CHAPTER 6
SUMMARY, CONCLUSIONS AND PERSPECTIVES NEDERLANDSE SAMENVATTING
Chapter 6
Summary, conclusions and perspectives
6. Summary Several drugs are biotransformed to active metabolites that can significantly contribute to their overall pharmacological or adverse effects. Tracking active metabolites is not only important to correctly interpret the pharmacological and/or adverse effects in preclinical studies but may also be used as a promising tool to identify potentially new drug candidates for drug discovery and development (1). In drug discovery and development it is important to have information on drug metabolism as early as possible. Knowledge of the metabolic pathways, metabolic stability, toxicity and the specific enzymes involved in the metabolism are all important information in the drug discovery and development process but also in planning human clinical studies. In chapter 1 the evolving role of drug metabolism in drug discovery and development is discussed. Because of the impact of biotransformation reactions on the fate and the effects of drugs, the preparative synthesis of drug metabolites is currently of primary importance in industry in order to assess potential pharmacological activities, toxicity, drug-drug interactions and to examine metabolic pathways (2). Full pharmacological assessment of biotransformation products usually requires organic synthesis to obtain sufficient amounts of pure compound. As an alternative to organic synthesis, biosynthetical approaches are upcoming methodologies. Biocatalysts (i.e. enzymes) can perform reactions with a high degree of regio- and stereoselectivity that often cannot be achieved by organic synthesis (3). Bacterial Cytochrome P450 BM3 appears as a powerful and versatile tool for the generation of high amounts of human relevant drug metabolites that can be isolated, identified and tested for toxicity and pharmacological activity. By rational re-design and directed evolution this enzyme can be engineered to catalyze reactions that are mimicking human P450s or to acquire completely novel catalytic properties (shown in this thesis). In recent years, P450 BM3 has emerged not only as an promising tool for the generation of drug metabolites and commercial products but also as an excellent model system to study general mechanistic aspects of P450 chemistry. The main aim of this thesis is to contribute to the development of a metabolic production and profiling platform encompassing the application of P450 BM3 mutants for the generation of physiologically relevant drug metabolites and the application of screening methods for their identification and pharmacological and toxicological characterization. Moreover it comprises mechanistic insights into P450 catalysis and chemistry and the geometry of the active site. The general strategy applied (Introduction, Figure 11) comprises genetic engineering of P450 BM3 mutants by different mutation strategies (site-directed, site-saturation, random mutagenesis), screening of mutant libraries (UPLC, fluorescent assay, LC-MS, GSH-
Chapter 6
Summary, conclusions and perspectives
trapping, HRS), upscaling and isolation of metabolites by Prep-LC, and structural 1 elucidation of isolated metabolites by H-NMR. One of the BM3 mutants which had been developed previously in our Molecular Toxicology group and which showed high activity in drug metabolism is BM3 M11, containing ten different amino acid substitutions compared to wild-type BM3. This BM3 M11 mutant was shown to be highly active in metabolizing a variety of drugs (e.g. clozapine, testosterone, MDMA, dextromethorphan, diclofenac) to human relevant metabolites (4-6), including reactive intermediates (7-9). In the present thesis, we have performed a saturation mutagenesis study in which the active-site residue at position 87 in BM3 M11 was mutated to all 20 possible amino acids. It was demonstrated that the type of amino acid at this position has strong effect on substrate selectivity when comparing a series of alkoxyresorufins (chapter 2), on the activity and regioselectivity of testosterone hydroxylation (chapter 2) and on the activity and regioselectivity of clozapine bioactivation (chapter 3). Twelve of the amino acid substitutions were not yet been reported previously in any BM3 variant. In chapter 2, a series of nine alkoxy-substituted substrates (methoxyresorufin to n-octoxyresorufin, and benzyloxyresorufin) were tested as diagnostic substrates. It was shown that mutation at position 87 dramaticaly affected not only the substrate selectivity but also the coupling efficiency of the enzyme. Interestingly, the coupling efficiency with these substrates was always less than 1% for all productive enzymes suggesting that alkoxyresorufins bind to BM3 at the active site mainly in a nonproductive orientation. Uncoupling of P450 is still a poorly understood process. Because the BM3 mutants contain amino acids at position 87 with different polarities and size, different modes of uncoupling might underly the high NADPH-consumptions observed. The mechanism by which alkoxyresorufine stimulated extremely high NADPH-consumption in the position 87 mutants and wild-type P450 BM3 therefore still remains to be elucidated. Testosterone was hydroxylated by the library of twenty mutants at position 87, at three different positions, as was shown previously in incubations with the triple mutant of P450 BM3, containing mutations R47L, F87V and L188Q (5). Structural identification of the metabolites by NMR revealed that two of the metabolites result from hydroxylation of the D-ring, at positions 15ß and 16ß; the third metabolite results from hydroxylation of the A-ring at position 2ß. With the triple mutant very poor regioselectivity was observed. With this library of mutants instead, big changes in metabolic profile were observed. For example, the mutant containing isoleucine at position 87 catalyzed 16ß-hydroxylation with very high selectivity whereas in case of the closely related leucine aminoacid testosterone hydroxylation was taking place predominantly at the position 2ß. Mutant M11 V87I is the first bacterial P450 able to selectively hydroxylate testosterone at position 16ß. Why these relatively small changes in
Chapter 6
Summary, conclusions and perspectives
amino acid side chain have such a large effect on regioselectivity remains to be established. In chapter 3 the site-saturation library of BM3 mutants at position 87 presented in chapter 2 was applied for the generation of reactive metabolites of clozapine. While in chapter 2 the library was screened only with non-charged substrates, in chapter 3 a positively charged molecule, i.e. clozapine, was evaluated. Clozapine is known to be involved in severe ADRs due to the formation of a reactive metabolite (10). Often reactive metabolites cannot be synthetised by organic chemistry, thus complicating the availability of reference compounds and standards. Also, the amounts of reactive metabolites formed by human P450s are generally too low to allow purification and NMR identification. In chapter 3, it was investigated whether BM3-mutants at position 87 could be applied for the generation of reactive metabolites in quantities allowing chemical characterization. Results showed that the nature of aminoacid at position 87 strongly determines both activity and regioselectivity of clozapine metabolism. The mutant containing Phe87 showed both high activity and high selectivity for the bioactivation pathway and was used for the large scale production of GST-dependent GSH conjugates by incubation in presence of glutathions S-transferase P1-1. Five human relevant GSH conjugates were produced in 1 high amounts enabling structural characterization by H-NMR. This results demonstrated the applicability of P450 BM3 mutants for the generation of human relevant reactive metabolites in sufficient amount to allow their structural elucidation by NMR. Detection and structural elucidation of reactive metabolites early in drug development is very critical for the development of safer drugs. In chapter 4, site-directed mutagenesis was applied to improve the regioselectivity of steroid hydroxylation by BM3 mutants. The strategy applied encompassed the restriction of the actve site size by mutating Ala82 with a Trp in order to reduce the substrate mobility, therefore improving the regioselectivity of steroid hydroxylation. The mutation A82W led to a < 42-fold increase in Vmax for 16hydroxylation of testosterone and norethisterone, and improved the coupling efficiency of the enzyme by a more efficient exclusion of water from the active site. Spin relaxation NMR was applied to rationalize the change in metabolic profile observed, showing that the mutation caused a change in the orientation of testosterone in M11 A82W as compared to the orientation in M11. In chapter 5, mutants of P450 BM3 were used to support drug development by producing human relevant drug metabolites analyzed for identity and bioaffinity assessment by the analytical high-resolution screening (HRS). A panel of BM3 mutants was applied for the generation of metabolic mixtures of TAK-715, a known p38 inhibitor. HRS screening allowed the identification and bioaffinity determination of all the metabolites produced. The high turnover rates of BM3 M11 and the convenient large scale production
Chapter 6
Summary, conclusions and perspectives
and purification protocols for this enzyme allowed semi-preparative production of the most abundant active metabolites that have been identified by NMR, while HRS allowed the determination of their IC50. These results showed that the combination of a catalytically diverse set of P450 BM3 mutants as a toolbox to diversify drugs and a HRSsystem capable to rapidly screen for affinity to p38-kinase might be a promising platform to generate potential novel lead compounds which might have improved physico-chemical properties (by improved water solubility) and desired pharmacological properties. 6.1. Conclusions and perspectives In this research, different mutagenesis techniques have been applied for the development of P450 BM3 mutants that are able to metabolize drug molecules to human relevant metabolites and that could be used as biocatalysts in drug discovery and synthesis. P450 mutants able to produce high amounts of drug metabolites are very useful biocatalysts to generate metabolites from novel drug candidates for structural, toxicological and pharmacological characterization. Also they can be used for the biosynthesis of known pharmacologically active compounds. In drug discovery, a library of diverse mutant P450s could be used to functionalize lead compounds in order to identify potential novel drugs and drug candidates. When the research described in this thesis started, in 2008, the so-called Metabolites in Safety Testing (MIST) guidelines were published by the FDA, leading several research groups to focus their efforts on the development of new biocatalysts for the generation of large amounts of human relevant drug metabolites (11). At that time the work of Van Vugt-Lussenburg et al. (4, 5, 12, 13) already had shown that by site-directed mutagenesis of specific amino acids in the active site of wild-type P450 BM3, the substrate spectrum could be expanded to accept drugs and drug-like molecules. Damsten et al. (8) showed that P450 BM3 mutants could also be applied for the generation of reactive drug metabolites, however structural elucidation of these metabolites was still hampered by the small amounts obtained. BM3 mutants that were able to hydroxylate steroids were already developed (5), however with very poor regioand steroselectivity. The platform presented in the introduction of this thesis (Chapter 1, Figure 11) has been applied for the regioselective steroid hydroxylation (chapter 2 and 4), for the generation of potentially toxic reactive metabolites (chapter 3), for the generation of bioactive drug metabolites (chapter 5). 6.1.1. Regioselective hydroxylation of steroids by P450 BM3 mutants The present thesis has shown that one of the current challenges in synthetic organic chemistry, namely the control of regio- and stereoselective oxidation of unactivated C-H bond of complex organic compounds (14-16) can be met by engineering P450 BM3 enzymes.
Chapter 6
Summary, conclusions and perspectives
In chapter 2 and chapter 4, it was shown that by single mutation in the active site of P450 BM3 (position 87 and position 82) dramatic changes in regio- and stereoselectivity of steroid hydroxylation could be obtained. Spin relaxation NMR was used as powerful tool to determine changes in orientation of testosterone in the active site of BM3 mutants showing different regioselectivity. The regio- and stereoselectivity of P450-mediated reactions depends upon the orientation of the substrate relative to the reactive iron-oxo species, which is in turn determined by the active-site configuration of the P450 enzyme (17). Small variations in the active site of P450s could alter or improve their substrate scope, regio- or stereoselectivity, activity and coupling efficiency (18). Interestingly, recently Venkatamaran et al. showed that a single active site mutation S72I in M01 A82W and in M11 V87I inverted the stereoselectivity of hydroxylation from 16 β to 16 α (19). Structure guided redesign of the active site can be used to manipulate the regioselectivity of P450 enzymes to obtain mutants able to target specific position in the steroid molecule, therefore creating novel hydroxysteroids. Spin-relaxation NMR, docking and molecular dynamics can be used as powerful tools to shed light on the origin of regio- and stereoselectivity. 6.1.2. Biosynthesis of reactive metabolites by P450 BM3 mutants Measuring the potential bioactivation of drugs and drug candidates leading to chemically reactive metabolites early in the drug discovery phase is often hampered by the difficulties in detecting and characterizing low levels of RIs (20). When this research started, Damsten et al. already showed that BM3 mutants were able to produce reactive metabolites from the drugs clozapine, diclofenac and acetominophen (8). However, the amounts obtained were still not sufficient for isolation and structural elucidation by NMR. Recently, Dragovic et al. (9) identified novel human relevant GST-dependent GSHconjugates for which unequivocal structural elucidation by NMR was still missing. Boerma et al. showed that P450 BM3 mutants with high capacity to activate drugs (clozapine, acetominophen and troglitazone) into relevant reactive metabolites can be employed to produce protein adducts to study the nucleophilic selectivity of highly reactive electrophiles (21). In chapter 3, by using BM3 mutant M11V87F, we were able to produce significant amounts of all major human relevant GSH conjugates of clozapine, for which the 1 structures were not yet elucidated unequivocally by H-NMR. This study confirmed the high potential of BM3 mutants as tool to assist the identification of reactive metabolites of drugs, the elucidation of novel bioactivation pathways and to generate high amounts of drug metabolites allowing the isolation of mg amounts of pure 1 reactive metabolites for their full structural elucidation by H-NMR. At current, predicting the potential of new chemical entities to generate IDRs is still not possible because of the lack of reliable pre-clinical models. However the formation of reactive metabolites and protein covalent binding are perceived as significant risk factors.
Chapter 6
Summary, conclusions and perspectives
In vitro screening tools for the generation and detection of reactive metabolites presented in chapter 3 can be applied as novel tools in the development of safer drugs. 6.1.3. Biosynthesis of active metabolites by P450 BM3 mutants Full characterization of metabolite profiles and elucidation of metabolite structures early in drug development is often hampered by difficulties in producing sufficient amounts of metabolites and in detecting active metabolites in a rapid and efficient way (22). The first problem is tackled by the application of highly active P450 BM3 mutants for the generation of high amounts of drug metabolites. The second problem is solved by the application of the High Resolution Screening (HRS), which allows structural characterization and bioaffinity determination of the metabolites formed in hyphenated mode (23). In chapter 5 an integrated strategy encompassing the use of BM3 mutants for generation of metabolic mixtures and the structural identification and bioaffinity assessment of metabolites with the HRS platform is presented. The combination of n hyphenated screening, MS analysis and NMR spectroscopy enabled full structure elucidation (except stereochemistry) and affinity determination of all the biotransformation products synthesized in semi-preparative amounts. The panel of BM3 mutants presented is highly suitable to be used in the drug development process as general reagents for lead diversification. The multidimensional screening approach described here truly adds valuable information to the more conventional chemical analysis methods. In a fast and efficient way, data was generated on both structure and biological activity of the metabolites formed. Furthermore, linking structural modifications by metabolism to changes in drug target affinity might be efficient tool to construct the pharmacophore model next to the more conventional medicinal chemistry approach of synthesizing structural analogues of lead compounds. 6.2. Future perspectives In this thesis we describe the development of a metabolite production and profiling platform, where P450 BM3 enzymes are engineered to mimic human P450s or to acquire novel catalytic properties, based on the metabolism of diagnostic substrates; libraries of site-directed, site-saturation or random mutants are then screened to select the best mutant for upscaling and production of large amounts of individual metabolites, that are purified by Prep-LC, allowing their structural elucidation by NMR and their pharmacolological evaluation. However, there are some unresolved issues that should be addressed in further research. The presented examples of P450 BM3 engineering amply demonstrate that the activity of this enzyme can be “tamed” for particular applications. Much has been made for the enzyme’s suitability in commercial-scale applications, and the realization of this goal is one of the researchers’ priorities. However the space-time yields and overall productivity of
Chapter 6
Summary, conclusions and perspectives
cultures still need to be extended considerably for industrial applications (24). Enhancing the longevity of the enzyme is one of the goals to achieve: total turnover numbers are limited by a range of factors, including the stability of the variants employed, intrinsic activity levels, and the response of the enzyme to specific substrates over extended periods of turnover, including the inhibition rates associated with each product (24). With non-natural substrates, too little is currently known about exit channels and product release, and whether degradation is due to denaturation or co-factor loss, heme modification or other factors. The commercial demands for enzymes that are functional in non-natural environments (elevated temperature, nonnative pH, high substrate and product concentrations, and organic solvents) are current challenges that have been only partially met. Moreover, the incubations performed in this study still required the addition of the expensive cofactor NADPH, which is not desirable for large scale incubations. This could be solved by performing whole-cell incubations, under non-lytic conditions, in order to allow the use of the endogenous NADPH supply of the cells. To perform whole-cell incubations, several issues need to be addressed: for example, many organic compounds do not readily cross cell membranes; this issue could be tackled adding permeability enhancing agents (e.g. EDTA), or by selecting mutant host strains with greater permeability, or by expressing BM3 on cell surface (25). Moreover, substrate or reaction products can be toxic for the host and over-oxidation can lead to the formation of many secondary or tertiary metabolites. A lot of research is currently done in the field of alternative oxidants, such as peroxides, metal powders and metal electrodes, to acquire simple, cost-effective and efficient ways of performing CYP-mediated reactions (26). Rational and directed evolution approaches have been successfully applied to engineer BM3 enzymes for enhanced activity, stability and expression in E.coli, as well as for altered substrate specificity and regio- and steroselectivity. However, the ability of BM3 to adopt new functions has mechanistic underpinnings that have yet to be fully elucidated. Recent advances in enzyme engineering have used a combination of random methods of directed evolution with elements of rational enzyme modification to successfully by-pass certain limitations of both directed evolution and rational design (27). Semi-rational approaches targeting multiple, specific residues to create “smart libraries” have been very successful (28, 29). Combinatorial alanine substitution has been successfully applied to generate P450 BM3 variants active with large substrates (30). Mutagenesis with un-natural aminoacids or insertion of cysteine residues that can be subsequently alkylated with different alkylating agents in the active site can be used to further improve the properties of the enzyme, opening up new chemistry not available with the standard twenty aminoacids (27). The synergy between computational and experimental BM3 research led to important mechanistic insights into the origin of regio- and stereoselectivity: recently, de Beer et al. successfully applied free energy calculations to get insight into the stereoselective
Chapter 6
Summary, conclusions and perspectives
hydroxylation of -ionones by engineered BM3 mutants (31). Moreover the role of protein plasticity in molecular dynamics simulation aimed at rationalize the regioselectivity in testosterone hydroxylation by BM3 mutants has been extensively studied (32). In conclusion, P450 BM3 has been presented as a valuable, versatile tool to support drug development by producing drug metabolites in such amounts that toxicological and pharmacological characterization is possible. Mutagenesis techniques are efficient tools to tailor the enzyme activity for a wide variety of applications. The biocatalytic potential of P450 BM3 mutants to generate human relevant and novel unique drug metabolites was demonstrated and these mutants were successfully used for the generation and structural characterization of reactive metabolites, for the functionalization of lead molecules and for the regioselective hydroxylation of steroid compounds. The combination of BM3 biosynthesis and HRS screening is a highly valuable platform to identify potential new lead compounds and to assess pharmacological properties of drug metabolites. The platform presented in this thesis can be applied in early stage drug discovery to expand the toolbox of the medicinal chemist for the generation and optimization of lead compounds and to detect potentially dangerous reactive metabolites for the development of safer drugs.
Chapter 6
Summary, conclusions and perspectives
References 1. Guanaratna, C. (2000) Drug Metabolism & Pharmacokinetics in Drug Discovery: A primer for Bioanalytical Chemists, Part I, Current Separations 19, 17-23. 2. Schroer, K., Kittelmann, M., and Lutz, S. (2010) Recombinant human cytochrome P450 monooxygenases for drug metabolite synthesis, Biotechnol Bioeng 106, 699-706. 3. Clouthier, C. M., and Pelletier, J. N. (2012) Expanding the organic toolbox: a guide to integrating biocatalysis in synthesis, Chemical Society reviews 41, 15851605. 4. Lussenburg, B. M., Babel, L. C., Vermeulen, N. P., and Commandeur, J. N. (2005) Evaluation of alkoxyresorufins as fluorescent substrates for cytochrome P450 BM3 and site-directed mutants, Anal Biochem 341, 148-155. 5. van Vugt-Lussenburg, B. M., Damsten, M. C., Maasdijk, D. M., Vermeulen, N. P., and Commandeur, J. N. (2006) Heterotropic and homotropic cooperativity by a drug-metabolising mutant of cytochrome P450 BM3, Biochem Biophys Res Commun 346, 810-818. 6. Reinen, J., Kalma, L. L., Begheijn, S., Heus, F., Commandeur, J. N., and Vermeulen, N. P. (2011) Application of cytochrome P450 BM3 mutants as biocatalysts for the profiling of estrogen receptor binding metabolites of the mycotoxin zearalenone, Xenobiotica; the fate of foreign compounds in biological systems 41, 59-70. 7. Damsten, M. C., de Vlieger, J. S., Niessen, W. M., Irth, H., Vermeulen, N. P., and Commandeur, J. N. (2008) Trimethoprim: novel reactive intermediates and bioactivation pathways by cytochrome p450s, Chemical research in toxicology 21, 2181-2187. 8. Damsten, M. C., van Vugt-Lussenburg, B. M., Zeldenthuis, T., de Vlieger, J. S., Commandeur, J. N., and Vermeulen, N. P. (2008) Application of drug metabolising mutants of cytochrome P450 BM3 (CYP102A1) as biocatalysts for the generation of reactive metabolites, Chem Biol Interact 171, 96-107. 9. Dragovic, S., Boerma, J. S., van Bergen, L., Vermeulen, N. P., and Commandeur, J. N. (2010) Role of human glutathione S-transferases in the inactivation of reactive metabolites of clozapine, Chem Res Toxicol 23, 1467-1476. 10. Fischer, V., Haar, J.A., Greiner, L., Lloyd, R.V., and Mason, R.P. (1991) Possible role of free radical formation in clozapine (Clorazil) induced agranulocytosis., Mol. Pharmacol. 40, 846-853. 11. Nedderman, A. N. (2009) Metabolites in safety testing: metabolite identification strategies in discovery and development, Biopharmaceutics & drug disposition 30, 153-162. 12. van Vugt-Lussenburg, B. M., Stjernschantz, E., Lastdrager, J., Oostenbrink, C., Vermeulen, N. P., and Commandeur, J. N. (2007) Identification of critical residues in novel drug metabolizing mutants of cytochrome P450 BM3 using random mutagenesis, J Med Chem 50, 455-461. 13. Stjernschantz, E., van Vugt-Lussenburg, B. M., Bonifacio, A., de Beer, S. B., van der Zwan, G., Gooijer, C., Commandeur, J. N., Vermeulen, N. P., and Oostenbrink, C. (2008) Structural rationalization of novel drug metabolizing mutants of cytochrome P450 BM3, Proteins 71, 336-352.
Chapter 6
14. 15. 16. 17.
18. 19.
20.
21.
22.
23.
24. 25.
26. 27.
28.
Summary, conclusions and perspectives
Newhouse, T., and Baran, P. S. (2011) If C-H bonds could talk: selective C-H bond oxidation, Angew Chem Int Ed Engl 50, 3362-3374. Chen, M. S., and White, M. C. (2007) A predictably selective aliphatic C-H oxidation reaction for complex molecule synthesis, Science 318, 783-787. Gaich, T., and Baran, P. S. (2010) Aiming for the ideal synthesis, The Journal of organic chemistry 75, 4657-4673. Zhang, K., El Damaty, S., and Fasan, R. (2011) P450 fingerprinting method for rapid discovery of terpene hydroxylating P450 catalysts with diversified regioselectivity, J Am Chem Soc 133, 3242-3245. Jung, S. T., Lauchli, R., and Arnold, F. H. (2011) Cytochrome P450: taming a wild type enzyme, Curr Opin Biotechnol 22, 809-817. Venkataraman, H., Beer, S. B., Bergen, L. A., Essen, N., Geerke, D. P., Vermeulen, N. P., and Commandeur, J. N. (2012) A single active site mutation inverts stereoselectivity of 16-hydroxylation of testosterone catalyzed by engineered cytochrome P450 BM3, Chembiochem 13, 520-523. Smith, D. A., and Obach, R. S. (2009) Metabolites in safety testing (MIST): considerations of mechanisms of toxicity with dose, abundance, and duration of treatment, Chemical research in toxicology 22, 267-279. Boerma, J. S., Vermeulen, N. P., and Commandeur, J. N. (2011) Application of CYP102A1M11H as a tool for the generation of protein adducts of reactive drug metabolites, Chemical research in toxicology 24, 1263-1274. Fura, A., Shu, Y. Z., Zhu, M., Hanson, R. L., Roongta, V., and Humphreys, W. G. (2004) Discovering drugs through biological transformation: role of pharmacologically active metabolites in drug discovery, J Med Chem 47, 43394351. Falck, D., de Vlieger, J. S., Niessen, W. M., Kool, J., Honing, M., Giera, M., and Irth, H. (2010) Development of an online p38alpha mitogen-activated protein kinase binding assay and integration of LC-HR-MS, Analytical and bioanalytical chemistry 398, 1771-1780. Whitehouse, C. J., Bell, S. G., and Wong, L. L. (2012) P450(BM3) (CYP102A1): connecting the dots, Chemical Society reviews 41, 1218-1260. Yim, S. K., Kim, D. H., Jung, H. C., Pan, J. G., Kang, H. S., Ahn, T., and Yun, C. H. (2010) Surface display of heme- and diflavin-containing cytochrome P450 BM3 in Escherichia coli: a whole cell biocatalyst for oxidation, J Microbiol Biotechnol 20, 712-717. Kumar, S. (2010) Engineering cytochrome P450 biocatalysts for biotechnology, medicine and bioremediation, Expert Opin Drug Metab Toxicol 6, 115-131. Chica, R. A., Doucet, N., and Pelletier, J. N. (2005) Semi-rational approaches to engineering enzyme activity: combining the benefits of directed evolution and rational design, Curr Opin Biotechnol 16, 378-384. Seifert, A., Vomund, S., Grohmann, K., Kriening, S., Urlacher, V. B., Laschat, S., and Pleiss, J. (2009) Rational design of a minimal and highly enriched CYP102A1 mutant library with improved regio-, stereo- and chemoselectivity, Chembiochem 10, 853-861.
Chapter 6
29.
30.
31.
32.
Summary, conclusions and perspectives
Seifert, A., Antonovici, M., Hauer, B., and Pleiss, J. (2011) An efficient route to selective bio-oxidation catalysts: an iterative approach comprising modeling, diversification, and screening, based on CYP102A1, Chembiochem 12, 1346-1351. Lewis, J. C., Mantovani, S. M., Fu, Y., Snow, C. D., Komor, R. S., Wong, C. H., and Arnold, F. H. (2010) Combinatorial alanine substitution enables rapid optimization of cytochrome P450BM3 for selective hydroxylation of large substrates, Chembiochem 11, 2502-2505. de Beer, S. B., Venkataraman, H., Geerke, D. P., Oostenbrink, C., and Vermeulen, N. P. (2012) Free Energy Calculations Give Insight into the Stereoselective Hydroxylation of alpha-Ionones by Engineered Cytochrome P450 BM3 Mutants, Journal of chemical information and modeling. de Beer, S. B., van Bergen, L. A., Keijzer, K., Rea, V., Venkataraman, H., Guerra, C. F., Bickelhaupt, F. M., Vermeulen, N. P., Commandeur, J. N., and Geerke, D. P. (2012) The role of protein plasticity in computational rationalization studies on regioselectivity in testosterone hydroxylation by cytochrome P450 BM3 mutants, Current drug metabolism 13, 155-166.
Samenvatting
SAMENVATTING Cytochromen P450 (P450 of CYP) vertegenwoordigen een grote superfamilie van hemebevattende monoxygenases die in nagenoeg alle organismen voorkomen. Bij de mens zijn P450's betrokken bij de biotransformatie (metabolisme) van 80% van de geneesmiddelen op de markt. Metabolieten die door P450 worden geproduceerd kunnen farmacologische activiteit vertonen of verantwoordelijk zijn voor de toxiciteit of andere ongewenste bijwerkingen van geneesmiddelen of geneesmiddelkandidaten. Het is om deze redenen dat het sinds enkele jaren verplicht is voor geneesmiddel registratie om de biologische eigenschappen van de belangrijkste metabolieten te karakteriseren. Er is daarom een grote behoefte aan systemen waarmee deze metabolieten in voldoende hoeveelheden kunnen worden geproduceerd zodat hun farmacologische en toxicologische eigenschappen in detail kunnen worden gekarakteriseerd. Tot nu toe werd productie van metabolieten meestal uitgevoerd door organische synthese of door grootschalige incubaties met menselijke P450's die zijn verkregen door heterologe expressie in E.coli, gist, insectcellen of zoogdiercellijnen. Echter, vanwege hun instabiliteit en intrinsiek lage activiteit is het rendement van metaboliet productie door menselijke P450's vaak erg laag en de kosten zeer hoog. In vergelijking met hun menselijke tegenhangers vertonen microbiële P450's over het algemeen een veel hogere stabiliteit en specifieke activiteit. Één van de meest bestudeerde microbiële P450's is cytochrome P450 BM3 (CYP102A1) van Bacillus megaterium. Bij dit microbiële P450 is het heme domein en het reductase domein gefuseerd in een enkele polypeptideketen waardoor het electronentransport van de cofactor NADPH naar het katalytische centrum zeer efficient is. Dit in tegenstelling tot de P450's van zoogdieren waar het heme domein en reductase domein als aparte membraangebonden eiwitten voorkomen. P450 BM3 is een zeer stabiel en oplosbaar enzym en is het meest actieve P450 dat tot nu toe in de natuur is gevonden. Omdat dit enzym op grote schaal kan worden geëxpresseerd in E.coli, en daaruit gemakkelijk kan worden gezuiverd, heeft P450 BM3 grote perspectieven voor de toepassing als biokatalysator voor metaboliet productie op grote schaal. Het feit dat er verschillende kristalstructuren van P450 BM3 met substraten zijn opgehelderd maakt het tevens mogelijk de substraatselectiviteit van dit enzym met gerichte site-directed mutagenese te manipuleren. Het in dit proefschrift beschreven onderzoek is uitgevoerd in de context van het Top Instituut Pharma-project 'MetStab' (D2-102). Het belangrijkste doel van het onderzoek was het leveren van een bijdrage aan de ontwikkeling van een platform waarmee, gebruik makend van een verzameling van hoog-actieve en katalytisch diverse BM3 mutanten en innovatieve on-line hoge resolutie screeningsmethodes, op een zeer efficiente manier metabolieten van geneesmiddelen of lead compounds kunnen worden geproduceerd, geidentificeerd en geprofileerd met betrekking tot hun affiniteit voor specifieke drug targets. Bovendien beoogde het onderzoek het verkrijgen van meer
Samenvatting
mechanistisch inzicht in de werking van de P450 en de geometrie en de topologie van de substraat bindingsplaats. In dit proefschrift zijn verschillende mutagenese technieken beschreven waarmee de enzymactiviteit van P450 BM3 kan worden gestuurd in de richting van verschillende toepassingen, zoals bioactivatie van geneesmiddelen tot reactieve metabolieten, stereoselectieve hydroxylering van steroïdes en productie van farmacologisch actieve geneesmiddel metabolieten. In hoofdstuk 1 wordt bediscussieerd waarom in de afgelopen jaren steeds meer aandacht wordt besteed aan de rol van metabolisme bij de ontdekking en ontwikkeling van nieuwe geneesmiddelen. Naast het feit dat metabolisme in belangrijke mate de farmacokinetiek van geneesmiddelen bepaalt, worden in dit hoofdstuk verschillende voorbeelden beschreven van geneesmiddelen die worden gemetaboliseerd tot farmacologisch zeer potente metabolieten die aanzienlijk bijdragen tot de algehele farmacologische werking van de geneesmiddelen. Anderzijds kan de biologische activiteit van metabolieten verantwoordelijk zijn voor toxiciteit of andere ongewenste bijwerkingen van geneesmiddelen. Karakterisering van actieve metabolieten is daarom van groot belang om zowel de farmacologische als toxicologische effecten beter te kunnen voorspellen in preklinische en klinische studies. Het feit dat metabolieten soms veel potenter zijn dan de uitgangsstof maakt dat structurele modificatie van lead compounds door middel van metabolisme ook een veelbelovend instrument kan zijn in het drug discovery proces, naast de meer conventionele benadering van organische synthese van structurele analoga. Controle van de regio-en stereoselectiviteit van de hydroxylering van nietgeactiveerde CH bindingen in complexe organische verbindingen is een van de uitdagingen in de synthetische organische chemie. In dit proefschrift is aangetoond dat genetisch gemodificeerde P450 BM3 enzymen in staat zijn steroides regio- en stereospecifiek te hydroxyleren. Door mutaties aan te brengen op specifieke posities in de substraat bindingsplaats van P450 BM3 werden grote veranderingen in regio-en stereoselectiviteit van hydroxylering van testosteron verkregen. In hoofdstuk 2 werd de rol van het aminozuur op positie 87 onderzocht, dat zich in de onmiddelijke nabijheid van het katalytische centrum van P450 BM3 bevindt. Deze studie werd uitgevoerd op een mutant van P450 BM3, mutant M11, die door een combinatie van site-directed en random mutagenese het vermogen heeft verworden om een groot aantal geneesmiddelen en steroides te metaboliseren met veel hogere activiteit dan de menselijke P450's. Met behulp van site-directed mutagenese werden alle 20 mogelijke aminozuren op positie 87 onderzocht. Hieruit bleek dat, afhankelijk van de aard van het aminozuur op positie 87, niet alleen de enzymactiviteit sterk veranderde maar ook de regioselectiviteit, hetgeen aantoont dat testosteron zich door de aminozuur-verandering op verschillende manieren orienteert in de active site.
Samenvatting
In hoofdstuk 4 werd de rol van het aminozuur op positie 82 op de regio- en stereoselectiviteit van steroid metabolisme onderzocht. Vervangen van het oorspronkelijke alanine-residue door het ruimtelijk grotere aminozuur tryptofaan resulteerde in een mutant dat testosteron en norethisteron met hoge regioselectiviteit op de 16ß-positie hydroxyleert. De regio-en stereoselectiviteit van P450-gemedieerde reacties wordt waarschijnlijk bepaald door de oriëntatie van het substraat ten opzichte van de reactieve ijzer-oxo species in het katalytische centrum van P450. Met behulp van spin relaxatie NMR, waarmee de kortste afstand van de waterstofatomen van een gebonden substraat ten opzichte van het heme-ijzer atoom van P450 kan worden bepaald, kon worden aangetoond dat de mutatie op positie 82 inderdaad tot een veranderde oriëntatie van testosteron in de actieve plaats van P450 BM3 heeft geleid, in overeenstemming was met de waargenomen verandering in regioselectiviteit. Veel schadelijke bijwerkingen van geneesmiddelen zijn het gevolg van P450afhankelijke vorming van hoog-reactieve metabolieten, reactieve intermediairen, die kunnen reageren met kritische macromoleculen in de cel. Al in een vroeg stadium van geneesmiddel ontwikkeling wordt daarom al onderzocht of bij metabolisme van een kandidaat geneesmiddel door P450's reactieve intermediairen ontstaan. Dit wordt veelal gedaan door de reactieve intermediaren in te vangen met het tripeptide glutathione (GSH) dat ook in elke lichaamscel als functie heeft de kritische macromoleculen te beschermen tegen reactieve producten. Echter, door de zeer lage niveau's waarmee de menselijke P450's de reactieve metabolieten produceren wordt de detectie en structuuropheldering van gevormde GSH-conjugaten sterk bemoeilijkt. In hoofdstuk 3 is de serie P450 BM3 mutanten met de verschillende aminozuren op positie 87 onderzocht op hun vermogen tot selectieve bioactivatie van clozapine. Dit geneesmiddel veroorzaakt bij sommige patienten ernstig bijwerkingen zoals agranulocytose en levertoxiciteit. In beide gevallen wordt een reactief nitrenium ion verantwoordelijk geacht voor de toxiciteit. Uit het onderzoek met de P450 BM3 mutanten bleek dat de mutant met een phenylalanine op positie 87, M11 V87F, clozapine zeer selectief en met hoge activiteit kan metaboliseren tot het reactieve nitrenium ion. Dit reactieve nitrenium ion bleek op verschillende manieren met GSH te kunnen reageren, resulterend in vijf verschillende GSH-conjugaten. Met behulp van mutant M11 V87F kon elk GSH-conjugaat van clozapine op voldoende schaal kon worden geproduceerd zodat 1 elke structuur kon worden opgehelderd met H-NMR. Deze studie bevestigt het grote potentieel van BM3 mutanten als hulpmiddel bij de identificatie van producten van reactieve metabolieten van geneesmiddelen, en, indirect, de aard van het reactieve intermediair. Dit nieuwe hulpmiddel voor de karakterisering van reactieve metabolieten kan tevens worden gebruikt om bioactivatie routes op te helderen van geneesmiddelen waarvan nog niet bekend is waarom ze toxische bijwerkingen vertonen. Uiteindelijk zal deze kennis een belangrijke rol spelen in de ontwikkeling van veiligere geneesmiddelen en kandidaat-geneesmiddelen.
Samenvatting
Zoals hierboven reeds aangegeven, kunnen bij het P450-afhankelijke metabolisme van geneesmiddelen soms producten ontstaan met een veel potentere farmacologische werking dan de uitgangsstof. Om deze reden kunnen P450's ook worden toegepast in de drug discovery fase, als alternatief voor organische synthese van structurele analoga van lead verbindingen. Tevens kan structurele modificatie door P450's leiden tot producten met verbeterde fysisch-chemische eigenschappen, zoals verbeterde wateroplosbaarheid. In hoofdstuk 5 wordt een geïntegreerde strategie gepresenteerd waarbij met behulp van verschillende P450 BM3 mutanten metabole mengsels van een verbinding worden gegenereerd, waarna de mengsels met een bioaffiniteitsplatform worden gescreened op aanwezigheid van metabolieten met hoge affiniteit voor een farmacologisch belangrijk receptoreiwit. Om het principe van deze nieuwe strategie aan te tonen was in dit onderzoek gekozen voor TAK-715. Dit geneesmiddel is een potente remmer van p38 kinase maar heeft als nadeel dat het zeer slecht wateroplosbaar is. Metabolisme van TAK-715 door P450 kan de wateroplosbaarheid verbeteren maar kan ook de affiniteit voor p38 kinase beinvloeden. Voor de identificatie van metabolieten van TAK-715 met hoge affiniteit voor p38 kinase werd gebruik gemaakt van HPLC gekoppeld aan zowel een massa spectrometer als een hoge resolutie screening (HRS) platform waarmee on-line de p38 kinase affiniteit van componenten van mengsels kan worden bepaald. Uit dit onderzoek bleek dat mutanten van P450 BM3 inderdaad in staat waren om TAK-715 zeer efficient te metaboliseren tot verschillende producten. Met behulp van het on-line bioaffiniteitsplatform kon worden aangetoond dat verschillende producten hoge bindingsaffiniteit hadden voor p38 kinase. Door grote schaal productie van de producten met het meest actieve P450 BM3 mutant kon de volledige structuur (behalve 1 stereochemie) van de actieve en niet-actieve producten worden bepaald met H-NMR spectroscopie en de affiniteit van elk product voor p38 kinase worden gekwantificeerd. Hieruit bleek dat sommige producten zelfs een hogere affiniteit vertoonden dan TAK-715 zelf. Concluderend: het in dit proefschrift beschreven onderzoek laat zien dat het enzymsysteem P450 BM3 op verschillende manieren een zeer waardevolle bijdrage kan leveren aan de ontwikkeling van nieuwe geneesmiddelen. Zo kunnen humaan-relevante geneesmiddel metabolieten met behulp van dit bacteriële enzym op grote schaal worden geproduceerd, zodat hun structuur kan worden opgehelderd en hun farmacologische en toxicologische eigenschappen in detail kunnen worden bestudeerd. De regio- en stereoselectiviteit van de enzymen, die kan worden gestuurd door site-directed mutagenese, maakt het mogelijk producten te maken die met organische synthese moeilijk toegankelijk zijn. Daarnaast heeft het geintegreerde platform van P450 BM3 mutanten in combinatie met hoge resolutie bioaffiniteitsscreening veel perspectief om te worden toegepast voor lead optimalisatie in het vroege stadium van geneesmiddelontwikkeli.
Appendices
LIST OF ABBREVIATIONS 15-OH-N
15-hydroxynorethisterone
15-OH-T
15-hydroxytestosterone
ADR
Adverse drug reaction
CLZ
Clozapine
CO
Carbon monoxide
FAD
Flavin adenine dinucleotide
FMN
Flavin mononucleotide
GSH
Glutathione (reduced)
GST
Glutathione S-transferase
HLM
Human liver microsomes
HPLC
High performance liquid chromatography
HRS
High resolution screening
IDR
Idiosyncratyc drug reaction
KPi
Potassium phosphate
NCE
New chemical entity
NET
Norethisterone
Ni-NTA agarose
Nickel nitroacetic acid agarose
NMR
Nuclear Magnetic Resonance
p38α
p38α mitogen-activated protein kinase
P450
Cytochrome P450 monooxygenase
P450 BM3
Cytochrome P450 BM3
RI
Reactive Intermediate
SD
Standard Deviation
SDS-PAGE
Sodium dodecyl sulfate polyacrylamide gel electrophoresis
T1 relaxation NMR
Spin relaxation NMR
TES
Testosterone
Tr
Retention time
UPLC
Ultra Performance liquid chromatography
Appendices
Appendices
