Structural Biology in Drug Metabolism and Drug Discovery
Targeting metabolic pathways in microbial pathogens: oxidative stress and anti-folate drug resistance in trypanosomatids ¨ W.N. Hunter1 , M.S. Alphey, C.S. Bond and A.W. Schuttelkopf Division of Biological Chemistry and Molecular Microbiology, School of Life Sciences, University of Dundee, Dundee DD1 5EH, U.K.
Abstract The large quantity of genomic, biochemical and metabolic data on microbial pathogens provides information that helps us to select biological problems of interest and to identify targets, metabolic pathways or constituent enzymes, for therapeutic intervention. One area of potential use in developing novel antiparasitic agents concerns the regulation of oxidative stress, and we have targeted the trypanothione peroxidase pathway in this respect. In order to characterize this pathway, we have determined crystal structures for each of its components, and are now studying enzyme–ligand complexes of the first enzyme, trypanothione reductase. Also with regard to trypanosomatids, a question that arose was: why do antifolates not provide useful therapies? The enzyme pteridine reductase has been shown to contribute to anti-folate drug resistance, and we have determined the enzyme structure and mechanism to understand this aspect of drug resistance.
Introduction Our knowledge of microbial biology is expanding, greatly assisted by the availability of genomic data. Such research is important, given the urgent requirement for new treatments to combat protozoan infections. This contributes an improved understanding of how existing anti-microbial drugs work, and which metabolic pathways and component enzymes represent potential new drug targets; that is, which enzymes are essential for the survival of parasites or bacteria, which are absent in the human host or present markedly differing substrate specificity [1]. An enzyme can be validated as a target, genetically or chemically (in the latter case, with a reliance on already having a potent inhibitor), and then exploited by a structure-based approach [2,3] to the development of new therapies. We are engaged in studies targeting trypanosomatid pathways in two areas: (i) oxidative stress and redox regulation, and (ii) pterin/folate metabolism.
Oxidative stress and redox regulation by the trypanothione peroxidase pathway The medically important trypanosomatids, as with other organisms living in an aerobic environment, are exposed to reactive oxygen species, such as hydrogen peroxide. These potentially destructive chemicals are eliminated by a combination of antioxidant enzymes. In mammalian cells, Key words: oxidative stress, pteridine reductase, trypanothione reductase, trypanosomatid. Abbreviations used: GP, glutathione peroxidase; GR, glutathione reductase; TrX, thioredoxin; TrR, TrX reductase; TryR, trypanothione reductase; T[SH]2 , reduced trypanothione; TryX, tryparedoxin; TryP, TryX peroxidase; Gspd, glutathionyl spermidine disulphide; DHFR, dihydrofolate reductase; TS, thymidylate synthase; PTR1, pteridine reductase; DHB, 7,8-dihydrobiopterin; SDR, short-chain dehydrogenase/reductase; MTX, methotrexate. 1
To whom correspondence should be addressed (e-mail
[email protected]).
the principal route of hydrogen peroxide detoxification involves glutathione peroxidase (GP) working in concert with NADPH, glutathione and glutathione reductase (GR). In addition, hydroperoxide metabolism in mammals, yeasts, plants and Plasmodium falciparum also utilizes thioredoxin (TrX) peroxidase, with reducing equivalents provided from a pathway involving TrX, TrX reductase (TrR) and NADPH. Trypanosomatids on the other hand rely on a unique system, the trypanothione peroxidase pathway, to regulate oxidative stress. Details of this pathway have been elucidated in the model trypanosomatid Crithidia fasciculata ([4,5]; see Figure 1). NADPH provides reducing input to trypanothione reductase (TryR), which maintains high levels of reduced trypanothione {T[SH]2 ; N1 ,N8 -bis (glutathionyl)spermidine}, a polyamine–peptide conjugate. T[SH]2 reduces tryparedoxin (TryX), which then reduces TryX peroxidase (TryP). The peroxidase then catalyses the final reduction of peroxides to water or alcohols. A recurring theme is the use of sulphur redox chemistry to pass reducing equivalents through the pathway. Interest in this pathway stems from the following two reasons. Firstly, this is a well-defined pathway to regulate oxidative stress, and serves as a general model. Secondly, since trypanosomatids are susceptible to oxidative stress, the pathway and its components are putative targets for structure-based drug discovery programmes. To further the understanding of this pathway, we have determined crystal structures of each of its components ([5–7]; Figure 1).
TryR: discovery of a new class of enzyme inhibitor TryR is unique to trypanosomatids, where it subsumes the role of GR. It has also been validated as a drug target [8]. C 2003
Biochemical Society
607
608
Biochemical Society Transactions (2003) Volume 31, part 3
Figure 1 The trypanothione peroxidase pathway The subscripts ‘ox’ and ‘red’ are used to depict oxidized or reduced forms of the individual components respectively. Ribbon diagrams of each component (not to scale) are shown under each part of the pathway.
Our previous work explained why TryR and GR are mutually exclusive with respect to their cognate substrates [6]. We subsequently determined the structures of oxidized Trypanosoma cruzi TryR (where Cys53 and Cys58 form a redox active disulphide) in complex with two physiological substrates, Try[S]2 and glutathionyl spermidine disulphide (Gspd), and the structure of a Cys58 →Ser mutant that mimics the reduced enzyme. This mutant was reacted with Try[S]2 and Gspd to form a disulphide bridge with Cys53 , the complexes were purified, crystallized and the structures were determined. These structures represent intermediates, and provide a series of snapshots of catalysis (W.N. Hunter, M.S. Alphey, C.S. Bond and ¨ A.W. Schuttelkopf, unpublished work) and direct experimental evidence regarding substrate discrimination of TryR and GR. The TryR–Try[S]2 complex provided a template to search for novel inhibitors. We concentrated on the interaction of the polyamine component of Try[S]2 with the hydrophobic patch and Glu19 of the TryR active site, since this is where the major differences occur between TryR and GR. A search of the Cambridge Structural Database identified potential leads and two natural products, cadabacine and lunarine, were selected for investigation on the basis of three criteria. Firstly, they have the spermidine moiety bound at either end by an amide group: a feature shared with Try[S]2 . Secondly, they are cyclic and of comparable size with Try[S]2 . Thirdly, their size and chemistry indicated they would not inhibit human GR. Molecular modelling suggested that both compounds would be selective TryR inhibitors. This C 2003
Biochemical Society
hypothesis has been confirmed experimentally in Professor Alan Fairlamb’s laboratory (Department of Biochemistry, University of Dundee) for lunarine [6], which does not inhibit human GR under conditions that inhibit TryR by 97%. Unexpectedly, the inhibition pattern is both timeand concentration-dependent, suggesting the formation of a non-covalent Michaelis-type complex, and then a covalent complex between lunarine and reduced TryR. The modest K i of 144 µM for lunarine is encouraging compared with the binding constant of 40 µM for Try[S]2 , and allows scope for improvement. Disulphide oxidoreductases, such as TryR, share similarities in their mechanism with cysteine proteases. Both types of enzymes are susceptible to covalent Michael addition, exemplified by lunarine in the case of TryR and vinyl sulphones in the case of T. cruzi cruzain [9,10]. Vinyl sulphone derivatives might therefore prove a useful lead for inhibition of TryR and other disulphide oxidoreductases; in particular, Plasmodium falciparum TrR.
TryX and TryP TryX, with a molecular mass of approx. 16 kDa, is significantly larger than TrX (12 kDa). TryX has a Trp39 -CysPro-Pro-Cys43 sequence near the N-terminus resembling the TrX-type Trp-Cys-Gly(or Ala)-Pro-Cys active-site motif, in which the vicinal cysteine residues form the redox-active disulphide. TryX and TrX share 20% sequence identity, and have a similar fold. However, the relationship of secondary structure with the linear amino acid sequences is different for each protein, producing a distinctive topology and on
Structural Biology in Drug Metabolism and Drug Discovery
the surface of the proteins different shapes and charge distributions [7]. These features will contribute to the interactions with other components of the different pathways in which they participate. The reduced TryP forms a decameric assembly with 52 symmetry (Figure 1). Secondary structure topology places TryP along with TryX and GP in a distinct subgroup of the thioredoxin superfamily. TryP has N- and C-terminal Val-Cys-Pro motifs, and is classified as a two-cysteine peroxiredoxin [5]. The molecular details at the active site support ideas about the enzyme mechanism, and comparisons with an oxidized two-cysteine peroxiredoxin have revealed structural alterations induced by the change in oxidation state. These include a difference in quaternary structure from dimer (oxidized form) to decamer (reduced form). The decameric assembly may prevent indiscriminate oligomerization, localize ten peroxidase active sites and contribute to both the specificity of reduction by the redox partner TryX and the attraction of peroxides into the active site. In theory, the disruption of the T[SH]2 –TryX and TryX– TryP associations offers the potential to curtail the ability of trypanosomatid parasites to regulate oxidative stress in the same way as inhibition of TryR would. However, since neither TryX nor TryP have yet been validated as drug targets, we have not invested any time in looking at protein–ligand interactions, preferring instead to concentrate on TryR.
Pteridine reductase (PTR1) and anti-folate drug resistance in trypanosomatids Pterins and folates are important cofactors in a wide range of metabolic pathways, including DNA and RNA synthesis, the initiation of protein synthesis and amino acid metabolism [11]. These compounds are essential for cell growth, and enzymes associated with this aspect of metabolism, in particular dihydrofolate reductase (DHFR) and thymidylate synthase (TS), are of interest as drug targets for anti-bacterial and anti-neoplastic therapy, and the treatment of certain parasitic diseases, such as malaria [11]. Trypanosomatids are auxotrophic for folates and pterins, and rely on specific transporters to acquire these essential metabolites required in critical pathways, including nucleic acid and protein biosynthesis. This suggests that anti-folates should serve as useful drugs against trypanosomatid infection, and the fact that they fail to do so has attracted attention. Following uptake, unconjugated and conjugated pterins must undergo two successive reductions, yielding the active ‘tetrahydro’ species. Two enzymes carry out these reactions in trypanosomatids. A bifunctional DHFR/TS enzyme is the major one responsible for the reduction of folate and dihydrofolate to tetrahydrofolate, required for the synthesis of thymidylate [12–15]. The second enzyme is PTR1, which carries out successive reductions of both conjugated and unconjugated pterins. For example, biopterin is reduced to 7,8-dihydrobiopterin (DHB), and then subsequently to
5,6,7,8-tetrahydrobiopterin, or folate can be reduced to 7,8dihydrofolate and then to 5,6,7,8-tetrahydrofolate. PTR1, a short-chain dehydrogenase/reductase (SDR) family member [16,17], is the only enzyme known to reduce biopterin in Leishmania, and gene knockouts have shown that it is essential for growth in vitro. Several crystal structures of Leishmania major PTR1–ligand complexes have been completed. The ternary complex with NADP+ and substrate DHB provides details of protein–pterin interactions, and defines a two-step reduction mechanism. The first step resembles other SDR-family members, although an aspartate residue (Asp181 ) replaces a serine in the mechanism, with implications for the pH-dependence of the reaction [17]. The second reduction is similar to DHFR, and provides an example of convergent evolution of mechanism, where two distinct enzymes utilize different components of the binary complexes to provide similar chemical interactions in support of catalysis. The complex with the archetypal antifolate compound methotrexate (MTX) describes a mode of inhibition and, in conjunction with the DHB complex, the structural basis for the enzyme’s broad substrate specificity. MTX binds in the active site in a different orientation to that used by DHB, and comparisons reveal that MTX is a more potent inhibitor of DHFR than PTR1, because the p-aminobenzoic acid group makes more extensive interactions with DHFR. These structures therefore provide an explanation at the molecular level of how PTR1, by acting as a metabolic by-pass, compromises drugs targeting DHFR. We are now using a combination of crystallography, molecular modelling and isothermal titration calorimetry to characterize enzyme–ligand interactions to support the development of novel PTR1 inhibitors. PTR1 can catalyse the same reaction as DHFR, but is less susceptible to inhibition by typical anti-folates [12]. However, DHFR inhibitors may become valuable therapies if the PTR1 by-pass can be blocked. It is the combination of PTR1 and DHFR/TS, rather than the individual enzymes, that is important. The ultimate goal of our research is to identify one molecule that effectively inhibits both PTR1 and DHFR/TS or, and more likely, compounds specific for each enzyme that can be used synergistically. Funded by the Wellcome Trust. C.S.B. is a Biotechnology and Biological Sciences Research Council (BBSRC) Sir David Phillips Research Fellow. We thank David Gourley for his contributions, and our collaborators, Alan Fairlamb and Steve Beverley, and their respective laboratories, for advice, support and stimulating interaction. Daresbury Laboratory and the European Synchrotron Radiation Facility (ESRF) gave access to synchrotron facilities, where Miroslav Papiz and Gordon Leonard provided excellent support.
References 1 Wang, C.C. (1997) Parasitology 114, 3–16 2 Kuntz, I.D. (1992) Science 257, 1078–1082 3 Verlinde, C.L.M.J. and Hol, W.G.J. (1994) Structure 2, 577–587 C 2003
Biochemical Society
609
610
Biochemical Society Transactions (2003) Volume 31, part 3
4 Hofmann, B., Hecht, H.-J. and Flohe, ´ L. (2002) Biol. Chem. 383, 347–364 5 Alphey, M.S., Bond, C.S., Tetaud, E., Fairlamb, A.H. and Hunter, W.N. (2000) J. Mol. Biol. 300, 903–916 6 Bond, C.S., Zhang, Y., Berriman, M., Cunningham, M., Fairlamb, A.H. and Hunter, W.N. (1999) Structure 7, 81–89 7 Alphey, M.S., Leonard, G.A., Gourley, D.G., Tetaud, E., Fairlamb, A.H. and Hunter, W.N. (1999) J. Biol. Chem. 274, 25613–25622 8 Tovar, J., Wilkinson, S., Mottram, J.C. and Fairlamb, A.H. (1998) Mol. Microbiol. 29, 653–660 9 Selzer, P.M., Pingel, S., Hsieh, I., Ugele, B., Chan, V.J., Engel, J.C., Bogyo, M., Russell, D.G., Sakanari, J.A. and McKerrow, J.H. (1999) Proc. Natl. Acad. Sci. U.S.A. 96, 11015–11022 10 Brinen, L.S., Hansell, E., Cheng, J.M., Roush, W.R., McKerrow, J.H. and Fletterick, R.J. (2000) Structure 8, 831–840 11 Then, R.L., Hartman, P.G., Kompis, I. and Santi, D. (1993) Chemistry and biology of pteridines and folates (Ayling, J.E., Nair, M.G. and Baugh, C.M., eds.), pp. 533–536, Plenum Press, New York
C 2003
Biochemical Society
12 Nare, B., Luba, J., Hardy, L. and Beverley, S.M. (1997) Parasitology 114, 101–110 13 Nare, B., Hardy, L.W. and Beverley, S.M. (1997) J. Biol. Chem. 272, 13883–13891 14 Carreras, C.W. and Santi, D.V. (1995) Annu. Rev. Biochem. 64, 721–762 15 Knighton, D.R., Kan, C.-C., Howland, E., Janson, C.A., Hostomska, Z., Welsh, K.M. and Matthews, D.A. (1994) Nat. Struct. Biol. 1, 1816–194 16 Hardy, L.W., Matthews, W., Nare, B. and Beverley, S.M. (1997) Exp. Parasitol. 87, 157–169 17 Gourley, D.G., Schuttelkopf, ¨ A.W., Leonard, G.A., Luba, J., Hardy, L.W., Beverley, S.M. and Hunter, W.N. (2001) Nat. Struct. Biol. 8, 521–525
Received 27 January 2003
