the intrinsic differences between class 1 and class 2 DHODHs suggests that a single strategy can be used for the development of selective DHODH inhibitors that target multiple trypanosomatid targets. Work is ongoing to validate DHODH as druggable target for multiple neglected tropical diseases such as Leishmaniasis, Sleeping Sickness and Chagas’ Diseases.

The genome sequencing projects of trypanosomatid organisms, which include human pathogenic species like T. cruzi [56], T. brucei [20] and Leishmania species [14,15] have had a major contribution in the characterization of several genes involved in the parasites’ biology. Also, they have allowed the identification of similarities and differences at molecular level of these related species. To date, the genome annotation process has assigned the putative function for DHODH enzyme to gene products of different trypanosomatids and reveals a high degree of sequence identity varying from 73 to 76% when comparing LmDHODH to homologous trypanosoma DHODHs and 83e96% when compared to other Leishmania species (Fig. 6). Trypanosomatid DHODH (TrypDHODHs) share around 50e57% of sequence identity when comparing to other members of class 1A DHODHs and 18e27% when comparing to members of class 2 DHODH. In particular, LmDHODH and the homologous human enzyme (HsDHODH) share approximately 25% of sequence identity. According to our previous work, there are significant differences between TrypDHODHs and the homologous human enzyme that could guide the selection of target sites for drug discovery [48]. By using TcDHODH as a model, topological analysis helped us to identify five different independent regions that display appropriate shape, volume and location to interact with ligands and thus could be exploited for the design of selective ligands against TrypDHODHs [53]. Named according to the distance from the active site, these sites comprise: the active site (S1); an extension of S1 site (S2); the site behind the catalytic loop (S3); an extension of S3 site (S4); and the cavity found at the dimeric interface (S5) (Fig. 5). Careful analyses of the crystal structures of LmDHODH-apo and LmDHODH-oro described in this work which reveals the lack of electron density from Leu129 to Tyr142 have allowed us to estimate the length of the flexible region of the active site loop (b4- aA loop; Fig. 5). In addition, the monitoring of the conformation dynamics of the active site loop, which comprises residues from S2, S3 and S5 sites (Fig. 5), has brought to our attention the plastic character of S2 to S5 pockets of interaction, as well as the different possible routes of accessing the active site, suggesting new possible strategies to be exploited in the design of TrypDHODHs inhibitors. When comparing LmDHODH against all available TrypDHODHs sequences and structures, we found that all sites (S1 to S5) are highly conserved in terms of conformation and chemical nature (Fig. 6). S1, S3 and S5 sites are found fully conserved and only few substitutions are found at S4 and S2 sites, the later recently validated as a new site of interaction for inhibitor development against class 1A DHODHs [53]. At the S4 site a single mutation can be observed in Leishmania species where the conserved Ile172 found in T. cruzi and T. brucei DHODHs is replaced by a Phe172 (Fig. 6). As a result, S4 site is found to be slightly shallower in Leishmania

4. Conclusion

Acknowledgments This work received financial support from Fundação de Amparo a Pesquisa do Estado de São Paulo (FAPESP), Grants 01/14583-0, 02/ 07032-0 and 04/14407-6. References [1] B. Liese, M. Rosenberg, A. Schratz, Neglected tropical diseases 1 programmes, partnerships, and governance for elimination and control of neglected tropical diseases, Lancet 375 (2010) 67e76. [2] Eradicating infectious diseases in 2010, Nature Reviews Microbiology 8 (2010). [3] J.O. Gyapong, M. Gyapong, N. Yellu, K. Anakwah, G. Amofah, M. Bockarie, S. Adjei, neglected tropical diseases 2 integration of control of neglected tropical diseases into health-care systems: challenges and opportunities, Lancet 375 (2010) 160e165. [4] M.C. Baker, E. Mathieu, F.M. Fleming, M. Deming, J.D. King, A. Garba, J.B. Koroma, M. Bockarie, A. Kabore, D.P. Sankara, D.H. Molyneux, Neglected tropical diseases 3 mapping, monitoring, and surveillance of neglected tropical diseases: towards a policy framework, Lancet 375 (2010) 231e238. [5] S.L. Croft, S. Sundar, A.H. Fairlamb, Drug resistance in leishmaniasis, Clinical Microbiology Reviews 19 (2006) 111. [6] H.C. Maltezou, Drug resistance in visceral leishmaniasis, Journal of Biomedicine and Biotechnology (2010). [7] B.L. Herwaldt, Leishmaniasis, Lancet 354 (1999) 1191e1199. [8] WHO, Global Plan to Combat Neglected Diseases 2008-2015 Geneva, Switzerland (2007). [9] J. Mishra, A. Saxena, S. Singh, Chemotherapy of leishmaniasis: Past, present and future, Current Medicinal Chemistry 14 (2007) 1153e1169. [10] E. Dumonteil, Vaccine development against Trypanosoma cruzi and Leishmania species in the post-genomic era, Infection Genetics and Evolution 9 (2009) 1075e1082. [11] S.L. Croft, V. Yardley, Chemotherapy of leishmaniasis, Current Pharmaceutical Design 8 (2002) 319e342. [12] S.L. Croft, G.H. Coombs, Leishmaniasis - current chemotherapy and recent advances in the search for novel drugs, Trends in Parasitology 19 (2003) 502e508. [13] S.L. Croft, M.P. Barrett, J.A. Urbina, Chemotherapy of trypanosomiases and leishmaniasis, Trends in Parasitology 21 (2005) 508e512. [14] A.C. Ivens, C.S. Peacock, E.A. Worthey, L. Murphy, G. Aggarwal, M. Berriman, E. Sisk, M.A. Rajandream, E. Adlem, R. Aert, A. Anupama, Z. Apostolou, P. Attipoe, N. Bason, C. Bauser, A. Beck, S.M. Beverley, G. Bianchettin, K. Borzym, G. Bothe, C.V. Bruschi, M. Collins, E. Cadag, L. Ciarloni, C. Clayton,

A.T. Cordeiro et al. / Biochimie 94 (2012) 1739e1748

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22] [23]

[24]

[25]

[26]

[27]

[28] [29]

[30]

R.M.R. Coulson, A. Cronin, A.K. Cruz, R.M. Davies, J. De Gaudenzi, D.E. Dobson, A. Duesterhoeft, G. Fazelina, N. Fosker, A.C. Frasch, A. Fraser, M. Fuchs, C. Gabel, A. Goble, A. Goffeau, D. Harris, C. Hertz-Fowler, H. Hilbert, D. Horn, Y.T. Huang, S. Klages, A. Knights, M. Kube, N. Larke, L. Litvin, A. Lord, T. Louie, M. Marra, D. Masuy, K. Matthews, S. Michaeli, J.C. Mottram, S. Muller-Auer, H. Munden, S. Nelson, H. Norbertczak, K. Oliver, S. O’Neil, M. Pentony, T.M. Pohl, C. Price, B. Purnelle, M.A. Quail, E. Rabbinowitsch, R. Reinhardt, M. Rieger, J. Rinta, J. Robben, L. Robertson, J.C. Ruiz, S. Rutter, D. Saunders, M. Schafer, J. Schein, D.C. Schwartz, K. Seeger, A. Seyler, S. Sharp, H. Shin, D. Sivam, R. Squares, S. Squares, V. Tosato, C. Vogt, G. Volckaert, R. Wambutt, T. Warren, H. Wedler, J. Woodward, S.G. Zhou, W. Zimmermann, D.F. Smith, J.M. Blackwell, K.D. Stuart, B. Barrell, P.J. Myler, The genome of the kinetoplastid parasite, Leishmania major, Science 309 (2005) 436e442. C.S. Peacock, K. Seeger, D. Harris, L. Murphy, J.C. Ruiz, M.A. Quail, N. Peters, E. Adlem, A. Tivey, M. Aslett, A. Kerhornou, A. Ivens, A. Fraser, M.A. Rajandream, T. Carver, H. Norbertczak, T. Chillingworth, Z. Hance, K. Jagels, S. Moule, D. Ormond, S. Rutter, R. Squares, S. Whitehead, E. Rabbinowitsch, C. Arrowsmith, B. White, S. Thurston, F. Bringaud, S.L. Baldauf, A. Faulconbridge, D. Jeffares, D.P. Depledge, S.O. Oyola, J.D. Hilley, L.O. Brito, L.R.O. Tosi, B. Barrell, A.K. Cruz, J.C. Mottram, D.F. Smith, M. Berriman, Comparative genomic analysis of three Leishmania species that cause diverse human disease, Nature Genetics 39 (2007) 839e847. P.J. Alcolea, A. Alonso, M.J. Gomez, I. Moreno, M. Dominguez, V. Parro, V. Larraga, Transcriptomics throughout the life cycle of Leishmania infantum: high down-regulation rate in the amastigote stage, International Journal for Parasitology 40 (2010) 1497e1516. J. Drummelsmith, V. Brochu, I. Girard, N. Messier, M. Ouellette, Proteome mapping of the protozoan parasite Leishmania and application to the study of drug targets and resistance mechanisms, Molecular & Cellular Proteomics 2 (2003) 146e155. K. Leifso, G. Cohen-Freue, N. Dogra, A. Murray, W.R. McMaster, Genomic and proteomic expression analysis of Leishmania promastigote and amastigote life stages: the Leishmania genome is constitutively expressed, Molecular and Biochemical Parasitology 152 (2007) 35e46. P.G. Nugent, S.A. Karsani, R. Wait, J. Tempero, D.F. Smith, Proteomic analysis of Leishmania mexicana differentiation, Molecular and Biochemical Parasitology 136 (2004) 51e62. M. Berriman, E. Ghedin, C. Hertz-Fowler, G. Blandin, H. Renauld, D.C. Bartholomeu, N.J. Lennard, E. Caler, N.E. Hamlin, B. Haas, W. Bohme, L. Hannick, M.A. Aslett, J. Shallom, L. Marcello, L.H. Hou, B. Wickstead, U.C.M. Alsmark, C. Arrowsmith, R.J. Atkin, A.J. Barron, F. Bringaud, K. Brooks, M. Carrington, I. Cherevach, T.J. Chillingworth, C. Churcher, L.N. Clark, C.H. Corton, A. Cronin, R.M. Davies, J. Doggett, A. Djikeng, T. Feldblyum, M.C. Field, A. Fraser, I. Goodhead, Z. Hance, D. Harper, B.R. Harris, H. Hauser, J. Hostetter, A. Ivens, K. Jagels, D. Johnson, J. Johnson, K. Jones, A.X. Kerhornou, H. Koo, N. Larke, S. Landfear, C. Larkin, V. Leech, A. Line, A. Lord, A. MacLeod, P.J. Mooney, S. Moule, D.M.A. Martin, G.W. Morgan, K. Mungall, H. Norbertczak, D. Ormond, G. Pai, C.S. Peacock, J. Peterson, M.A. Quail, E. Rabbinowitsch, M.A. Rajandream, C. Reitter, S.L. Salzberg, M. Sanders, S. Schobel, S. Sharp, M. Simmonds, A.J. Simpson, L. Talton, C.M.R. Turner, A. Tait, A.R. Tivey, S. Van Aken, D. Walker, D. Wanless, S.L. Wang, B. White, O. White, S. Whitehead, J. Woodward, J. Wortman, M.D. Adams, T.M. Embley, K. Gull, E. Ullu, J.D. Barry, A.H. Fairlamb, F. Opperdoes, B.G. Barret, J.E. Donelson, N. Hall, C.M. Fraser, S.E. Melville, N.M. El-Sayed, The genome of the African trypanosome Trypanosoma brucei, Science 309 (2005) 416e422. C. Verlinde, E.A. Merritt, F. Vandenakker, H.D. Kim, I. Feil, L.F. Delboni, S.C. Mande, S. Sarfaty, P.H. Petra, W.G.J. Hol, Protein Crystallography and Infectious-diseases, Protein Science 3 (1994) 1670e1686. H.K. Majumder, Drug Targets in Kenotoplastid Parasites 625 (2008) 158. N.S. Carter, P. Yates, C.S. Arendt, J.M. Boitz, B. Ullman, Purine and pyrimidine metabolism in Leishmania, Drug Targets in Kinetoplastid Parasites 625 (2008) 141e154. P.S. Monzani, S. Trapani, O.H. Thiemann, G. Oliva, Crystal structure of Leishmania tarentolae hypoxanthine-guanine phosphoribosyltransferase, Bmc Structural Biology 7 (2007). M. Silva, C. Silva, J. Iulek, G. Oliva, O.H. Thiemann, Crystal structure of adenine phosphoribosyltransferase from Leishmania tarentolae: potential implications for APRT catalytic mechanism, Biochimica Et Biophysica Acta-Proteins and Proteomics 1696 (2004) 31e39. T. Annoura, T. Nara, T. Makiuchi, T. Hashimoto, T. Aoki, The origin of dihydroorotate dehydrogenase genes of kinetoplastids, with special reference to their biological significance and adaptation to anaerobic, parasitic conditions, Journal of Molecular Evolution 60 (2005) 113e127. T.L. Arakaki, F.S. Buckner, J.R. Gillespie, N.A. Malmquist, M.A. Phillips, O. Kalyuzhniy, J.R. Luft, G.T. DeTitta, C. Verlinde, W.C. Van Voorhis, W.G.J. Hol, E.A. Merritt, Characterization of Trypanosoma brucei dihydroorotate dehydrogenase as a possible drug target; structural, kinetic and RNAi studies, Molecular Microbiology 68 (2008) 37e50. B.A. Fox, D.J. Bzik, De novo pyrimidine biosynthesis is required for virulence of Toxoplasma gondii, Nature 415 (2002) 926e929. J. Baldwin, A.M. Farajallah, N.A. Malmquist, P.K. Rathod, M.A. Phillips, Malarial dihydroorotate dehydrogenase, Journal of Biological Chemistry 277 (2002) 41827e41834. O. Bjornberg, P. Rowland, S. Larsen, K.F. Jensen, Active site of dihydroorotate dehydrogenase A from Lactococcus lactis investigated by chemical modification and mutagenesis, Biochemistry 36 (1997) 16197e16205.

1747

[31] E. Zameitat, Z. Gojkovic, W. Knecht, J. Piskur, M. Loeffler, Biochemical characterization of recombinant dihydroorotate dehydrogenase from the opportunistic pathogenic yeast Candida albicans, Febs Journal 273 (2006) 3183e3191. [32] E. Zameitat, A.J. Pierik, K. Zocher, M. Loeffler, Dihydroorotate dehydrogenase from Saccharomyces cerevisiae: spectroscopic investigations with the recombinant enzyme throw light on catalytic properties and metabolism of fumarate analogues, Fems Yeast Research 7 (2007) 897e904. [33] P.R. Feliciano, A.T. Cordeiro, A.J. Costa, M.C. Nonato, Cloning, expression, purification, and characterization of Leishmania major dihydroorotate dehydrogenase, Protein Expression and Purification 48 (2006) 98e103. [34] S.P. Liu, E.A. Neidhardt, T.H. Grossman, T. Ocain, J. Clardy, Structures of human dihydroorotate dehydrogenase in complex with antiproliferative agents, Structure with Folding & Design 8 (2000) 25e33. [35] S.G. Couto, M.C. Nonato, A.J. Costa-Filho, Defects in vesicle core induced by Escherichia coli dihydroorotate dehydrogenase, Biophysical Journal 94 (2008) 1746e1753. [36] S.G. Couto, M.C. Nonato, A.J. Costa-Filho, Site directed spin labeling studies of Escherichia coli dihydroorotate dehydrogenase N-terminal extension, Biochemical and Biophysical Research Communications 414 (2011) 487e492. [37] R.L. Fagan, B.A. Palfey, Roles in binding and Chemistry for conserved active site residues in the class 2 dihydroorotate dehydrogenase from Escherichia coli, Biochemistry 48 (2009) 7169e7178. [38] M. Nagy, F. Lacroute, D. Thomas, Divergent evolution of pyrimidine biosynthesis between anaerobic and aerobic yeasts, Proceedings of the National Academy of Sciences of the United States of America 89 (1992) 8966e8970. [39] A.T. Cordeiro, P.R. Feliciano, M.C. Nonato, Crystallization and preliminary Xray diffraction analysis of Leishmania major dihydroorotate dehydrogenase, Acta Crystallographica Section F-Structural Biology and Crystallization Communications 62 (2006) 1049e1051. [40] A.G.W. Leslie, The integration of macromolecular diffraction data, Acta Crystallographica Section D-Biological Crystallography 62 (2006) 48e57. [41] P. Evans, Scaling and assessment of data quality, Acta Crystallographica Section D-Biological Crystallography 62 (2006) 72e82. [42] P. Rowland, F.S. Nielsen, K.F. Jensen, S. Larsen, The crystal structure of the flavin containing enzyme dihydroorotate dehydrogenase A from Lactococcus lactis, Structure 5 (1997) 239e252. [43] P. Emsley, K. Cowtan, Coot: model-building tools for molecular graphics, Acta Crystallographica Section D-Biological Crystallography 60 (2004) 2126e2132. [44] G.N. Murshudov, A.A. Vagin, E.J. Dodson, Refinement of macromolecular structures by the maximum-likelihood method, Acta Crystallographica Section D-Biological Crystallography 53 (1997) 240e255. [45] V.B. Chen, W.B. Arendall, J.J. Headd, D.A. Keedy, R.M. Immormino, G.J. Kapral, L.W. Murray, J.S. Richardson, D.C. Richardson, MolProbity: all-atom structure validation for macromolecular crystallography, Acta Crystallographica Section D-Biological Crystallography 66 (2010) 12e21. [46] W.L. Delano, The PyMOL Molecular Graphics System, DeLano Scientific LLC, San Carlos, CA, USA. [47] P. Rowland, O. Bjornberg, F.S. Nielsen, K.F. Jensen, S. Larsen, The crystal structure of Lactococcus lactis dihydroorotate dehydrogenase A complexed with the enzyme reaction product throws light on its enzymatic function, Protein Science 7 (1998) 1269e1279. [48] M.P. Pinheiro, J. Iulek, M.C. Nonato, Crystal structure of Trypanosoma cruzi dihydroorotate dehydrogenase from Y strain, Biochemical and Biophysical Research Communications 369 (2008) 812e817. [49] D.K. Inaoka, K. Sakamoto, H. Shimizu, T. Shiba, G. Kurisu, T. Nara, T. Aoki, K. Kita, S. Harada, Structures of Trypanosoma cruzi dihydroorotate dehydrogenase complexed with substrates and products: atomic resolution insights into mechanisms of dihydroorotate oxidation and fumarate reduction, Biochemistry 47 (2008) 10881e10891. [50] Y. Liu, Z.Q. Gao, C.P. Liu, J.H. Xu, L.F. Li, C.N. Ji, X.D. Su, Y.H. Dong, Structure of the putative dihydroorotate dehydrogenase from Streptococcus mutans, Acta Crystallographica Section F-Structural Biology and Crystallization Communications 67 (2011) 182e187. [51] B. Walse, V.T. Dufe, B. Svensson, I. Fritzson, L. Dahlberg, A. Khairoullina, U. Wellmar, S. Al-Karadaghi, The structures of human dihydroorotate dehydrogenase with and without inhibitor reveal conformational flexibility in the inhibitor and substrate binding sites, Biochemistry 47 (2008) 8929e8936. [52] A.E. Wolfe, M. Thymark, S.G. Gattis, R.L. Fagan, Y.-c. Hu, E. Johansson, S. Arent, S. Larsen, B.A. Palfey, Interaction of benzoate pyrimidine analogues with class 1A dihydroorotate dehydrogenase from Lactococcus lactis, Biochemistry 46 (2007) 5741e5753. [53] J. Cheleski, J.R. Rocha, M.P. Pinheiro, H.J. Wiggers, A.B.F. da Silva, M.C. Nonato, C.A. Montanari, Novel insights for dihydroorotate dehydrogenase class 1A inhibitors discovery, European Journal of Medicinal Chemistry 45 (2010) 5899e5909. [54] M.B. Ottosen, O. Bjornberg, S. Norager, S. Larsen, B.A. Palfey, K.F. Jensen, The dimeric dihydroorotate dehydrogenase A from Lactococcus lactis dissociates reversibly into inactive monomers, Protein Science 11 (2002) 2575e2583. [55] J. Shi, J. Dertouzos, A. Gafni, D. Steel, B.A. Palfey, Single-molecule kinetics reveals signatures of half-sites reactivity in dihydroorotate dehydrogenase A catalysis, Proceedings of the National Academy of Sciences of the United States of America 103 (2006) 5775e5780. [56] N.M. El-Sayed, P.J. Myler, D.C. Bartholomeu, D. Nilsson, G. Aggarwal, A.N. Tran, E. Ghedin, E.A. Worthey, A.L. Delcher, G. Blandin, S.J. Westenberger, E. Caler, G.C. Cerqueira, C. Branche, B. Haas, A. Anupama, E. Arner, L. Aslund, P. Attipoe,

1748

A.T. Cordeiro et al. / Biochimie 94 (2012) 1739e1748 E. Bontempi, F. Bringaud, P. Burton, E. Cadag, D.A. Campbell, M. Carrington, J. Crabtree, H. Darban, J.F. da Silveira, P. de Jong, K. Edwards, P.T. Englund, G. Fazelina, T. Feldblyum, M. Ferella, A.C. Frasch, K. Gull, D. Horn, L.H. Hou, Y.T. Huang, E. Kindlund, M. Ktingbeil, S. Kluge, H. Koo, D. Lacerda, M.J. Levin, H. Lorenzi, T. Louie, C.R. Machado, R. McCulloch, A. McKenna, Y. Mizuno, J.C. Mottram, S. Nelson, S. Ochaya, K. Osoegawa, G. Pai, M. Parsons, M. Pentony, U. Pettersson, M. Pop, J.L. Ramirez, J. Rinta, L. Robertson, S.L. Salzberg, D.O. Sanchez, A. Seyler, R. Sharma, J. Shetty, A.J. Simpson, E. Sisk,

M.T. Tammi, R. Tarteton, S. Teixeira, S. Van Aken, C. Vogt, P.N. Ward, B. Wickstead, J. Wortman, O. White, C.M. Fraser, K.D. Stuart, B. Andersson, The genome sequence of Trypanosoma cruzi, etiologic agent of Chagas disease, Science 309 (2005) 409e415. [57] F. Corpet, Multiple sequence alignment with hierarchical-clustering, Nucleic Acids Research 16 (1988) 10881e10890. [58] P. Gouet, E. Courcelle, D.I. Stuart, F. Metoz, ESPript: analysis of multiple sequence alignments in PostScript, Bioinformatics 15 (1999) 305e308.