Structural Insights into Mycobacterium tuberculosis Rv2671 Protein as a Dihydrofolate Reductase Functional Analogue Contributing to para-Aminosalicylic Acid Resistance.

Mycobacterium tuberculosis (Mtb) Rv2671 is annotated as a 5-amino-6-ribitylamino-2,4(1H,3H)-pyrimidinedione 5'-phosphate (AROPP) reductase (RibD) in the riboflavin biosynthetic pathway. Recently, a strain of Mtb with a mutation in the 5' untranslated region of Rv2671, which resulted in its overexpression, was found to be resistant to dihydrofolate reductase (DHFR) inhibitors including the anti-Mtb drug para-aminosalicylic acid (PAS). In this study, a biochemical analysis of Rv2671 showed that it was able to catalyze the reduction of dihydrofolate (DHF) to tetrahydrofolate (THF), which explained why the overexpression of Rv2671 was sufficient to confer PAS resistance. We solved the structure of Rv2671 in complex with the NADP(+) and tetrahydrofolate (THF), which revealed the structural basis for the DHFR activity. The structures of Rv2671 complexed with two DHFR inhibitors, trimethoprim and trimetrexate, provided additional details of the substrate binding pocket and elucidated the differences between their inhibitory activities. Finally, Rv2671 was unable to catalyze the reduction of AROPP, which indicated that Rv2671 and its closely related orthologues are not involved in riboflavin biosynthesis.

High-resolution structures of Trypanosoma brucei pteridine reductase ligand complexes inform on the placement of new molecular entities in the active site of a potential drug target.

Pteridine reductase (PTR1) is a potential target for drug development against parasitic Trypanosoma and Leishmania species. These protozoa cause serious diseases for which current therapies are inadequate. High-resolution structures have been determined, using data between 1.6 and 1.1 Šresolution, of T. brucei PTR1 in complex with pemetrexed, trimetrexate, cyromazine and a 2,4-diaminopyrimidine derivative. The structures provide insight into the interactions formed by new molecular entities in the enzyme active site with ligands that represent lead compounds for structure-based inhibitor development and to support early-stage drug discovery.

Structures of dihydrofolate reductase-thymidylate synthase of Trypanosoma cruzi in the folate-free state and in complex with two antifolate drugs, trimetrexate and methotrexate.

The flagellate protozoan parasite Trypanosoma cruzi is the pathogenic agent of Chagas disease (also called American trypanosomiasis), which causes approximately 50,000 deaths annually. The disease is endemic in South and Central America. The parasite is usually transmitted by a blood-feeding insect vector, but can also be transmitted via blood transfusion. In the chronic form, Chagas disease causes severe damage to the heart and other organs. There is no satisfactory treatment for chronic Chagas disease and no vaccine is available. There is an urgent need for the development of chemotherapeutic agents for the treatment of T. cruzi infection and therefore for the identification of potential drug targets. The dihydrofolate reductase activity of T. cruzi, which is expressed as part of a bifunctional enzyme, dihydrofolate reductase-thymidylate synthase (DHFR-TS), is a potential target for drug development. In order to gain a detailed understanding of the structure-function relationship of T. cruzi DHFR, the three-dimensional structure of this protein in complex with various ligands is being studied. Here, the crystal structures of T. cruzi DHFR-TS with three different compositions of the DHFR domain are reported: the folate-free state, the complex with the lipophilic antifolate trimetrexate (TMQ) and the complex with the classical antifolate methotrexate (MTX). These structures reveal that the enzyme is a homodimer with substantial interactions between the two TS domains of neighboring subunits. In contrast to the enzymes from Cryptosporidium hominis and Plasmodium falciparum, the DHFR and TS active sites of T. cruzi lie on the same side of the monomer. As in other parasitic DHFR-TS proteins, the N-terminal extension of the T. cruzi enzyme is involved in extensive interactions between the two domains. The DHFR active site of the T. cruzi enzyme shows subtle differences compared with its human counterpart. These differences may be exploited for the development of antifolate-based therapeutic agents for the treatment of T. cruzi infection.

Mutant Gly482 and Thr482 ABCG2 mediate high-level resistance to lipophilic antifolates.

Cellular uptake of hydrophilic antifolates proceeds via the reduced folate carrier whereas lipophilic antifolates enter cells by diffusion. Recently we have shown that transfectant cells overexpressing the mutant G482 ABCG2 displayed 120-6,250-fold resistance to hydrophilic antifolates than untransfected cells upon 4 h drug exposure, but lost almost all their antifolate resistance upon 72 h drug exposure (Shafran et al. in Cancer Res 65:8414-8422, 2005). Here we explored the ability of the wild type (WT) R482-as well as the mutant G482-and T482 ABCG2 to confer resistance to lipophilic antifolate inhibitors of dihydrofolate reductase (trimetrexate, piritrexim, metoprine and pyrimethamine) and thymidylate synthase (AG337, AG377 and AG331). Lipophilic antifolate resistance was determined using growth inhibition assays upon 72 h drug exposure. Cells overexpressing these mutant efflux transporters displayed up to 106-fold resistance to lipophilic antifolates relative to untransfected cells; this resistance was reversed by the specific and potent ABCG2 efflux inhibitor Ko143. In contrast, cells overexpressing the WT R482 ABCG2 exhibited either no or only a low-level of lipophilic antifolate resistance. These results provide the first evidence that overexpression of the mutant G482- and T482 but not the WT R482 ABCG2 confers a high-level of resistance to lipophilic antifolates. The high membrane partitioning of lipophilic antifolates along with the large confinement of ABCG2 to the plasma membrane suggest that these mutant ABCG2 transporters may possibly recognize and extrude lipophilic antifolates from the lipid bilayer. The potential implications to cancer chemotherapy as well as the mechanism of anticancer drug extrusion by these mutant exporters are discussed.

A new nonlinear mixture response surface paradigm for the study of synergism: a three drug example.

A flexible approach to response surface modeling for the study of the joint action of three active anticancer agents is used to model a complex pattern of synergism, additivity and antagonism in an in vitro cell growth assay. The method for determining a useful nonlinear response surface model depends upon a series of steps using appropriate scaling of drug concentrations and effects, raw data modeling, and hierarchical parameter modeling. The method is applied to a very large in vitro study of the combined effect of Trimetrexate (TMQ), LY309887 (LY), and Tomudex (TDX) on inhibition of cancer cell growth. The base model employed for modeling dose-response effect is the four parameter Hill equation [1]. In the hierarchical aspect of the final model, the base Hill model is treated as a function of the total amount of the three drug mixture and the Hill parameters, background B, dose for 50% effect D50, and slope m, are understood as functions of the three drug fractions. The parameters are modeled using the canonical mixture polynomials from the mixture experiment methodologies introduced by Scheff [2]. We label the model generated a Nonlinear Mixture Amount model with control observations, or zero amounts, an "NLMAZ" model. This modeling paradigm provides for the first time an effective statistical approach to modeling complex patterns of local synergism, additivity, and antagonism in the same data set, the possibility of including additional experimental components beyond those in the mixture, and the capability of modeling three or more drugs.

Dihydrofolate reductase mutant with exceptional resistance to methotrexate but not to trimetrexate.

Two double (F31A/F34A, I60A/L67G) and one quadruple (F31A/F34A/I60A/L67G) mutant murine dihydrofolate reductases were constructed and evaluated for their ability to impart antifolate resistance. Both I60A/L67G and F31A/F34A/I60A/L67G were found to be unstable and devoid of catalytic activity. The K(i) values for F31A/F34A, methotrexate (MTX), bis-MTX, and PT-523 were found to be 10100-, 4410-, and 617-fold higher than the wild-type enzyme, respectively, but only 13.5-fold higher for trimetrexate (TMTX). These findings suggest that F31A/F34A could be used for gene therapy to render normal cells resistant to MTX but sensitive to TMTX.

Analysis of two polymorphic forms of a pyrido[2,3-d]pyrimidine N9-C10 reversed-bridge antifolate binary complex with human dihydrofolate reductase.

The results of the crystal structure determination of human dihydrofolate reductase (hDHFR) as a binary complex with the potent N9-C10 reversed-bridge antifolate inhibitor 2,4-diamino-6-[N-(3',4',5'-trimethoxybenzyl)-N-methylamino]pyrido[2,3-d]pyrimidine (1) are reported for two independent polymorphic rhombohedral R3 lattices [R3(1) and R3(2)]. Data from these two crystal forms were refined to 1.90 A resolution for complex R3(1), with R = 0.186 for 9689 data, and to 1.80 A resolution for complex R3(2), with R = 0.194 for 13 305 data. Changes in the loop geometry between the two structures reflects contact differences in the packing environments in the two R3 lattices. The largest changes (between 0.5 and 1.7 A) are observed for the loop regions encompassing residues 16-25, 40-48, 81-89, 99-108, 143-148 and 161-169. Comparison of the intermolecular contacts of these loops reveals that the R3(2) lattice is more tightly packed, as reflected in its smaller V(M) value and smaller solvent content. The conformation of inhibitor (1) is similar in both structures and the N9-C10 bridge geometry is more similar to that observed for the normal C9-N10 bridge of trimetrexate (TMQ) than to the other N9-C10 reversed-bridge antifolates previously reported. The effect of the N9-C10 reversed-bridge geometry is to distort the bridge from coplanarity with the pyrido[2,3-d]pyrimidine ring system and to twist the C10 methylene conformation towards a gauche conformation. This also influences the conformation of the methoxybenzyl ring, moving it away from a trans position and placing the 5'-methoxy group deeper within the hydrophobic pocket made by Leu60, Pro61 and Asn64 of the hDHFR active site.

Design, synthesis, computational prediction, and biological evaluation of ester soft drugs as inhibitors of dihydrofolate reductase from Pneumocystis carinii.

A series of lipophilic soft drugs structurally related to the nonclassical dihydrofolate reductase (DHFR) inhibitors trimetrexate and piritrexim have been designed, synthesized, and evaluated in DHFR assays, with special emphasis on the inhibition of P. carinii DHFR. The best inhibitors, encompassing an ester bond in the bridge connecting the two aromatic systems, were approximately 10 times less potent than trimetrexate and piritrexim. The metabolites were designed to be poor inhibitors. Furthermore, molecular dynamics simulations of three ligands in complex with DHFR from Pneumocystis carinii and from the human enzyme were conducted in order to better understand the factors determining the selectivity. A correct ranking of the relative inhibition of DHFR was achieved utilizing the linear interaction energy method. The soft drugs are intended for local administration. One representative ester was selected for a pharmacokinetic study in rats where it was found to undergo fast metabolic degradation to the predicted inactive metabolites.

The antifolate trimetrexate: observation of the enzyme-binding conformation.

The crystal structure of the title compound contains four 2, 4-diamino-5-methyl-6-[(3,4,5-trimethoxyanilino)methyl]quinazoline molecules, two dimethyl sulfoxide molecules and three water molecules in the asymmetric unit, i.e. 4C(19)H(23)N(5)O(3).-2C(2)H(6)OS.3H(2)O. All four quinazoline molecules adopt trans,-gauche conformations. An extensive hydrogen-bond network involving N. N base-pairing interactions, as well as the dimethyl sulfoxide and water molecules, stabilizes the crystal structure.

Structure and dynamics in solution of the complex of Lactobacillus casei dihydrofolate reductase with the new lipophilic antifolate drug trimetrexate.

We have determined the three-dimensional solution structure of the complex of Lactobacillus casei dihydrofolate reductase and the anticancer drug trimetrexate. Two thousand seventy distance, 345 dihedral angle, and 144 hydrogen bond restraints were obtained from analysis of multidimensional NMR spectra recorded for complexes containing 15N-labeled protein. Simulated annealing calculations produced a family of 22 structures fully consistent with the constraints. Several intermolecular protein-ligand NOEs were obtained by using a novel approach monitoring temperature effects of NOE signals resulting from dynamic processes in the bound ligand. At low temperature (5 degrees C) the trimethoxy ring of bound trimetrexate is flipping sufficiently slowly to give narrow signals in slow exchange, which give good NOE cross peaks. At higher temperature these broaden and their NOE cross peaks disappear thus allowing the signals in the lower-temperature spectrum to be identified as NOEs involving ligand protons. The binding site for trimetrexate is well defined and this was compared with the binding sites in related complexes formed with methotrexate and trimethoprim. No major conformational differences were detected between the different complexes. The 2,4-diaminopyrimidine-containing moieties in the three drugs bind essentially in the same binding pocket and the remaining parts of their molecules adapt their conformations such that they can make effective van der Waals interactions with essentially the same set of hydrophobic amino acids, the side-chain orientations and local conformations of which are not greatly changed in the different complexes (similar chi1 and chi2 values).

Synthesis and biological activities of conformationally restricted, tricyclic nonclassical antifolates as inhibitors of dihydrofolate reductases.

Seven novel tricyclic pyrimido[4,5-c][2,7]naphthyridones 5-8 and the corresponding naphthyridines 9-11 were synthesized as conformationally restricted inhibitors of dihydrofolate reductase (DHFR) and as antitumor and/or antiinfectious agents. The analogues were designed to orient the side chain trimethoxyphenyl group in different conformationally defined positions in order to explore the effect of the side chain orientation on binding affinity and selectivity for DHFR from various species. The semirigid orientations were achieved by bridging the C5 and N10 of compound 12 with a N-ethyl bridge and by variation of the position of double bonds in rings B and C as well as substitution at the 2',6'-positions of the phenyl ring. The synthesis of compounds 5-11 were accomplished by cyclocondensation of the appropriate keto ester (as the biselectrophile) with 2,4,6-triaminopyrimidine to afford the lactam 5. The dehydrolactams 6 and 7 were prepared by air oxidation and PtO2-catalyzed dehydrogenation of 7, respectively. The dichloro dehydro lactam 8 was obtained by refluxing lactam 5 and/or 6 in POCl3 or a mixture of POCl3/PCl5. Compounds 9-11 were obtained by two methods, direct borane reduction of lactam 5 or 6 or thiation of the dipivoylated lactam 15 followed by reductive dethiation. Compounds 9-11 were interconverted by air oxidation or PtO2-catalyzed reduction/oxidation, respectively. The compounds were evaluated as inhibitors of DHFR from Pneumocystis carinii (pc) and Toxoplasma gondii (tg) with rat liver (rl) serving as the reference mammalian enzyme. In the lactam series 5-8, the most unsaturated analogue 7 showed an IC50 of 86 nM against rlDHFR, almost 100-fold more active than 5 and 3-fold more active than 6. The 2',6'-dichloro dehydro lactam 8 was less active than the corresponding dehydro lactam 6 against rlDHFR. In the naphthyridine series 9-11, the dehydro analogue 10 was more active than 9 against rlDHFR. The fully reduced analogue 11 (as a mixture of cis and trans isomers) was the most active in the naphthyridine series. The analogues were, in general, more inhibitory against rlDHFR than against pcDHFR, or tgDHFR, and thus lacked selectivity. In addition, they were less potent than the bicyclic compounds trimetrexate 3 (TMQ) and piritrixim 4 (PTX).

Conformational analysis of the lipophilic antifolate trimetrexate.

The conformational properties of the lipophilic antifolate trimetrexate (TMQ) were calculated and compared to the structurally-analogous prototypical antifolate methotrexate (MTX) using both empirical force-field and AM1 quantum mechanical methods. The conformational preferences of TMQ and MTX are diametrically opposed with respect to the bridge-system set of torsion angles tau 1, tau 2: TMQ prefers gauche, trans while MTX prefers approximately trans, gauche. These predictions are consistent with the observed crystal structures of TMQ (i.e., tau 1 = 79 degrees, tau 2 = 178 degrees) and of DHFR-bound MTX (i.e., tau 1 = -157 degrees, tau 2 = 57 degrees in L. casei). The crystal structure of MTX.4H2O deviates from this pattern with tau 1 closer to cis (i.e., 39 degrees) than the predicted trans, yet this near-cis conformation is driven by intermolecular hydrogen-bonding and electrostatic forces operative in the MTX crystal. As a consequence of these strong intermolecular forces, MTX incurs 1.8 kcal/mole in conformational-strain energy in its crystalline form. In contrast, TMQ experiences virtually no conformational strain in its crystalline form. This disparity is attributed to two distinctions between TMQ and MTX: (i) MTX crystallizes as a zwitterion while TMQ crystallizes as the free base, and (ii) the hydrophilic glutamate tail in MTX is replaced by three lipophilic trimethoxy groups in TMQ. The corresponding conformational-strain energy of DHFR-bound MTX is 2.0 kcal/mole while that of DHFR-bound TMQ is only 0.65 kcal/mole based on the assumption that the latter adopts the same bridge conformation as the former. This cost in conformational-strain energy for TMQ and MTX is paid at the expense of their respective free energies of binding of DHFR. Consequently, the present study offers the possibility of designing a new class of antifolates which are conformationally strain-free when bound to DHFR and thereby more effective as chemotherapeutic agents.