UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of HsaC and HsaD, an oxygenase and a hydrolase in the cholesterol catabolic pathway… Yam, Katherine 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2011_fall_yam_katherine.pdf [ 15.5MB ]
Metadata
JSON: 24-1.0072179.json
JSON-LD: 24-1.0072179-ld.json
RDF/XML (Pretty): 24-1.0072179-rdf.xml
RDF/JSON: 24-1.0072179-rdf.json
Turtle: 24-1.0072179-turtle.txt
N-Triples: 24-1.0072179-rdf-ntriples.txt
Original Record: 24-1.0072179-source.json
Full Text
24-1.0072179-fulltext.txt
Citation
24-1.0072179.ris

Full Text

Characterization of HsaC and HsaD, an oxygenase and a hydrolase in the cholesterol catabolic pathway of Mycobacterium tuberculosis by Katherine Yam  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Biochemistry & Molecular Biology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2011 © Katherine Yam, 2011  ABSTRACT Mycobacterium tuberculosis (Mtb) is the leading cause of mortality from bacterial infection. A cholesterol degradation pathway identified in Mtb is implicated in the pathogen’s survival in the host. This pathway includes an Fe(II)-containing extradiol dioxygenase, HsaC, and a meta-cleavage product (MCP) hydrolase, HsaD, which are predicted to catalyze the cleavage of DHSA (3,4-dihydroxy-9,10-seconandrost-1,3,5(10)triene-9,17-dione) and subsequent hydrolysis of DSHA (4,5-9,10-diseco-3-hydroxy-5,9,17trioxoandrosta-1(10),2-diene-4-oic acid), respectively. HsaC and HsaD were expressed in E. coli, purified, and characterized. Substrates were obtained by biotransformation of cholesterol using a ΔhsaC mutant of Rhodococcus jostii RHA1. From steady-state kinetic studies, purified HsaC efficiently cleaved the proposed steroid metabolite, DHSA (kcat/Km = 15 ± 2 µM-1s-1), better than the biphenyl catechol, DHB, or a synthetic analogue, DHDS. Two halogenated substrates, 2’,6’-diCl DHB and 4-Cl DHDS, inactivated HsaC with partition coefficients < 50. Structures of HsaC:DHSA at 2.1 Å revealed predominantly bidentate binding of the catechol to the active site iron, as has been reported in similar enzymes. A high-throughput colorimetric assay was developed to screen for small molecular inhibitors of HsaC. 4-chloro-N-methyl-N-(4-[(4-methylpiperidin-1-yl) carbonyl] phenyl) benzene-1-sulfonamide and gedunin were identified as potent inhibitors of HsaC with Kic values of 450 ± 50 nM and 80 ± 10 nM, respectively. Purified HsaD had higher specificity for DSHA (kcat/Km = 3.3 ± 0.3 x 104 M-1s-1) than for the biphenyl metabolite and a synthetic analogue. The catalytically impaired S114A variant of HsaD bound DSHA with a Kd of 51 ± 2 µM. The S114A:DSHA species absorbed maximally at 456 nm, 60 nm red-shifted versus  ii  the DSHA enolate. Crystal structures of the S114A:DSHA complex at 1.9 Å identified the trapped intermediate as a 2-oxo, 6-oxido species. These data indicate that the catalytic serine catalyzes enol-to-keto tautomerization as well as C-C bond hydrolysis. While both ΔhsaC and ΔhsaD mutants of M. bovis BCG did not grow on cholesterol, the presence of an additional carbon source restored growth of the ΔhsaD mutant but only partial growth of the ΔhsaC mutant, likely due to toxic oxidized metabolites. Overall, the kinetic and structural characterization of these mycobacterial cholesterol-degrading enzymes provides novel insights into a disease of global importance.  iii  PREFACE Parts of this thesis have been published in peer-reviewed journals. The identification of the cholesterol catabolic pathway in Rhodococcus jostii RHA1 and Mycobacterium tuberculosis appeared in the Proceedings of the National Academy of Sciences of USA (Van der Geize, R., Yam, K., Heuser, T., Wilbrink, M. H., Hara, H., Anderton, M. C., Sim, E., Dijkhuizen, L., Davies, J. E., Mohn, W. W., and Eltis, L. D. “A gene cluster encoding cholesterol catabolism in a soil actinomycete provides insight into Mycobacterium tuberculosis survival in macrophages” (2007) Proc Natl Acad Sci U S A 104, 1947-1952). In this study, I was responsible for cloning hsaC and hsaD from M. tuberculosis H37Rv genomic DNA and performing the preliminary activity tests with raw extracts. Dr. Van der Geize wrote the paper. This published work is located in section 2.2.1. Characterization of HsaC appeared in PLoS Pathogens (Yam, K. C.*, D'Angelo, I.*, Kalscheuer, R., Zhu, H., Wang, J. X., Snieckus, V., Ly, L. H., Converse, P. J., Jacobs, W. R., Jr., Strynadka, N., and Eltis, L. D. “Studies of a Ring-Cleaving Dioxygenase Illuminate the Role of Cholesterol Metabolism in the Pathogenesis of Mycobacterium tuberculosis” (2009) PLoS Pathog 5, e1000344 (*shared first authorship)). In this study, I was responsible for the HsaC purification, DHSA purification, and all kinetic aspects of this work. I also contributed to interpreting the structural data. Dr. D’Angelo and I wrote the paper. This published work is located in sections 3.1.1-3 and 3.2. Characterization of HsaD appeared in the Journal of Biological Chemistry (Lack, N. A., Yam, K. C., Lowe, E. D., Horsman, G. P., Owen, R. L., Sim, E., and Eltis, L. D.  iv  "Characterization of a carbon-carbon hydrolase from Mycobacterium tuberculosis involved in cholesterol metabolism." (2010). J Biol Chem 285(1): 434-443). In this study, I was responsible for HsaD purification, purification of meta-cleavage products, DSHA and HOPODA, determination of their chemical and physical properties, and all kinetic characterization. I also contributed to interpreting the structural data. I proposed a revised reaction mechanism for HsaD. Jie Liu helped construct the S114A variant. Dr. Lack and I wrote the paper. This published work is located in sections 3.1.4 and 3.4. The high-throughput screen to identify inhibitors of HsaC was performed in collaboration with Duncan Browman and Tom Pfeifer (Center for Drug Research and Development). These results will be part of a manuscript to be submitted shortly. I was responsible for the design of the high-throughput screening assay, the screening of the KD2 and DiverSet libraries, the data analysis, and the inhibitor studies of PPB sulfonamide and gedunin. This work can be located in section 3.3.  v  TABLE OF CONTENTS ABSTRACT.............................................................................................................................. ii	
   PREFACE.................................................................................................................................iv	
   TABLE OF CONTENTS ........................................................................................................vi	
   LIST OF TABLES ....................................................................................................................x	
   LIST OF FIGURES .................................................................................................................xi	
   LIST OF ABBREVIATIONS ...............................................................................................xiv	
   ACKNOWLEDGEMENTS ..................................................................................................xvi	
   CHAPTER 1: INTRODUCTION............................................................................................1	
   1.1 Steroids..............................................................................................................................1	
   1.1.1 Cholesterol .................................................................................................................1	
   1.2 Bacterial catabolism of steroids ........................................................................................2	
   1.2.1 Cholesterol catabolic pathway ...................................................................................3	
   1.2.2 Side chain degradation ...............................................................................................4	
   1.2.3 A, B ring degradation.................................................................................................5	
   1.2.4 C, D ring degradation.................................................................................................6	
   1.2.5 Distribution of steroid catabolic genes in actinomycetes ..........................................9	
   1.3 Mycobacterium tuberculosis ...........................................................................................10	
   1.3.1 Mtb strains and other model organisms ...................................................................10	
   1.3.2 Infection cycle of Mtb ..............................................................................................11	
   1.3.3 Role of cholesterol catabolism in Mtb virulence .....................................................14	
   1.4 Enzymes of the meta-cleavage pathway .........................................................................18	
   1.4.1 Extradiol dioxygenases ............................................................................................18	
   1.4.2 Meta-cleavage product (MCP) hydrolases...............................................................22	
   1.5 Aim of this study .............................................................................................................27	
   CHAPTER 2: MATERIALS AND METHODS ..................................................................29	
   2.1 Chemicals and reagents ...................................................................................................29	
   2.1.1 Commercially and privately sourced .......................................................................29	
   2.1.2 Preparation of DHSA ...............................................................................................30	
   2.1.3 Preparation and characterization of HOPODA and DSHA .....................................31	
   vi  2.1.4 Oligonucleotides and DNA sequencing...................................................................32	
   2.2 Manipulation of DNA .....................................................................................................33	
   2.2.1 Cloning of Mtb genes...............................................................................................33	
   2.2.2 Mutagenesis of hsaD................................................................................................34	
   2.3 Growth of bacteria...........................................................................................................34	
   2.4 Protein purification..........................................................................................................35	
   2.4.1 Purification of HsaC ................................................................................................35	
   2.4.2 Purification of HsaD and the S114A variant ...........................................................36	
   2.5 Protein analysis and handling..........................................................................................37	
   2.5.1 Handling of HsaC ....................................................................................................37	
   2.6 Steady-state kinetic assays ..............................................................................................38	
   2.6.1 Oxygraph assay for HsaC ........................................................................................38	
   2.6.2 Spectrophotometric assay for HsaC.........................................................................39	
   2.6.3 Determination of partition ratio ...............................................................................40	
   2.6.4 Reversible inhibition of HsaC..................................................................................40	
   2.6.5 Steady-state kinetic measurements of HsaD............................................................41	
   2.7 Stopped-flow spectrophotometry of HsaD or S114A with HOPDA ..............................42	
   2.8 Characterization of S114A:DHSA complex ...................................................................43	
   2.8.1 Determination of Kd .................................................................................................43	
   2.8.2 Determination of half-life ........................................................................................44	
   2.9 High-throughput screening (HTS) for inhibitors of HsaC ..............................................44	
   2.9.1 Development of a HTS activity assay......................................................................44	
   2.9.2 Primary screen .........................................................................................................45	
   2.9.3 Counter screen .........................................................................................................46	
   2.9.4 Dose-response curves...............................................................................................46	
   2.10 In vitro growth of BCG wild-type and knockouts on cholesterol .................................47	
   2.10.1 Generation of BCG standard inoculum..................................................................47	
   2.10.2 Growth of BCG on various carbon sources ...........................................................48	
   CHAPTER 3: RESULTS .......................................................................................................50	
   3.1 Characterization of cholesterol metabolites and analogues ............................................50	
    vii  3.1.1 Accumulation of DHSAs by the ΔhsaC mutant of RHA1.......................................50	
   3.1.2 Production of DHSA................................................................................................52	
   3.1.3 Chemical properties of DHSA .................................................................................54	
   3.1.4 Chemical properties of HOPODA and DSHA.........................................................54	
   3.2 Characterization of HsaC ................................................................................................56	
   3.2.1 Expression and purification of HsaC .......................................................................56	
   3.2.2 Substrate specificity of HsaC...................................................................................57	
   3.2.3 Inactivation of HsaC by chlorinated catechols ........................................................58	
   3.2.4 Oxidative inactivation of HsaC................................................................................59	
   3.3 HTS for inhibitors of HsaC .............................................................................................59	
   3.3.1 HTS activity assay ...................................................................................................59	
   3.3.2 Actives identified from HTS....................................................................................61	
   3.3.3 Inhibitor studies of HsaC .........................................................................................67	
   3.4 Characterization of HsaD ................................................................................................69	
   3.4.1 Expression and purification of HsaD .......................................................................69	
   3.4.2 Substrate specificity of HsaD...................................................................................70	
   3.4.3 Transient kinetic studies of HsaD ............................................................................70	
   3.4.4 Characteristics of S114A in solution .......................................................................72	
   3.5 In vitro growth of BCG on cholesterol............................................................................75	
   CHAPTER 4: DISCUSSION .................................................................................................78	
   4.1 The role of HsaC and HsaD in cholesterol catabolism ...................................................78	
   4.2 Physiological substrates of Hsa enzymes may be CoA thioesters ..................................81	
   4.3 Insights into substrate specificity of HsaC and HsaD from their structures ...................83	
   4.4 Inhibitors of HsaC ...........................................................................................................89	
   4.5 HsaD mechanism.............................................................................................................95	
   4.6 Role of cholesterol metabolism in Mtb pathogenicity.....................................................99	
   4.6.1 Cholesterol catabolism may be essential for Mtb pathogenicity .............................99	
   4.6.2 Cholesterol metabolism may modulate the host’s immune response ....................104	
   4.7 Future directions............................................................................................................108	
   BIBLIOGRAPHY .................................................................................................................111	
    viii  APPENDIX: CRYSTALLOGRAPHIC DATA .................................................................124	
    ix  LIST OF TABLES Table 1. Summary of transcriptomic data for the cholesterol pathway genes in Mtb..............17	
   Table 2. Oligonucleotides used in this study............................................................................33	
   Table 3. Half-lives of catechols................................................................................................54	
   Table 4. Properties of MCPs ....................................................................................................55	
   Table 5. Steady-state kinetic parameters of HsaC with various catecholic substrates.............58	
   Table 6. IC50 values of hits from the HsaC screening campaign. ............................................66	
   Table 7. Competitive inhibition parameters of two reversible inhibitors of HsaC ..................67	
   Table 8. Steady-state kinetic parameters of HsaD with HOPDA, HOPODA and DSHA. ......70	
   Table 9. Crystallographic properties, X-ray diffraction data, and refinement statistics for HsaC........................................................................................................................................124	
   Table 10. Summary of data-collection and refinement statistics for S114A and S114A complexes of HsaD. ................................................................................................................126	
   Table 11. Torsional angles of the refined 2-enol and 2-oxo, 6-oxo tautomers of HOPDA, HOPODA and DSHA in the S114A variant of HsaD.............................................................127	
    x  LIST OF FIGURES Figure 1. Chemical structure of cholesterol ...............................................................................2	
   Figure 2. The cholesterol catabolic gene cluster in M. tuberculosis..........................................4	
   Figure 3. The proposed cholesterol degradation pathway .........................................................8	
   Figure 4. The life cycle of Mtb ................................................................................................12	
   Figure 5. Structure of the BphC monomer and active site.......................................................19	
   Figure 6. The proposed mechanism for extradiol dioxygenases..............................................21	
   Figure 7. Secondary structure diagram of the canonical α/β hydrolase fold ...........................23	
   Figure 8. Possible hydrolytic mechanisms of BphD................................................................26	
   Figure 9. GC-MS spectrum of TMS-derivatized 17-isopropionyl-DHSA ..............................51	
   Figure 10. The catecholic substrates used in characterizing HsaC. .........................................51	
   Figure 11. GC-MS spectrum of TMS-derivatized DHSA .......................................................52	
   Figure 12. HPLC trace of the ΔhsaC RHA1 culture supernatant incubated with cholesterol. 53	
   Figure 13. MCPs used in characterizing HsaD. .......................................................................55	
   Figure 14. SDS-PAGE analysis of HsaC purification .............................................................57	
   Figure 15. Representative progress curves for the positive and negative controls in the HsaC screening assay..........................................................................................................................61	
   Figure 16. Flow-chart of the HsaC high-throughput screening of the DiverSet and KD2 libraries .....................................................................................................................................63	
   Figure 17. Structures of polycyclic sulfonamides that were top actives identified from HsaC screening of the DiverSet library ..............................................................................................64	
   Figure 18. Dose-response curves for six hits from the HsaC screening campaign of the KD2 and DiverSet libraries ...............................................................................................................65	
   Figure 19. The dose-response curve for 4-Cl DHDS...............................................................66	
   Figure 20. Dixon plots of steady-state analysis of PPB sulfonamide and gedunin, two reversible inhibitors of HsaC ....................................................................................................68	
   xi  Figure 21. SDS-PAGE analysis of HsaD purification .............................................................69	
   Figure 22. Time-resolved spectra of the hydrolysis of HOPDA by HsaD ..............................71	
   Figure 23. Time-resolved spectra from single turnover of HOPODA by HsaD......................72	
   Figure 24. Characterization of the S114A:DSHA complex by UV-visible absorption spectroscopy..............................................................................................................................73	
   Figure 25. Decay of S114A:DSHA complex at 456 nm..........................................................74	
   Figure 26. Time-resolved spectra of HOPDA binding by HsaD S114A .................................74	
   Figure 27. Growth of wild-type M. bovis BCG, ΔhsaC, and ΔhsaD mutants on acetate, cholesterol or both acetate and cholesterol. ..............................................................................77	
   Figure 28. Growth of a ΔhsaC mutant of M. tuberculosis on cholesterol and in mice............80	
   Figure 29. Side chains of potential cholesterol metabolites.....................................................82	
   Figure 30. Structure of HsaC:DHSA complex.........................................................................84	
   Figure 31. DHSA in the HsaC active site ................................................................................85	
   Figure 32. Secondary structure and B-factor of the S114A variant of HsaD. .........................86	
   Figure 33. Overlay of the MCP substrates HOPDA, HOPODA and DSHA in complex with the S114A variant of HsaD .......................................................................................................87	
   Figure 34. Cavity of HsaC and HsaD S114A can accommodate substrates with longer side chains ........................................................................................................................................88	
   Figure 35. Compounds bound by human 11β-HSD1 ..............................................................92	
   Figure 36. Lowest binding energy conformation of gedunin in HsaC active site by automated docking......................................................................................................................................93	
   Figure 37. Steroid skeleton of potential sterol anti-tuberculars from plant extracts................95	
   Figure 38. The 1,2,4-triazole-3-thione compound, an inhibitor of NAT and a potential antitubercular ..................................................................................................................................95	
   Figure 39. The proposed mechanism of HsaD.........................................................................97	
   Figure 40. Possible conformations of DSHA in the HsaD S114A active site. ........................98	
   Figure 41. Different tautomers of 5,8-diF HOPDA fit to the electron density ........................98	
    xii  Figure 42. Survival of SCID mice after intravenous infection with 104 CFUs wild-type H37Rv, the ΔhsaD mutant or the complemented ΔhsaD mutant, respectively......................102	
   Figure 43. Growth of the ΔhsaC mutant in the guinea pig model of tuberculosis.................103	
   Figure 44. Structure of 1,25(OH)2D3 (calcitriol) and of selected analogues with immunoregulatory properties..................................................................................................107	
    xiii  LIST OF ABBREVIATIONS 4-AD  4-androstadiene-3,17-dione  3β-HSD  3β-hydroxysteroid dehydrogenase  BphC  2,3-dihydroxybiphenyl 1,2-dioxygenase  BphD  2-hydroxy-6-oxo-6-phenylhexa-2,4-dienoic acid hydrolase  DHB  2,3-dihydroxybiphenyl  DHDS  2,3-dihydroxy-6-methyl-7,8-dihydrostilbene  DHSA  3,4-dihydroxy-9,10-seco-nandrost-1,3,5(10)-triene-9,17-dione  DMSO  dimethyl sulfoxide  DOHNAA  9,17-dioxo-1,2,3,4,10,19-hexanorandro-stan-5-oic acid  DSHA  4,5–9,10-diseco-3-hydroxy-5,9,17-trioxoandrosta-1(10),2-diene-4-oic acid  DTT  dithiothreitol  E:Sred  red-shifted intermediate  GC-MS  gas chromatography-mass spectrometry  HEPES  4-(2-hydroxyethyl)-1-piperazineethanesulphonic acid  HHD  2-hydroxy-hexa-2,4-dienoic acid  HOPDA  2-hydroxy-6-oxo-6-phenylhexa-2,4-dienoic acid  HOPODA  8-(2-chlorophenyl)-2-hydroxy-5-methyl-6-oxoocta-2,4-dienoic acid  HPLC  high-performance liquid chromatography  HsaC  3,4-DHSA dioxygenase  HsaD  4,9-DSHA hydrolase  HTS  high-throughput screening  IPTG  isopropyl-β-D-thio-galactoside  xiv  LB  lysogeny broth  MCP  meta-cleavage product  Mtb  Mycobacterium tuberculosis  PCB  polychlorinated biphenyl  PPB sulfonamide  4-chloro-N-methyl-N-(4-[(4-methylpiperidin-1-yl) carbonyl] phenyl) benzene-1-sulfonamide  RHA1  Rhodococcus jostii RHA1  SCID  severe combined immunodeficient  TB  tuberculosis  xv  ACKNOWLEDGEMENTS I would like to thank my supervisor Prof. Lindsay Eltis for his guidance, patience and fearless leadership. I am very fortunate for all the research opportunities and learning experiences that you have opened up to me. My committee members, Prof. Charles Thompson and Prof. Natalie Strynadka, have been a phenomenal source of wisdom. Thank you for all your insightful comments. I would like to thank all the past and present members of the Eltis lab: Dr. Pascal Fortin, Dr. Geoff Horsman, Dr. Carly Huitema, Dr. Sachi Okamoto, Dr. Rahul Singh, Dr. Israel Casabon, Dr. Tim Machonkin, Dr. Hao-Ping Chen, Antonio Ruzzini, Joseph Roberts, and Jonathan Penfield for their expertise and many helpful discussions. I owe many thanks to Jie Liu, Christine Florizone and Gordon Stewart for their technical assistance. I would like to especially thank Jenna Capyk, Dr. Carola Dresen and Dr. Elitza Tocheva for their enduring friendship throughout the years. I am privileged to work with some of the most brilliant researchers in the world: Dr. Rainer Kalscheuer and Prof. William R. Jacobs Jr. (Albert Einstein College of Medicine, Bronx, NY), Prof. Victor Snieckus and Dr. Jian-Xin Wang (Queen’s University, Kingston, ON), Dr. Igor D’Angelo and Dr. Leo Lin (UBC), Dr. Nathan Lack and Prof. Edith Sims (University of Oxford, UK), and Dr. Duncan Browman and Dr. Tom Pfeifer (CDRD). Their contributions have greatly benefitted the HsaC and HsaD projects.  xvi  I would like to acknowledge financial support from the Natural Sciences and Engineering Research Council (NSERC) of Canada, the Michael Smith Foundation for Health Research, and the University of British Columbia. Although I have been far from home, I wish to thank my family and friends across Canada who were with me in spirit and made me feel loved. Last, but not least, thank you to my best friend, Christopher, who has always supported and encouraged me when I was working through the biggest project of all: life.  xvii  CHAPTER 1: INTRODUCTION 1.1 Steroids Steroids are a family of tetracyclic triterpenoid lipids that play an important role in membrane function and regulate a vast number of fundamental biological processes including metabolism, immune function, homeostasis, and reproduction. Hundreds of steroids have been identified including those from insects (ecdysteroids), plants (phytosterols), and fungi (ergosterols) (1). The most well known vertebrate steroids are sex hormones, corticosteroids and cholesterol. Interestingly, bacteria do not appear to synthesize steroids, despite their metabolic diversity and distribution (2). However, some bacteria synthesize a different class of triterpenoids, the pentacyclic hopanoids that play a role in membrane stabilization (3). Steroids consist of a four-ringed nucleus and often include a branched alkyl side chain of varying complexity. They fall into several families based on the various functional groups attached to a core structure. Sterols, represented by cholesterol, are a subset of steroids that have a hydroxyl group at position 3 on the cholestane skeleton (Figure 1). 1.1.1 Cholesterol Cholesterol is the principal sterol synthesized by animals, where it is a precursor to fat-soluble vitamins and steroid hormones, and is an essential component of mammalian cell membranes that modulates fluidity (4,5). Cholesterol levels vary by cell type, but are reported to be as high as 45% of total lipids in erythrocyte cell membranes (6). Cholesterol is  1  characterized by a carbon skeleton of four fused rings, labeled A to D, and a side chain consisting of eight carbons (Figure 1).  Figure 1. Chemical structure of cholesterol. Rings are labeled A to D. Carbons are numbered according to the standard nomenclature of steroids. 1.2 Bacterial catabolism of steroids The earliest account of microbial steroid degradation can be traced back to the mid1900’s, when species of the genus Nocardia (formerly Proactinomyces (7)) were found to utilize cholesterol as a sole carbon source (8). Since the early 1960’s, a variety of bacteria have been recognized to degrade steroids aerobically via a process that involves aromatization of ring A (9). Moreover, the knowledge gained from the bacterial transformation of sitosterol, a plant steroid, has been important in the synthesis of steroid drugs (10). Indeed, better understanding of steroid catabolism has been driven by the motivation to engineer strains for the biotransformation of steroids. Aspects of aerobic steroid catabolism have been documented in bacteria from the genera Nocardia (11), Pseudomonas (12), and Mycobacterium (13). Genes for steroid  2  degradation were first identified in Comamonas testosteroni TA441 (14) and Rhodococcus erythropolis SQ1 (15), based on their catabolism of testosterone and 4-androstene-3-17dione, respectively. Many members of the suborder Corynebacterineae, including those mentioned above, possess exceptional catabolic diversity. In fact, members of Rhodococcus, a genus of soil-dwelling microbes, have long been known to degrade a range of naturally occurring steroids including cholesterol and phytosterols (16). Recent genomic studies have demonstrated that the Rhodococcus jostii RHA1 genome has four clusters of genes potentially encoding distinct steroid catabolic pathways (17). All predicted pathways involve aromatization of ring A and are rich in oxygenases. Three of these steroid pathways have unknown substrates, and only the cholesterol catabolic pathway is the best characterized to date (18). 1.2.1 Cholesterol catabolic pathway A cholesterol catabolic pathway has been best studied in R. jostii RHA1 and Mycobacterium tuberculosis (Mtb), which possess similar gene clusters (Figure 2) responsible for uptake of the cholesterol and the degradation of the side-chain, rings A and B, and the partial degradation of rings C and D (Figure 3). A multi-component ATP-dependent uptake system, encoded by the mce4 locus, transports cholesterol in actinomycetes including RHA1 and Mtb (19,20). Once imported, it is unclear if degradation of the cholesterol side chain or the rings occurs first. The accumulation of 4-androstene-3-17-dione (4-AD) and 1,4androstadiene-3,17-dione (ADD) (Figure 3, 1, 2, respectively) by Mycobacterium sp. mutants without appreciable degradation of the steroid rings suggests that the side-chain degradation occurs prior to sterol ring degradation (13). However, a knockout of a ring-degrading enzyme  3  of RHA1 was found to accumulate metabolites with various side chain lengths suggesting that, in at least some strains, side chain and ring degradation occur concurrently (21). Regardless of the order of these processes, many of the genes involved in steroid catabolism have yet to be identified, and many of the pathway enzymes are poorly characterized, particularly those involved in degrading the bicycloalkanone originating from rings C and D. In addition, many steroid-degrading enzymes have been misannotated in other bacteria, due to the fact that the ring-opening enzymes have high sequence similarity to the enzymes that degrade biphenyl.  Figure 2. The cholesterol catabolic gene cluster in M. tuberculosis. Genes in the physical map are colour-coded according to predicted and confirmed function: yellow, uptake; red, side-chain degradation; blue, cleavage of rings A and B; orange, degradation of DOHNAA proprionate moiety; green, degradation of rings C and D; and purple, regulation. Grey arrows represent conserved hypothetical genes. White arrows represent genes for which no homologue was found in R. jostii RHA1 and are likely not involved in cholesterol catabolism. 1.2.2 Side chain degradation The 8-carbon branched alkyl side chain of cholesterol is degraded by successive cycles that resemble the β-oxidation of fatty acids (22), as initially deduced in Nocardia (23). Recent molecular, genetic, and spectroscopic data support the concept that the initiation of sterol side-chain degradation occurs by oxidation of C26 of cholesterol catalyzed by  4  functionally redundant cytochrome P450s in Mtb: CYP125 (24,25), and CYP142 (26). Once the side chain is activated by an acyl CoA-synthetase, repetition of the cycle requires an acylCoA dehydrogenase, an enoyl-CoA hydratase, a hydroxyacyl-CoA dehydrogenase and a thiolase. Degradation of the cholesterol side chain results in the release of two molecules of propionyl-CoA and one molecule of acetyl-CoA (Figure 3). Mtb and RHA1 have multiple sets of enzymes necessary to perform these β-oxidation-like reactions (Figure 2, highlighted in red) and their genes are highly up-regulated during growth on cholesterol (18). However, specific roles have yet to be assigned to these enzymes in most cases. 1.2.3 A, B ring degradation Elucidation of the cholesterol ring-degrading enzymes was aided largely by their close similarity to enzymes involved in the catabolism of testosterone by C. testosteroni (14). Degradation of the steroid nucleus in Nocardia (27) and Mtb (28) is initiated by a NAD(P)+dependent 3β-hydroxysteroid dehydrogenase (3β-HSD) to produce 4-cholestene-3-one. Cholesterol oxidases in Streptomyces and R. equi perform the same function (29,30); their mechanism and structure are nearly identical. In the case of complete removal of the sterol side chain, 4-cholestene-3-one is converted to 4-androstadiene-3,17-dione (4-AD) (Figure 3, 1). Further metabolism of 4-AD to 9α–hydroxy-1,4-androstadiene- 3,17-dione (9-OHADD) (Figure 3, 3) involves the dehydrogenation of ring A by 3-ketosteroid-Δ1-dehydrogenase (KstD), followed by 9α-hydroxylation by 3-ketosteroid-9α-monooxygenase (KshAB), a twocomponent Rieske oxygenase. 9-OHADD undergoes aromatization and cleavage of ring B via a non-enzymatic reverse-aldol reaction to produce 3-hydroxy-9,10-secondandrost1,3,5(10)-triene-9,17-dione (3-HSA) (Figure 3, 4). Ring A of 3-HSA is hydroxylated by  5  HsaAB, a two-component enzyme comprised of an oxygenase and reductase and requires molecular oxygen and NADH. The reaction yields a catecholic 3,4-dihydroxy-9,10seconandrost-1,3,5 (10)-triene-9,17-dione (DHSA) (Figure 3, 5). The formation of this catechol moiety by HsaAB is essential for the subsequent ring-cleaving steps. An extradiol dioxygenase (HsaC) catalyses the oxygenolytic ring A cleavage of DHSA. An MCP hydrolase, HsaD, then cleaves the C-5:C-6 bond of 4,5-9,10-diseco-3-hydroxy-5-9-17trioxoandrosta-1(10),2-diene-4-oic acid (DSHA) (Figure 3, 6) through the addition of water, resulting  in  2-hydroxyhexadienoate  (2-HHD)  and  9,17-dioxo-1,2,3,4,10,19-  hexanorandrostan-5-oic acid (DOHNAA) (Figure 3, 7) as products. The steps catalyzed by HsaC and HsaD are typical of the aerobic catabolism of aromatic compounds by bacteria. 2HHD is transformed to central metabolites such as propionyl-CoA and pyruvate involving the successive actions of a hydratase (HsaE), an aldolase (HsaF), and an acetaldehyde dehydrogenase (HsaG). 1.2.4 C, D ring degradation Members of various genera including Nocardia, Pseudomonas, and Mycobacterium are capable of catabolizing DOHNAA, the cholesterol catabolite comprising rings C and D (31). However, the step-wise catabolic pathway and the enzymes involved have yet to be completely elucidated. The degradation of DOHNAA likely occurs by the removal of 2carbon atoms from the propionic acid side chain, similar to β-oxidation, as demonstrated in R. equi (32) and Comamonas testosteroni (33). The enzymes catalyzing the opening of rings C and D are poorly characterized but likely involves Baeyer-Villiger oxidation either before or after attack of the side chain (34). The genome of RHA1 reveals two genes encoding a  6  Baeyer-Villiger monooxygenase and a lactone hydrolase (Ro06698 and Ro06693, respectively) that are predicted to act on ring D of DOHNAA (17,18); these genes are typically associated with cycloalkanone fission. The corresponding enzymes in Mtb have not yet been identified.  7  COOH CH2OH  CYP125 or CYP142  C A  O2 NADH + H+  D  B  O  COOH  CYP125 or CYP142 FadD19, FadE26, Hsd4AB, Ltp3 FadD17, FadE27, EchA19, Ltp4, FadA5  NAD+ H2O  3!HSD  HO  Cholesterol  COOH  R  NAD+  R  R  KstD  R  KshAB  NADH NAD(P)H O + H+  O  1  2  O2 NADH + H+  NAD+ H2O  O  OH HO  O  3  4 O2 NADH + H+  HsaAB  H2O  HsaEFG OH  propionyl-CoA + pyruvate  HsaD  +  OH R  ?  R  R  HOOC  H2O  O  HsaC O  O  HOOC  O2  HO OH  6  O HOOC  5  7  Figure 3. The proposed cholesterol degradation pathway. Side chain degradation and A-B ring degradation are depicted independently, but these processes are likely concurrent. It is unknown whether CYP125/CYP142 or 3β−HSD acts on cholesterol first. Side chain degradation is thought to occur through a β-oxidation-like process starting with thioesterification of the acid group. The identity of the R-group is unknown but may be a partially degraded side chain and/or a CoA-thioester. If R represents a ketone moiety, then 1 = 4-AD, 2 = ADD, 3 = 9-OHADD, 4 = HSA, 5 = DHSA, 6 = DSHA and 7 = DOHNAA. Brackets indicate a non-enzymatic step. Bold arrows indicate enzymatic steps for which there is evidence in Mtb. Dashed arrows indicate multiple hypothetical reactions. The highlighted box indicates enzyme reactions that are characterized in this study. 8  1.2.5 Distribution of steroid catabolic genes in actinomycetes Examination of genomes of mycolic acid-containing actinomycetes indicates that the cholesterol catabolic pathway is widespread among these bacteria. Indeed, the pathway has been demonstrated in R. jostii RHA1, but only predicted to be present based on gene sequence alignment and organization in all rhodococci whose genomes have been sequenced (R. opacus B4, R. erythropolis PR4, and R. equi), Nocardia farcinia IFM10152, Gordonia bronchialis DSM43247, and Gordonia sp. KTR9 (H-P Chen and L. Eltis, personal communication). Interestingly, the pathway has not been found in any Corynebacterium species based on genome sequence analyses. The cholesterol catabolic pathway has been recently confirmed in Mtb, the causative agent of tuberculosis (TB) and the most devastating infection agent of mortality worldwide. Among intracellular pathogens, the pathway is predicted to occur in M. bovis, causing TB in cattle and humans and sharing >99.95% sequence identity with Mtb (35); M. avium ssp. paratuberculosis which causes Johne’s disease in livestock and is associated with Crohn’s disease in humans (36); and Rhodococcus equi, which primarily infects foals but is an emerging pathogen in immunocompromised humans (37). The cluster is absent in M. leprae, an obligate intracellular pathogen causing leprosy. Cholesterol and cholesterol degradation enzymes seem to play an important, but still unknown role in the pathogenicity of some of these actinomycetes (38,39). Further genomic studies of these and other cholesterol-utilizing organisms may help elucidate the role of cholesterol in these pathogens.  9  1.3 Mycobacterium tuberculosis As mentioned in the previous section, Mtb is a GC-rich, obligate aerobe belonging to the Corynebacterineae suborder of actinomycetes. Mtb can be classified as an acid-fast Gram-positive organism due to the high mycolic-acid content in its cell envelope (40). These mycolic acids occur in the inner leaflet of the outer membrane, and confer resistance to dehydration, alkali conditions and chemical disinfectants (41). This resistance extends to common antibiotics, which limits the number of effective therapeutics to treat Mtb infection. Currently, 2 billion people are infected with TB with 8 million progressing to active disease and 1.7 million deaths per year (42). TB infections are on the rise due to the mutually acceleratory effect of HIV and TB on co-infected individuals, as well the emergence of multidrug-resistant TB strains (MDR-TB). MDR-TB is resistant to the first-line drugs used to treat TB: isoniazid and rifampicin. Extensively-drug resistant TB strains (XDR-TB) are MDR strains that also exhibit resistance to second-line drugs (e.g., kanamycin and fluoroquinolones) (43) and have been reported on all continents. In 2009, the first US case of extremely-drug resistant TB (XXDR-TB) resistant to all current TB chemotherapeutics was reported (44). There is an urgent need for novel therapeutics that will combat these emerging strains. 1.3.1 Mtb strains and other model organisms The study of Mtb pathogenicity and physiology requires appropriate Mtb strains and model organisms. The H37Rv strain is the most commonly used reference strain for tuberculosis research (45). Clinical isolates such as the highly-transmissible strain,  10  CDC1551, gives valuable insight into the induction of the immune response induction (46). Unfortunately, the slow generation time and virulence of these strains make them inconvenient organisms for genetic manipulation. One model organism more amenable to in vitro study is another member of the M. tuberculosis complex, M. bovis Bacillus CalmetteGuérin (BCG). BCG is an avirulent strain, well-known as a tuberculosis vaccine. The loss of virulence in BCG can be attributed to a missing chromosomal “region of difference” (RD1) (47). Nevertheless, BCG is a well-suited model for the study of Mtb since the two genomes share a high degree of similarity (>99.9%). Other proposed models for Mtb have included non- and mildly-pathogenic fast-growing mycobacteria: M. smegmatis (48) and M. marinum (49). M. marinum has 85% sequence identity to 3000 Mtb gene orthologues, however, ~30% of Mtb genes lack the conserved orthologues in M. smegmatis (50). Due to low pathogenicity in mammalian macrophages and moderate gene conservation, their use as model organisms has been limited. Regardless, the cholesterol catabolic pathway has been predicted by sequence alignment in all of the abovementioned organisms. 1.3.2 Infection cycle of Mtb Infection with Mtb begins when bacilli are inhaled as airborne droplets. In the lung, the bacteria are phagocytosed by alveolar macrophages. The mycobacteria-containing phagosomes accumulate the early endocytic surface GTP-binding protein Rab5 but fail to develop the late endosomal signal, Rab7, signifying a block in phagosomal maturation (51). This maturation arrest is critical for Mtb persistence and development of the sites of infection known as the granuloma. The balance between containment of infection and the progression to active disease is determined at the granuloma, which is comprised of a core of infected  11  macrophages surrounded by foamy macrophages, mononuclear phagocytes and lymphocytes (52). Maturing granulomas gradually develop a fibrous sheath of collagen and other extracellular matrix components, forming a hypoxic center (Figure 4). The disease progression is marked by necrosis and liquification of the center giving it its cheese-like (caseum) appearance. Caseation is triggered when the host becomes immuno-compromised, and quickly leads to rupturing of the granuloma and release of infectious bacteria. Formation of the granuloma may be a host response to wall off the infection. However, there is mounting evidence that these structures and their caseation may be pathogen-mediated (53). The mechanism of persistence in the granuloma is poorly understood, but host cholesterol is implicated in this process (39,54). Figure 4 (next page). The life cycle of Mtb. The schematic illustrates the phagocytosis of the Mtb bacilli by alveolar macrophages; recruitment of the mononuclear cells to the site of infection; formation of the granuloma; and progression to active disease. Reprinted by permission from Macmillan Publishers Ltd: Nature Immunology (D.G. Russell, P-J Cardona, M-J Kim, S. Allain, and F. Altare. Foamy macrophages and the progression of the human tuberculosis granuloma. Nature Immunology (2009) Vol 10, 943-948.), copyright (2009).  12  13  1.3.3 Role of cholesterol catabolism in Mtb virulence Several lines of evidence suggest that cholesterol is important for Mtb pathogenesis. Firstly, host cholesterol levels are thought to be involved in macrophage infection (53). Cholesterol appears to play a role in bacterial uptake by macrophages; mycobacteria were found to invade host macrophages at the site of cholesterol-rich domains and depleting membrane cholesterol inhibited mycobacterial uptake (55,56). Related to this, the complement receptor 3 (CR3)-dependent signaling pathway mediated by cholesterol-rich lipid rafts is implicated in the ability of neutrophils to internalize M. kansasii (57). Interactions between CR3 and glycosylphosphatidylinositol-associated proteins mediate mycobacterial entry into macrophages without activating them. Moreover, cholesterol depletion leads to mycobacterial phagosome maturation and fusion with lysosomes (39). Lipid transporters are implicated in mediating Mtb virulence in macrophage infection. Finally, cholesterol is thought to rearrange the membrane-cytoskeleton network during phagocytosis (58) Secondly, cholesterol appears to be a significant, and perhaps essential growth substrate for Mtb in newly infected macrophage and in caseating granulomas, particularly as the latter were found to contain high concentrations of cholesterol in both humans and mice (54,59). Mtb studies using 14C-labeled cholesterol at C4 and C26, suggest that part of the ring structure is utilized as an energy source whereas the side chain is utilized in the biosynthesis of phthiocerol dimycocerosate (PDIM), a virulence-associated lipid (19). Side chain utilization for the biosynthesis of methyl-branched fatty acids is also supported by increased  14  intracellular propionyl-CoA as a result of Mtb cholesterol degradation, which alters the flux of the bacterium’s catabolic and anabolic pathways (60). A third role for Mtb cholesterol metabolism may be in modulating the host’s immune system. Elevated serum cholesterol caused deficient priming of the adaptive immune response of ApoE deficient mice to TB infection (61). β-Androstene steroids such dehydroepiandrosterone (DHEA) and 3β,17β-androstenediol (AED), conferred protection against Mtb infection in a mouse model (62). Hydroxysteroids are also important signaling molecules in mammalian systems. Specifically, 27-hydroxycholesterol, generated by the transformation of cholesterol by Mtb cytochrome P450, CYP142, plays a role in at least three processes: regulating cholesterol uptake and metabolism as a liver X receptor (LXR) ligand (63,64); preventing accumulation of cholesterol in macrophages and regulating macrophage differentiation (65); and modulating the activity of human estrogen receptors (66). Several genome-wide and targeted mutagenesis approaches have identified cholesterol catabolic genes of Mtb as being essential for virulence; however, the precise role of cholesterol catabolism in infection and persistence remains unknown. Transcriptomic studies have identified 23 and 29 Mtb cholesterol catabolic genes that are specifically upregulated during growth on cholesterol (67) (Table 1, identified by shaded boxes in “C” column) and survival in IFNγ-activated macrophages (68) (Table 1, identified by shaded boxes in “M” column), respectively. Transposon-insertion mutagenesis has been used to identify genes that are necessary for survival of Mtb in macrophages under conditions that model the immune response. Of 126 genes that were identified, 11 are cholesterol catabolic genes (69) (Table 1, identified by shaded boxes in “T” column). Another study identified 196  15  genes specifically required for infection and survival in a mouse model of TB, 18 of which were subsequently identified as cholesterol catabolic genes (70) (Table 1, identified by shaded boxes in “E” column). Targeted mutagenesis of some genes in the cholesterol catabolic pathway has extended the predictions of several genome-wide studies and provided more insight into the roles of specific genes. Studies of an Mtb mutant defective in the mce4-encoded transporter demonstrated that cholesterol uptake is essential for chronic infection in murine lungs and IFNγ-activated macrophages (19). Most strikingly, ΔkshA and ΔkshB mutants did not kill immunocompromised SCID mice and these strains were rapidly killed by immunocompetent mice (71). An Mtb strain lacking functional CYP125, responsible for catalyzing the first oxidation of the sterol side chain, was deficient in cholesterol metabolism due to accumulated cholest-4-en-3-one, toxic to CDC1551 cells (24). Lastly, 3β-HSD, the first committed step in ring degradation, is absolutely required for in vitro growth of Mtb on cholesterol, but interestingly, the virulence phenotype of a Δ3βhsd strain was similar to that of the wild-type (72). It is unclear whether these different phenotypes reflect the different roles of specific genes and/or the toxicity of cholesterol metabolites. It is also possible that the phenotypes reflect differences in the bacterial or animal strains used in the studies.  16  Table 1. Summary of transcriptomic data and gene deletions for the cholesterol pathway genes in Mtb. Shaded boxes indicate: C, at least 1.5-fold induced by 3 h growth on cholesterol (67); M, induced in macrophages (68); T, essential for survival in macrophages (69); E, essential for survival in mice (70). Highlighted genes indicate the enzyme products that are characterized in this study.  17  1.4 Enzymes of the meta-cleavage pathway A critical step in the bacterial catabolism of aromatic compounds is the cleavage of the thermodynamically stable aromatic ring. In aerobic catabolism, this reaction is often catalyzed by dioxygenases that act on catecholic intermediates (73), which are benzene rings possessing hydroxyl groups on adjacent carbon atoms. Ring-cleaving dioxygenases catalyze the incorporation of both atoms of oxygen from O2 into the catechol cleavage product and are classified on the basis of their mode of cleavage: intradiol dioxygenases cleave the bond between the hydroxylated carbons, and extradiol dioxygenases cleave a bond adjacent to the two hydroxylated carbons (73). The result of either cleavage is a ring-opened, muconic semialdehyde that can be hydrolyzed by a C-C bond hydrolase typical of the bacterial metacleavage pathway. Products are readily degraded to TCA cycle metabolites. 1.4.1 Extradiol dioxygenases Extradiol dioxygenases utilize a non-heme ferrous iron or another divalent metal ion to catalyze ring cleavage. All enzymes characterized to date have been classified into one of three superfamilies depending on their structural fold. Type I extradiol dioxygenases are members of the vicinal-oxygen-chelate (VOC) superfamily (74,75). Due to the intensive study of the biphenyl degradation (bph) pathway responsible for the bacterial transformation of polychlorinated biphenyls (PCBs), BphC from Burkholderia xenovorans LB400 (EC 1.13.11.39) is one of the best characterized Type I extradiol dioxygenases, cleaving 2,3dihydroxybiphenyl (DHB) in the third step of the bph pathway (76). The hsaC gene, originally annotated as bphC in Mtb (77) encodes an enzyme sharing 35% amino acid  18  sequence identity with BphCLB400. As mentioned above, HsaC has been proposed to catalyze the cleavage of DHSA, a catecholic intermediate in Mtb cholesterol catabolism.  1.4.1.1 Structure of type I extradiol dioxygenases Crystal structures of Fe(II)-dependent type I extradiol dioxygenases are now available in the ferrous (active) (78,79) and ferric (inactive) forms (80). The majority of type I enzymes identified are two–domain and each conserved domain structure consists of two copies of a βαβββ structural motif (Figure 5A). The iron-binding 2-His 1-carboxylate facial triad is located in the C-terminal domain and is the site of catechol cleavage (Figure 5B).  Figure 5. Structure of the BphC monomer and active site. A) Ribbon representation of BphCLB400 monomer. Backbone is colour-ramped from blue (N-terminus) to red (Cterminus). B) Active site showing the active site residues. Oxygen and nitrogen atoms are coloured red and blue, respectively. Fe atom is represented by an orange sphere.  19  1.4.1.2 Mechanism of extradiol dioxygenases In the proposed mechanism, based on biochemical, spectroscopic, kinetic and structural studies (73,81,82), the catecholic substrate binds first to the enzyme’s Fe(II) centre in a bidentate manner, displacing two solvent ligands (Figure 6). Spectroscopic data demonstrate that BphC binds DHB as a monoanion, deprotonating O2 but not O3, consistent with X-ray absorption spectroscopy (XAS) studies of a related enzyme. Activated by substrate binding, the ferrous centre binds O2. Biochemical studies support a mechanism involving iron-mediated transfer of an electron from the substrate to the O2, yielding a semiquinone-Fe(II)-superoxide intermediate (83). The reaction of species leads to the formation of an Fe(II)–bound alkylperoxo intermediate (84). The latter undergoes heterolytic O-O bond cleavage and Criegee rearrangement involving 1,2-alkenyl migration to produce an unsaturated lactone intermediate and an Fe(II)-bound hydroxide (Figure 6). Hydrolysis of the lactone affords the ring-cleaved product (85). Several of the proposed intermediates were recently substantiated in structural studies of homoprotocatechuate 2,3-dioxygenase (HPCD) and a slow substrate, 4-nitrocatechol (86). Nevertheless, some steps of the catalytic cycle remain unclear, including the multi-step binding of the catecholic substrate (87).  20  Figure 6. The proposed mechanism for extradiol dioxygenases. Roles of active site residues depicted. For clarity, the displacement of solvent species from the ferrous center is not depicted explicitly. Adapted with permission from Critical Reviews in Biochemistry and Molecular Biology, Vol 41, F.H. Vaillancourt, J.T. Bolin and L.D. Eltis. The ins and outs of ring-cleaving dioxygenases (2006) pp 209-246. 1.4.1.3 Inhibition of extradiol dioxygenases Extradiol dioxygenases are subject to inactivation in the absence of substrate and exhibit two forms of substrate inhibition: reversible substrate inhibition and a mechanismbased inactivation. Mechanism-based inactivation involves the dissociation of superoxide from the ternary complex prior to formation of the iron-alkylperoxo intermediate resulting in oxidation of the active site Fe(II) to produce an inactive ferric centre. Studies of the mechanism-based inactivation of BphC by chlorinated metabolites (88) and by substrate analogues that sterically occlude the O2-binding site (89) have led to the identification of  21  2′,6′-diCl DHB, a metabolite from PCB degradation as a potent inhibitor (Ki = 7 ± 1 nM). Based on this knowledge, potential inhibitors of HsaC as lead compounds in the development of novel TB therapeutics may be designed. 1.4.2 Meta-cleavage product (MCP) hydrolases Meta-cleavage product (MCP) hydrolases belong to a family of enzymes that are involved in the aerobic degradation of aromatic compounds, hydrolyzing vinylogous 1,5diketones arising from the extradiol ring cleavage of catechols (90). In the aerobic biphenyl catabolic pathway, BphD catalyzes the hydrolysis of 2-hydroxy-6-oxo-6-phenylhexa-2,4dienoate (HOPDA) (91). A proposed MCP hydrolase in the Mtb cholesterol catabolic pathway was originally annotated as bphD, but has since been renamed hsaD (18). HsaD is proposed to catalyze the hydrolysis of DSHA, the product of the HsaC-catalyzed reaction. HsaD shares modest amino acid sequence identity with other MCP hydrolases, including MhpC from Escherichia coli (36 %), which is involved in phenylproprionate degradation (92), and BphD from Burkholderia xenovorans LB400 (44 %) (93). 1.4.2.1 Structure of MCP hydrolases Crystal structures of MCP hydrolases, including those of BphD from B. xenovorans LB400 and MhpC from E. coli reveal that these enzymes belong to the α/β-hydrolase superfamily and contain a conserved Ser-His-Asp catalytic triad as well as an “oxyanion” hole, similar to a number of other hydrolytic enzymes (94). The α/β-hydrolase fold consists of seven parallel and one anti-parallel β-strand, forming a twisted β-sheet, surrounded by six α-helices (Figure 7). Many members of this enzyme family contain loop insertions that are 22  important secondary structural elements for defining the shape of and regulating accessibility to the enzyme active site (95). The insertions occur in several locations but often follow the β6-strand. The substrate-binding pocket comprises two subsites: a hydrophilic polar (P) subsite that binds the dienoate moiety of the MCP, and a hydrophobic non-polar (NP) subsite that accommodates the remainder of the MCP.  Figure 7. Secondary structure diagram of the canonical α/β hydrolase fold. The α-helices and β-strands are represented by white cylinders and gray arrows, respectively. The location of the catalytic triad is indicated by black dots. Dashed lines indicate the location of possible insertions. Reprinted from Current Opinion in Structural Biology, Vol 9, M. Nardini and B. W. Dijkstra, α/β Hydrolase fold enzymes: the family keeps growing, 732-737. Copyright (2000), with permission from Elsevier 1.4.2.2 Mechanism of MCP hydrolases Members of the α/β-hydrolase superfamily use the catalytic triad in one of two fundamental ways to catalyze reactions. In some hydrolases, the catalytic serine acts as a nucleophile to generate a covalent acyl-enzyme intermediate (Figure 8A) (96-98). It is  23  reminiscent of the “classical” nucleophilic mechanism, however, the role of the His-acid dyad as activator of the serine nucleophile is not shared among all members of the superfamily. Other members, including a dioxygenase, use the absolutely conserved histidine to act as a general base in the activation of small molecules (Figure 8B) (99). In this context, the mechanism of the MCP hydrolases has been the topic of some debate as researchers have sought direct evidence for either a nucleophilic or general base mechanism. Previous work has favoured a general base mechanism involving a gem-diol intermediate (Figure 8B). The most convincing evidence for this intermediate has been obtained from two types of studies. In the first, 18O from H218O was incorporated into the MCP of MhpC, and was exchanged into a non-hydrolyzable substrate analog (100). In the second, a line broadened signal was assigned to the gem-diol intermediate while monitoring MhpC and BphD variant-catalyzed turnover of 13C6-HOPDA by 13C NMR spectroscopy (101). The signal for this intermediate was nevertheless very weak, arguably consistent with its transient nature. Very recently, conclusive evidence for an acyl-enzyme intermediate was obtained. A trapped benzoylated-serine adduct of BphD H265Q has been observed by crystallography and formation of the covalent adduct by the wild-type enzyme during turnover has been identified by mass-spectrometry (A. Ruzzini and L.D. Eltis, in preparation). This evidence is consistent with a number of other observations. Firstly, no solvent molecule suitable for attack at the C6 carbonyl was observed in the active site in the S112A:HOPDA structure (102) . Secondly, the Ser112 hydroxyl of BphD is well positioned for a nucleophilic attack at the C6 carbonyl when it is modeled into the S112A:HOPDA structure. Thirdly, product inhibition experiments imply that the dienoate product, HPD in the case of HOPDA  24  cleavage, is released prior to benzoate as observed by stopped-flow spectrophotometry (102). Overall, the evidence is consistent with a nucleophilic mechanism in which rapid acylation of the catalytic serine is followed by rate-limiting deacylation. Regardless of the nature of the C-C bond hydrolysis, the catalytic triad of MCP hydrolases have another important catalytic role: tautomerization of the enolate substrate to a keto intermediate prior to hydrolysis (103). Formation of the diketone is essential to generate the electron sink required for C-C bond cleavage. Nevertheless, features of the MCP hydrolase mechanism remain to be resolved. First, the nature of a transient intermediate observed in the transformation of HOPDA by BphD, E:Sred, is unclear (93). E:Sred possesses a red-shifted absorption spectrum (λmax = 492 nm) with respect to that of the free HOPDA enolate (λmax = 434 nm). X-ray diffraction studies of a similar intermediate (λmax = 506 nm) trapped using the S112A variant of BphD revealed torsion angles that were most consistent with a ketonized tautomer of HOPDA (E:Sk) (93). Nevertheless, the refinements could not exclude the 2-enol tautomer, and a strained enolate (E:Sse) more satisfactorily accounts for the red-shifted absorption spectrum as strain of the double bond out of planarity raises the ground state energy level but not the excited state (104). Second, the roles of catalytic residues remain unclear. Histidine has also been previously proposed to protonate C5 initiating tautomerization (102), however, inspection of crystallographic data suggests the serine is also well positioned for protonation. Furthermore, while the active site histidine has been shown to be essential in forming the E:Sred intermediate (102), its role in hydrolysis has not been well-established.  25  Gly42  Arg190  N H  Met133 NH  Trp266 Asn111  Ser112 OH Ph  NHis265  CO2 O  OH  HOPDA  Arg190  Gly42 N H  Met133  CO2  NH Ph  Trp266 O  Ser112  O OH  H  Asn111  NHis265  A  B  Gly42  Arg190 Ph  NH Met133  CO2 Met133  O  NH  Gly42  Trp266 O  NH  H  CO2  O  Trp266 O  Asn111  OH Ser112  Arg190 Ph  NH  Asn111  OH H  Ser112  NHis265  NHis265  H2O Gly42  Arg190 Ph  NH  HR  Met133  Met133  O  NH  Gly42  Trp266 O  O  HS  NH  NHis265  CO2  O  Trp266 H  Asn111  O  O  Asn111  H H  O  H  Ser112  Arg190 Ph  NH  CO2  Ser112  NHis265  HPD  H2O  CO2  H  OH Gly42  Gly42 Met133  Met133 O  NH  acyl-enzyme  Ph  NH  H O O  Ser112  Arg190 Ph  NH  N H  CO2  O HO  Trp266 O  gem-diol  Asn111  OH  H  H  Ser112  NHis265  NHis265  O Ph  OH  Benzoate  Products  Figure 8. Possible hydrolytic mechanisms of BphD. Schematic showing the hydrolysis of HOPDA that can occur via nucleophilic mechanism (A) or general base mechanism (B). Residue numbers correspond to those in BphD.  26  1.4.2.3 Inhibition of MCP hydrolases BphD represents a bottleneck in the degradation of PCBs by the bph pathway because some chlorinated HOPDA congeners are poorly transformed (105,106). Studies with purified enzyme revealed that HOPDAs bearing substituents on the dienoate (position 3, 4 or 5) are poor substrates for BphD likely due to steric perturbations in the active site leading to restricted ability to catalyze tautomerization (107). For example, 3-Cl HOPDA and 4-Cl HOPDA inhibit BphDLB400 (107) and BphDP6 from R. globerulus P6 (108). 3-Cl HOPDA is a particularly potent inhibitor of BphD due to its high stability and relatively strong competitive inhibition of the enzyme. Steady-state and transient-state kinetics of BphD and a catalytically deficient S112A variant with a series of 3-substituted HOPDA suggests that sterically bulky 3-substituents, but not electronegativity, limit turnover due to impaired E:Sred formation and tautomerization (109). Crystallographic analysis of the S112A:3-Cl HOPDA and 3,10-diF HOPDA complexes reveal that 3-substituted HOPDAs bind in a coplanar, catalytically non-productive mode. 1.5 Aim of this study The overall objective of this thesis is to characterize HsaC and HsaD and confirm that they are involved in cholesterol catabolism essential for Mtb survival. To achieve this goal, HsaC and HsaD were produced in E. coli and characterized. To facilitate the characterization of these enzymes, key cholesterol metabolites and their analogues were obtained through biotransformation using mutant strains of R. jostii RHA1. The enzymes’ specificities (kcat/Km) for the steroid metabolites and related compounds were determined.  27  Crystallographic studies of enzyme:substrate complexes for HsaC and the catalytically impaired S114A variant of HsaD were performed in collaboration with Prof. Natalie Strynadka (UBC) and Prof. Edith Sim (Oxford), respectively. In order to assess the role of HsaC and HsaD in cholesterol metabolism, I compared growth of wild-type, ΔhsaC and ΔhsaD BCG mutants on cholesterol and other carbon sources. In search of a potent inhibitor of HsaC, a high-throughput assay was designed to screen natural and synthetic chemical libraries. Screened hits were characterized with respect to their inhibitory effect on HsaC in dose-response curves and steady-state inhibition assays. Structural data, presented in the discussion, aided the interpretation of kinetic and biochemical data. Overall, the study has provided insight into the function of a member of the class of Type I extradiol dioxygenases and a member of the α/β hydrolase superfamily and helped shed light on the metabolic role of two enzymes that are important to Mtb cholesterol metabolism. These cholesterol catabolic enzymes may play critical roles in infection and persistence of this pathogen in the host. Thus, the elucidation of their substrate specificities and design of small molecule inhibitors will provide novel insights into a disease of global importance.  28  CHAPTER 2: MATERIALS AND METHODS 2.1 Chemicals and reagents 2.1.1 Commercially and privately sourced DHB, DHDS and 2′,6′-diCl DHB were synthesized by a combined directed ortho metalation cross-coupling strategy (110) and were gifts from Prof. Victor Snieckus (Department of Chemistry, Queen’s University, Kingston, Ontario). HOPDA was produced enzymatically from DHB following procedures described previously (107). Ferene S was from ICN Biomedicals Inc. (Cosa Mesa, CA). Restriction endonucleases and Antarctic Phosphatase were purchased from New England Biolabs (Beverly, MA). T4 DNA ligase and Taq DNA polymerase were from Amersham Biosciences (Baie d’Urfe, QC) and Promega (Madison,  WI),  respectively.  4-Chloro-N-methyl-N-(4-[(4-methylpiperidin-1-  yl)carbonyl]phenyl) benzene-1-sulfonamide (referred to as PPB sulfonamide in this manuscript) and gedunin were purchased from Chembridge (San Diego, CA) and Tocris Bioscience (Ellisville, MO), respectively. N,O-Bis(trimethylsilyl)trifluoroacetamide with 1% trimethylchlorosilane (BSTFA+TMCS) (99:1) were purchased from Supelco (Bellefonte, PA). Buffers for protein purification and characterization were prepared using water purified on a Barnstead NANOpure UV apparatus to a resistivity greater than 17 MΩ·cm. All the other chemicals were of analytical grade and used without further purification.  29  2.1.2 Preparation of DHSA DHSA was produced by incubating a culture of the ΔhsaC mutant of R. jostii RHA1 (18) with cholesterol. Briefly, several colonies of the mutant were used to inoculate 100 ml W minimal salt medium (111) containing 20 mM pyruvate. At mid-log phase (OD600 of 1.0), 50 ml of preculture was used to inoculate 5 litres W media containing 20 mM pyruvate and 0.5 mM cholesterol. When the culture attained an OD600 of 1.5, the cells were harvested by centrifugation and the pellet was resuspended in 0.5 litres W media containing 0.5 mM cholesterol in a 2-litre baffled flask. Production of metabolites in culture supernatant was monitored using HPLC. When the culture supernatant contained the maximum amount of DHSA, the supernatant was collected by centrifugation, acidified to ~pH 6 using 0.5% orthophosphoric acid, and then extracted twice with 0.5 volume of ethyl acetate. The ethyl acetate fractions were pooled, dried with anhydrous magnesium sulphate, and evaporated to dryness with a rotary evaporator. The residue was dissolved in either a 70:30 or a 44:56 mixture of methanol/water containing 0.5% phosphoric acid and purified using a Waters model 2695 HPLC (Milford, MA) equipped with a Prodigy 10 µm ODS-Prep column (4.6 x 250 mm; Phenomenex, Torrance, CA). Fractions containing 17-isopropionyl-DHSA (tR ~38 min) and DHSA (tR ~21 min) could be eluted using the 70:30 methanol/0.5% phosphoric acid mixture at a flow rate of 1 mL/min. The eluate was monitored at 280 nm. Fractions containing pure DHSA (tR ~35 min) were collected using an solvent mixture of 44:56 methanol/0.5% phosphoric acid at a flow rate of 1 mL/min, added to 10 volumes of water, and extracted as described above.  30  To identify the metabolites, 17-isopropionyl-DHSA and DHSA were derivatized using BSTFA+TMCS and analyzed by using an HP 6890 series gas chromatograph (Agilent Technologies Canada Inc., Mississauga, ON) fitted with an HP 5973N mass-selective detector (Agilent Technologies Canada Inc., Mississauga, ON) and a 30 m × 250 µm HP5MS Agilent column in electron ionization mode. The operating conditions were: TGC (injector), 280 °C; TMS (ion source), 230 °C; oven time program (T0 min), 104 °C; T2 min, 104 °C; T14.4 min, 290 °C (heating rate 15 °C·min−1); and T29.4 min, 290 °C. The non-enzymatic transformation of DHB, DHDS and DHSA were determined by monitoring their absorbance spectra in 20 mM HEPES, 80 mM NaCl, pH 7.0 at 25 ˚C over 16 h. The experiments were performed using 0.3 mM of each catechol, in duplicate. The half-lives were determined by fitting a single-phase decay equation to the averaged increase at 400 nm, corresponding to the reported absorption spectrum of o-benzoquinone (112), using GraphPad Prism software (GraphPad Software, San Diego, CA). The extinction coefficient of DSHA was determined using an oxygraph assay, described below. 2.1.3 Preparation and characterization of HOPODA and DSHA The meta-cleavage products (MCPs) HOPODA and DSHA were produced by dissolving the corresponding catechol, DHDS and DHSA, respectively, in a small volume of ethanol and diluting to the desired volume with 100 mM potassium phosphate, pH 7.5 (ethanol constituted less than 0.1% of the solution). To this solution a sufficient amount of HsaC was added to completely transform the catechol to the corresponding MCP. Upon the addition of HsaC to a diluted sample, complete cleavage was verified by ensuring that there  31  was no further increase in absorbance of the ring-cleaved products at 396 nm (ε = 6.8 and 3.8 mM−1 cm−1 for HOPODA and DSHA, respectively). The MCP-containing solution was extracted into ethyl acetate, dried and evaporated as described above and stored at -20 °C. Solutions of MCPs for enzyme kinetics were prepared in 100 mM potassium phosphate pH 7.5 and used within 2 h. To determine pKa values, HOPODA and DSHA were prepared as described above in unbuffered water. HsaC was removed by ultrafiltration using a stirred cell equipped with an YM10 membrane (Amicon, Billerica, MA). Solutions of 0.5 mM MCP were acidified to pH 3 using 12 N HCl. This solution was titrated with aliquots of 50 mM NaOH, and the pH was determined after the addition of each aliquot. The pKa values were determined from plots of pH versus the amount of base added. The non-enzymatic transformation of MCPs in solution (100 mM potassium phosphate, pH 7.5, at 25 °C) was determined by monitoring the absorption spectra. The experiments were performed using two concentrations of each MCP (0.03, 0.3 mM). The half-lives were determined by fitting equations to the decay at 396 nm using Microsoft Excel (Microsoft Corp., Redwood, WA). 2.1.4 Oligonucleotides and DNA sequencing The oligonucleotides used in this study were purchased from Integrated DNA Technologies (Coralville, IA) and are presented in Table 2.  32  2.2 Manipulation of DNA DNA was propagated, purified, digested, ligated and amplified using standard protocols (113). PCR was performed using a RoboCycler GRADIENT 96 (Stratagene, La Jolla, CA). The nucleotide sequence of cloned amplicons was verified at the Nucleic Acid Protein Service Unit (NAPS), University of British Columbia. 2.2.1 Cloning of Mtb genes The hsaC and hsaD genes were amplified from the genomic DNA of M. tuberculosis H37Rv by PCR using Expand High Fidelity DNA polymerase (Roche Diagnostics, Indianapolis, IN) and either of the following forward and reverse primer pairs: Hcmt-F and Hcmt-R or Hdmt-F and Hdmt-R (Table 2). The amplicons were digested with NdeI and BamHI and cloned into similarly digested pT7–7 (114). The nucleotide sequences of the cloned genes were confirmed to yield pT7HC1 and pT7HD1. HsaC and HsaD were produced using E. coli GJ1158 transformed with pT7HC1 and pT7HD1, respectively. Table 2. Oligonucleotides used in this study Primer  Sequencea  Hcmt-F  CGACTAGCATATGAGCATCCGGTCGC  Hcmt-R  CGGGATCCCTGAGCCGACATCGTTTG  Hdmt-F  CGACGTACATATGACAGCTACCGAGGAATTG  Hdmt-R  CAGGATCCTCATCTGCCACCTCCCAG  HdS114A  CTGGTGGGCAACGCGTTGGGCGGGG  Comments Forward primer for hsaC H37Rv NdeI Reverse primer for hsaC H37Rv BamHI Forward primer for hsaD H37Rv NdeI Reverse primer for hsaD H37Rv BamHI Ser114 (TCG) to Ala114 (GCG)  a  The recognition sequences for the indicated restriction sites are underlined. The substituted nucleotide is highlighted in bold font.  33  2.2.2 Mutagenesis of hsaD Ser114 of HsaD was substituted with alanine to generate the S114A variant using the QuikChange II-E site-directed mutagenesis kit (Stratagene, La Jolla, CA) as per the manufacturer’s instructions. In this reaction, the template was pT7HD1, described in Section 2.2.1, and the primer was HdS114A (Table 2). The nucleotide sequence of the mutated gene was confirmed using an ABI 373 Stretch DNA sequencer (Applied Biosystems, Foster City, CA) and the Big-Dye Terminator v3.1 kit (Applied Biosystems, Foster City, CA). The S114A variant was subcloned into pEMBL18 (115) downstream of the Plac promoter as an XbaI fragment, yielding pEMHSA. 2.3 Growth of bacteria HsaC and HsaD were produced in E. coli strain GJ1158 containing pT7HC1 or pT7HD1, respectively. The cells were grown at 30 °C in salt-depleted lysogeny broth (10 g of tryptone extract, 5 g of yeast extract/liter) supplemented with 100 µg/ml ampicillin. For cells expressing hsaC, culture media was supplemented (10 ml per liter) with an HClsolubilized solution of minerals containing 14.1 mM MgCl2·6H2O, 2.5 mM CaCO3, 4.27 mM FeSO4·7H2O, 0.625 mM ZnSO4·7H2O, 0.625 mM MnSO4·H2O, 0.125 mM CuSO4·5H2O, 0.125 mM CoCl2·5H2O, 0.125 mM H3BO3, 262.5 mM MgSO4, 10 mM CaCl2, and 0.1 mM thiamine. Cultures were grown to an optical density of 0.5 at 600 nm, at which point NaCl was added to a final concentration of 300 mM. The cultures were incubated for an additional 20 h, and then the cells were harvested by centrifugation. The S114A variant was produced using E. coli strain DH5α containing pEMHSA. The cells were grown at 30 °C in LB broth  34  supplemented with 100 µg/ml ampicillin. The culture was grown to an optical density of 0.5 at 600 nm, and isopropyl-1-thio-β-d-galactopyranoside (IPTG) was added to a final concentration of 0.5 mM. The culture was incubated for an additional 20 h before the cells were harvested. Mycobacterium bovis bacillus Calmette-Guérin Pasteur (BCG) (Pasteur Vaccine strain) was a gift from Dr. Rainer Kalscheuer. Deletion mutants (ΔhsaC and ΔhsaD) were generated by specialized transduction according to the protocol described previously (116) and were a gift from Dr. William Jacobs Jr. 2.4 Protein purification 2.4.1 Purification of HsaC HsaC was purified anaerobically using a two-column protocol. Briefly, cells from 3 litres of E. coli GJ1158 culture containing pT7HC1 were resuspended in 30 ml of 10 mM TRIS, pH 7.5 containing 1 mM MgCl2, 1 mM CaCl2 and 0.1 mg/ml DNAse I and disrupted using a French Press operated at 20,000 psi. The cell debris was removed by ultracentrifugation (120,000 g x 45 min). The clear supernatant fluid (~40 ml) was decanted, referred to as the raw extract, and divided into two equal portions. Each portion was loaded onto a column packed with Source15 Phenyl resin (2 x 9 cm) and equilibrated with 10 mM TRIS, pH 7.5 containing 1 M ammonium sulphate. The column was operated at a flow rate of 5 ml/min. The enzyme activity was eluted with a linear gradient of 1 to 0 M ammonium sulphate over 8 column volumes. Fractions (10 ml) containing activity from the two runs were concentrated to 10 ml with a stirred cell concentrator equipped with a YM10 membrane  35  (Amicon, Billerica, MA) and loaded onto a Mono Q anion exchange column (1 x 8 cm) equilibrated with 10 mM TRIS, pH 7.5 containing 5% t-butanol, 2 mM dithiothreitol (DTT) and 0.25 mM ferrous ammonium sulphate. The column was operated at a flow rate of 2 ml/min. The enzyme activity was eluted with a linear gradient of 0.2 to 0.4 M NaCl over 20 column volumes. Fractions exhibiting activity were combined, exchanged into the column equilibration buffer, concentrated to 20-25 mg/ml protein, and flash frozen as beads in liquid N2. Purified HsaC was stored at –80ºC for several months without significant loss of activity. 2.4.2 Purification of HsaD and the S114A variant The cell pellet obtained from 4 liters of culture was resuspended in 40 ml of 20 mM HEPES, pH 7.5, and disrupted using three passages through a Emulsiflex C-5 cell disrupter (ATA Scientific, Sutherland, Australia) operating at a pressure of 15,000 p.s.i. The cell debris was removed by ultracentrifugation at 120,000 × g for 60 min. The supernatant was removed and filtered through a 0.45 µm cellulose filter (Sartorius AG, Göttingen, Germany) to yield 45 ml of raw extract. The raw extract was loaded onto 28 ml of Source 15Q anion exchange resin (GE Healthcare, Uppsala, Sweden) packed in an AP-2 column (Waters Corp., Milford, MA) equilibrated with 20 mM HEPES, pH 7.5. The proteins were eluted using a linear gradient of 70–220 mM NaCl in 280 ml. HsaD eluted at 140 mM NaCl. Fractions of 10 ml containing activity (in the case of the wild-type enzyme) or the protein of the expected size (in the case of the variant) were pooled and concentrated to 10 ml using an Amicon stirred cell concentrator equipped with an YM10 ultrafiltration membrane (Millipore, Billerica, MA). The HsaD-containing solution was brought to 1 M ammonium sulphate and the sample was briefly centrifuged to harvest the pellet. The cloudy white precipitate was  36  washed twice with 1 M ammonium sulphate solution; exchanged into 20 mM HEPES, pH 8.5, to resolubilize the enzyme; concentrated to >20 mg/ml; flash frozen as beads in liquid nitrogen; and stored at −80 °C. Coomassie Blue-stained denaturing gels revealed that preparations of the wild-type and variant proteins were >97% pure. 2.5 Protein analysis and handling Protein concentration was determined using the micro BCA protein assay (Pierce) according to kit instructions. SDS-PAGE was performed using separating gels containing 12% acrylamide (113). Gels were stained using Coomassie Brilliant Blue. 2.5.1 Handling of HsaC Aliquots of HsaC were thawed anaerobically immediately before use and exchanged into 20 mM HEPES, 80 mM NaCl (I = 0.1), pH 7.0 containing 5% t-butanol using a desalting column. Samples of HsaC were further diluted for enzyme kinetics using the same buffer containing 0.1 mg/ml BSA and 2 mM DTT, except in the inactivation experiments. For the latter, enzyme was diluted in the same buffer without DTT and were used within two hours. HsaC activity was verified at the beginning and end of each set of experiments. Protein and iron concentrations were evaluated colorimetrically using the Bradford method (117) and Ferene S (118), respectively.  37  2.6 Steady-state kinetic assays 2.6.1 Oxygraph assay for HsaC Activity of HsaC was routinely measured by following the consumption of O2 using a Clark-type polarographic electrode (Yellow Springs Instruments model 5031, Yellow Springs, OH). Experiments were performed in a thermojacketted Model RC1 respiration chamber (Cameron Instrument Company, Port Aransas, TX) equipped with a circulating water bath. The electrode signal was amplified using a Model OM200 O2 meter (Cameron Instrument Company, Port Aransas, TX) and recorded on a microcomputer equipped with a PC-LPM-16 multifunction board and Virtual Bench Data Logger (National Instruments, Austin, TX). Data were recorded every 0.1 s. Initial velocities were determined from progress curves by analyzing the data using Microsoft Excel (Microsoft Corp., Redmond, WA). For steady-state kinetic experiments and the determination of Km of O2, reaction buffers containing different concentrations of dissolved O2 were prepared; buffers were bubbled with humidified mixtures of O2 and N2 gases for at least 15 min prior to the experiment and transferring them to the reaction chamber. Ultra high purity O2 (100% or 10% O2 in N2) and 5.0 grade N2 were mixed in the desired proportions with a stainless steel model 561 gas proportioner (Concoa Virginia Beach, VA). The gas mixture was humidified by bubbling through a 5 x 10 cm column of buffer. The concentration of dissolved O2 in the reaction mixture was verified using the O2 electrode.  38  Standard steady-state reactions were performed in a total volume of 1.3 ml 20 mM HEPES, 80 mM NaCl, (I = 0.1 M) pH 7.0, 25.0 ± 0.1 ºC equilibrated with 5% O2 in N2 (103 ± 3 µM dissolved O2). The concentration of catecholic substrate ranged from 1µM to 60 µM. The reaction was initiated by injecting 2-10 µL of an appropriate dilution of enzyme preparation into the reaction chamber. The amount of active HsaC was defined by the iron content of the sample and was used in calculating kinetic parameters including specificity and catalytic constants. Buffers were used within 24 hours of preparation. Stock solutions were prepared fresh daily and stored under nitrogen on ice. Calibration of the O2 electrode was established by using standard concentrations of DHB with addition of excess BphC and zeroing with equilibrated buffer spiked with excess sodium hydrosulfite. Steady-state rate equations were fit to the data using the least squares and dynamic weighting options of LEONORA (119). One unit of enzymatic activity was defined as the quantity of enzyme required to consume 1 µmol of O2 per minute. 2.6.2 Spectrophotometric assay for HsaC Cleavage of catecholic substrates was also measured by following the rate of appearance of the ring-cleaved product using a Cary 5000 spectrophotometer equipped with a thermojacketed cuvette holder (Varian Inc., Walnut Creek, CA). Initial velocities were determined from a least-squares analysis of the linear portion of the progress curves using the kinetics module of the Cary WinUV software.  39  2.6.3 Determination of partition ratio Partition ratios expressing the number of substrate molecules consumed per enzyme molecule inactivated were determined spectrophotometrically for DHB, 2′,6′-diCl DHB, DHDS, and DHSA by following the appearance of the ring-cleaved products at 434 nm (ε = 23.4 mM−1cm−1), 391 nm (ε = 36.5 mM−1cm−1), 396 nm (ε = 6.3 mM−1cm−1), and 392 nm (ε = 7.6 mM−1cm−1), respectively. The partition ratio for 4-Cl DHDS was determined by oxygraph electrode due to a very low extinction coefficient. Partition ratios were determined under saturating substrate conditions ([S] >> Km). The amount of enzyme added to the reaction cuvette was such that the enzyme was completely inactivated before 15 % of either the substrate or O2 was consumed in the reaction mixture. The partition ratio was calculated by dividing the amount of product formed by the amount of active HsaC added to the assay (88).  Partition Ratio =  €  # substrate molecules consumed # enzyme molecules inactivated  2.6.4 Reversible inhibition of HsaC Steady-state inhibition assays were performed spectrophotometrically essentially as described in Section 2.6.2 except that 50 nM HsaC and 20 mM HEPES, 80 mM NaCl, pH 8.0, 25 ˚C were used. The concentration of PPB sulfonamide was varied between 0 and 32 µM and the concentration of DHB was varied from 2.5 to 500 µM. For steady state kinetic studies with gedunin, the concentration of inhibitor was varied from 0 to 500 nM while the concentration of DHB was varied as previously mentioned. Inhibition parameters were  40  evaluated by fitting Equation 1 to the data using the least-squares and dynamic weighting options of LEONORA (119).  v=  Vmax *[S] ⎛ [I] ⎞ K m ⎜1+ ⎟ + [S] ⎝ K ic ⎠  Equation 1  where Vmax is the maximum initial velocity, [S] is the substrate concentration, Km is the  € Michaelis-Menten constant, [I] is the inhibitor concentration and Kic is the competitive inhibition constant. To investigate whether compounds were reversible inhibitors of HsaC, 10 µM sulfonamide or 2 µM gedunin were incubated with 5 µM HsaC in 100 µL 20 mM HEPES, 80 mM NaCl pH 8.0 buffer at room temperature up to 1h aerobically and anaerobically. The enzyme:inhibitor solution was then diluted 50 times into 1 mL of the same air-saturated buffer before addition of 100 µM DHB to initiate the reaction. Activity of the incubated HsaC was monitored by following the production of HOPDA in pH 8.0 buffer at 434 nm (ε = 33.2 mM−1 cm−1) on a Cary 5000 spectrophotometer (Varian Inc., Walnut Creek, CA). Activity obtained was compared to enzyme without incubation with either compound and diluted to the same concentration. 2.6.5 Steady-state kinetic measurements of HsaD The enzyme-catalyzed hydrolysis of the MCP was monitored using a Varian Cary 5000 spectrophotometer equipped with a thermostatted cuvette holder (Varian Inc., Walnut Creek, CA) maintained at 25.0 ± 0.5 °C. The amount of enzyme used in each assay was  41  adjusted so that the progress curve was linear for at least 2 min. Initial velocities were determined from a least squares analysis of the progress curves using the kinetics module of the Cary WinUV software. Specificity experiments were performed in a total volume of 1.0 ml of potassium phosphate (I = 0.1 M), pH 7.5, at 25.0 ± 0.1 °C. The reaction was initiated by adding 5 µL of an appropriately diluted enzyme preparation to the reaction cuvette. The reactions were monitored at the wavelengths corresponding to the enolate. Initial velocities were determined over the following ranges of substrate concentration: 0.5–30 µM HOPDA, 10–300 µM HOPODA, and 2–60 µM DSHA. Steady-state kinetic parameters were evaluated by fitting the appropriate equations to the data using the least squares and dynamic weighting options of LEONORA (119). The kcat of S114A and WT HsaD for DSHA were determined by monitoring the decrease in absorbance at 405 nm using a 96-well plate reader (Victor plate reader, Perkin Elmer, Woodbridge, ON, Canada) at 25˚C. Assays were carried out in a total volume of 50 µL of 100 mM ionic strength potassium phosphate, pH 7.5, 25.0 ± 0.1 ˚C containing 5 µM S114A or 0.16 µM HsaD and 200 µM DSHA. Reactions were monitored over 30 min. The non-enzymatic decay of DSHA was subtracted from the progress curves and fitted using Excel (Microsoft Corp., Redmond, WA). 2.7 Stopped-flow spectrophotometry of HsaD or S114A with HOPDA Single turnover reactions for HsaD or S114A with HOPDA were performed with a 10:1 ratio of enzyme (20 µM) to substrate (2 µM) in potassium phosphate buffer (I = 0.1 M,  42  pH 7.5). Single turnover reactions of HsaD with HOPODA were performed with a 2:1 ratio of enzyme (16 µM) to substrate (8 µM) in the same buffer. Single turnover reactions were monitored using an SX.18MV stopped-flow reaction analyzer (Applied Photophysics Ltd., Leatherhead, UK) equipped with a photodiode array detector. The drive syringe chamber and optical cell were maintained at 25 °C by a recirculating water system. Multiple wavelength time-resolved spectra from single shots were acquired using the Pro-Data SX software (Applied Photophysics Ltd., Leatherhead, UK). Spectra shown are representative of at least five replicates of the stopped-flow drive syringe. Double or single-exponential equations were fit to the decrease in absorbance at 434 nm and 396 nm observed in the turnover of HOPDA and HOPODA, respectively, by HsaD using Origin 8.1 (OriginLab Corp, Northampton, MA). 2.8 Characterization of S114A:DHSA complex 2.8.1 Determination of Kd The dissociation constant of the S114A:DSHA complex was determined by titrating a 200 µL solution of potassium phosphate (I = 0.1 M), pH 7.5 at 25.0 ± 0.1 °C containing 5 µM DSHA with S114A and monitoring the change in absorbance at 456 nm. The spectra were recorded using a Cary 5000 spectrophotometer. The dissociation constant was evaluated by fitting the binding equation (120) to the data using the curve fitting program and the nonlinear statistical analyses of R.  43  2.8.2 Determination of half-life The half-life of S114A·DSHA was determined by monitoring the decay of the complex at 456 nm. The experiment was carried out in a total volume of 150 µL of potassium phosphate (I = 0.1 M), pH 7.5, at 25.0 ± 0.1 °C containing 14 µM S114A and 6 µM DSHA. A single-exponential equation was fit to the decrease at 456nm using the SX18MV stoppedflow reaction analyzer software (Applied Photophysics, Ltd. Leatherhead, U.K.) 2.9 High-throughput screening (HTS) for inhibitors of HsaC 2.9.1 Development of a HTS activity assay A high-throughput assay measuring HsaC activity was developed to screen for inhibitors. HsaC activity was measured by monitoring production of HOPDA. Moreover, HsaC for this assay was purified from cell extract using a SourceQ anion exchange column. Protein preparations obtained from this purification contained > 85% HsaC as judged using SDS-PAGE. Due to the oxygen-lability of HsaC activity, aliquots of frozen enzyme were thawed at the beginning of each day of screening and stock solutions were prepared under anaerobic conditions and stored on ice until use. DHB stock solutions were also prepared anaerobically daily and stored on ice to limit oxidation. HsaC (50 nM) was added to 10 µM DHB using a Microfill apparatus (BioTek, Winooski, VT) to a final volume of 50 µL buffer (20 mM HEPES, 80 mM NaCl, pH 8.0, 0.1 mg/mL BSA) in clear flat-bottomed 384-well plates (Corning, Corning, NY). Absorbance at 434 nm was measured before addition of HsaC (pre-read) and 8 min after addition of HsaC,  44  (read) using a VarioSkan (Thermo Scientific, Hudson, NH). Subtraction of the pre-reads from the reads allowed correction for naturally coloured compounds. Reactions were done at room temperature (~22 °C). The Km (Michaelis constant) under HTS conditions was evaluated using 0 to 200 µM DHB. Each plate included wells containing enzyme and substrate as positive controls and wells containing buffer instead of HsaC as negative controls. To determine the robustness of the assay, the Z′ factor was calculated according to Equation 2 (121):  Zʹ′ = 1 −  (3σc + +3σc −) µ c + − µc −  Equation 2  where σ and µ are the standard deviation and mean, respectively, of signal for both positive € (c+) and negative (c-) controls.  2.9.2 Primary screen Two libraries were screened: the 4,761-compound Known Drugs 2 (KD2) library (Prestwick, Washington, DC; Microsource, Gaylordsville, CT; Sigma-Aldrich, Oakville, ON; and, BIOMOL, Plymouth Meeting, PA), and the 49,940-compound DiverSet library (ChemBridge Corporation, San Diego, CA). Compounds were pinned into the destination 384-well plates using FP3 pins (Thermo Scientific, Hudson, NH) arrayed in a 384-pin head and a PlateMate Plus (Thermo Scientific, Hudson, NH). Each well contained 16.7 µM DHB in 30 uL 20 mM HEPES, 80 mM NaCl, pH 8.0. Approximately 5 µM of each compound was deposited in each well. Data were analyzed by GraphPad Prism software (GraphPad Software, San Diego, CA). Compounds were ranked by percent inhibition (% I) calculated by equation 3: 45  %I =  ( µc + −ΔAbs434 ) × 100% µc +  Equation 3  where µc+ is the average absorbance of the positive control wells of the plate and ΔAbs434 is € the corrected read. The top actives (>47% inhibition) were identified as primary hits and  were selected for counter screening. 2.9.3 Counter screen The top compounds of the DiverSet and KD2 libraries were selected for further testing. In the counter screen, compounds were pre-incubated with HsaC for 15 min at 25 ˚C in the reaction buffer before initiation of the reaction with substrate. Plates were read at 434 nm before and after HsaC addition to normalize for coloured compounds. 2.9.4 Dose-response curves The dose-response curves of the hits from primary screening were determined by HsaC activity assays in the presence of seven concentrations of inhibitor (0.033-30 µM) and 10 µM DHB. Inhibitor stock solutions were prepared in DMSO to a concentration of 5 mM. The final DMSO concentration was not more than 2% in each well, which did not affect enzyme activity. Reaction components were added and incubated similarly as the counter screen. 2’6’diCl-DHB and 4-Cl DHDS were also included in this analysis. The reads used for IC50 determination were taken 7 min post-addition of DHB. Each point was calculated as a percentage of HsaC inhibition based on the reaction rate in the absence of inhibitor. IC50 values for compounds exhibiting the sigmoidal semilogarithmic dose-response curves within the range of inhibitor concentrations tested were calculated using a 4-parameter non-linear  46  regression analysis (equation 4) and GraphPad Prism software (GraphPad Software, San Diego, CA).  %I = b +  a −b ⎛ [ I ] ⎞ s 1+ ⎜ ⎟ ⎝ IC 50 ⎠  Equation 4  where %I is the percent inhibition, a is the highest percent inhibition, b is the lowest percent  €  inhibition, [I] is the concentration of inhibitor, and s is the slope factor. For the cases where enzyme inhibition was not 0% and 100% at the lowest and highest concentrations, respectively, of inhibitor tested, a normalized non-linear regression analysis (equation 5) was employed:  %I =  €  100 ⎛ [ I ] ⎞ s 1+ ⎜ ⎟ ⎝ IC 50 ⎠  Equation 5  where %I is the percent inhibition, [I] is the concentration of inhibitor, and s is the slope factor. 2.10 In vitro growth of BCG wild-type and knockouts on cholesterol 2.10.1 Generation of BCG standard inoculum Standard inocula were prepared using 50 mL Middlebrook 7H9 broth (Fisher Scientific, Ottawa, Canada) with 0.05% tyloxapol (Sigma-Aldrich, Oakville, Canada) into roller bottles. After autoclaving, medium was supplemented with oleic acid (0.006 %),  47  bovine serum albumin (5 g/L), catalase (4 mg/L), sodium chloride (0.81 g/L) and sodium acetate (10 mM). The cultures were placed on a roller apparatus moving at 2 rpm at 37 ˚C and allowed to incubate until an optical density of 0.5 at 600 nm was reached. Cells were harvested, washed twice with 0.8 g/L sodium chloride and resuspended in Middlebrook 7H9 liquid media with 0.08% DMSO. Cultures were frozen in 1 mL aliquots in eppendorfs to be used as standard inoculum for further growth experiments. 2.10.2 Growth of BCG on various carbon sources To test for in vitro growth of wild-type BCG or knockouts strains (ΔhsaC and ΔhsaD), 100 mL Middlebrook 7H9 broth (Fisher Scientific, Ottawa, Canada) with 0.5% tyloxapol was prepared in roller bottles and autoclaved. If the medium contained cholesterol, it was added to the medium dissolved in tyloxapol prior to autoclaving. Cultures containing either no additional carbon source, 10 mM actetate, 0.5 mM cholesterol or both 10 mM acetate plus 0.5 mM cholesterol were prepared in triplicate and inoculated with the 1 mL standard inoculum of wild-type BCG or knockout strains (ΔhsaC and ΔhsaD) directly from a frozen glycerol stock, as described above. Growth was monitored by three methods: measuring absorbance at 600 nm (OD600); counting colony forming units (CFU); and determining protein concentration by BCA assay. CFU’s were determined by plating serial dilutions of cultures onto Middlebrook 7H10 agar (DifCo) supplemented with 10% (v/v) OADC and 0.5% (v/v) glycerol. To determine protein concentrations, cells were pelleted from 1 mL of culture to remove all supernatant and washed once with sterile saline (8 g NaCl per liter of distilled water). Pellets were boiled for 10 min in 100 µL of 1M NaOH before  48  protein concentrations were determined by BCA assay (Thermo Scientific, Rockford, IL) following manufacturer’s instructions.  49  CHAPTER 3: RESULTS 3.1 Characterization of cholesterol metabolites and analogues 3.1.1 Accumulation of DHSAs by the ΔhsaC mutant of RHA1 Bioinformatic analyses of the genes up-regulated in RHA1 during growth on cholesterol led to the predictions that HsaC catalyzes the extradiol cleavage of the catecholic ring of DHSA to DSHA and that HsaD hydrolyzes the latter MCP to HHD and DOHNAA (Figure 3) (18). To test this hypothesis, a ΔhsaC mutant of RHA1 was generated by Drs. Thomas Heuser and Matthew C. Anderton (18). The ΔhsaC mutant was incubated with cholesterol and metabolites in the medium were monitored using HPLC. Cultures of the ΔhsaC mutant incubated with cholesterol developed a pinkish colour after ~8 hours, indicative of the production of oxidized catechols and their condensation products. Catecholic metabolites were initially identified by formation of the yellow-coloured MCPs after addition of HsaC, obtained from expression in E. coli and purification, to culture supernatants and fractions. HPLC analyses of ΔhsaC culture supernatants identified two metabolites with different retention times (C18 column, 70:30 methanol/0.5% phosphoric acid solvent mixture) that absorbed at 280 nm, and disappeared upon addition of HsaC, and were thus transformed to yellow-coloured MCPs. GC-MS analysis of the first metabolite (Rt ~38 min) yielded fragments with m/z ratios of 193, 294, 381 and 590 (Figure 9; derivatized with TMS), consistent with 17-isopropionyl-DHSA (Figure 10). GC-MS analysis of the second metabolite (Rt ~21 min) yielded fragments with m/z ratios of 193, 251, 294 and 460 (Figure 11), consistent with DHSA (Figure 10).  50  Figure 9. GC-MS spectrum of TMS-derivatized 17-isopropionyl-DHSA. The parent ion (m/z = 590) corresponds to the complete derivatized molecule. The wavy line denotes the fragmentation of 17-isopropionyl-DHSA to yield one of the major daughter ions (m/z = 294). . COOH O Cl Cl  O  O Cl  OH  OH  OH  Cl  OH  OH  OH OH  OH  DHB  OH  OH  DHDS  DHSA  OH  2'6'-diCl DHB  Cl  4-Cl DHDS  OH  17-IsopropionylDHSA  Figure 10. The catecholic substrates used in characterizing HsaC.  51  Figure 11. GC-MS spectrum of TMS-derivatized DHSA. The parent ion (m/z = 460) corresponds to the complete derivatized molecule. The wavy line denotes the fragmentation of DHSA to yield one of the major daughter ions (m/z = 294). . 3.1.2 Production of DHSA The biotransformation of cholesterol to DHSA using the RHA1 ΔhsaC mutant was optimized for studies of Mtb HsaC and HsaD. To optimize conditions for DHSA production in W minimal media, the following parameters were varied: cholesterol concentration (0.5 - 2 mM), culture volume (0.5 – 2 L), aeration technique (baffled vs. non-baffled flasks), density of cell suspension (OD600 of 2-4), and incubation time. Even at the lowest concentration of cholesterol used (0.5 mM), most of the cholesterol remained insoluble as a cloudy suspension in the media consistent with its reported low solubility (<5 µM) and even lower critical micelle concentration in water (<40 nM) (122). DHSA production at an approximate rate of 52  0.1 mg/hour was monitored by HPLC. In general, maximum amounts of DHSA were obtained using a culture of ΔhsaC RHA1 cells pre-grown on 20 mM pyruvate and 0.5 mM cholesterol to exponential growth phase (OD600 = 0.5) in W minimal salt medium to induce the cholesterol catabolic pathway. The culture was concentrated 10-fold to an OD600 of 4 before incubation with 0.5 mM cholesterol. DHSA was purified using HPLC (Prodigy 10 µm ODS-Prep column, 44:56 methanol/0.5% phosphoric acid solvent mixture; Figure 12). This approach yielded approximately 1 mg of purified DHSA from a 1 L cell suspension.  Figure 12. HPLC trace of the ΔhsaC RHA1 culture supernatant incubated with cholesterol. Metabolites were eluted from a Prodigy 10 µm ODS-Prep column using a 44:56 methanol/0.5% phosphoric acid solvent mixture at flow rate of 1 mL/min. Absorbance was monitored at 280 nm. The asterisk indicates the peak corresponding to DHSA, as confirmed by GC-MS. The peak corresponding to 17-isopropionyl DHSA is not shown in this trace.  53  3.1.3 Chemical properties of DHSA The stability of DHSA and related catechols in air-saturated 20 mM HEPES, 80 mM NaCl, pH 7 at 25 ºC was evaluated by monitoring solutions at 400 nm, the absorption maximum of o-benzoquinone (112). Increases could be described by a single exponential decay, yielding similar half-lives for DHSA (1.5 h) and DHDS (2.3 h), a synthetic analogue. These dialkylated catechols are thus more susceptible to oxidation than DHB, the biphenyl catechol (5.5 h; Table 3). Table 3. Half-lives of catechols Substratea  t½ (h) b  DHB  5.5  DHDS  2.3  DHSA  1.5  a  Structures of the tested substrates are shown in Figure 10. Determined in air-saturated 20 mM HEPES, 80 mM NaCl, pH 7.0 at 25 ºC. Standard errors are < 15 %. b  3.1.4 Chemical properties of HOPODA and DSHA MCPs (Figure 13) were generated by incubating catechols with >20-fold HsaC. The absorption spectra of MCPs are strongly pH-dependent due to the 2-OH group, for which the pKa is typically ~7.3. The enol has an absorption maximum ~330 nm while the enolate anion is yellow coloured with an absorption maximum ~400 nm. The 38 nm blue-shift of the enolate of HOPODA and DSHA compared to HOPDA can be explained by the presence of the two-carbon bridge which reduces the degree of conjugation. The electron-donating  54  methyl group at C5 is likely a minor contributor to the decrease in extinction coefficient and enolate blue shift as shown in the 7 mM-1 cm-1 and 3 nm decrease, respectively, of 3-Me HOPDA compared to HOPDA (109). Finally, MCPs undergo a non-enzymatic transformation in aqueous buffer to gem-diol lactones (101). The half-lives of HOPODA and DSHA were less than 30 h, significantly less than that of HOPDA (Table 4). This is consistent with the shorter half-life of 3-Me HOPDA (30 h) (109). These data are summarized in Table 4 together with the pKa values of the enolic hydroxyl groups as determined by titration and the extinction coefficients of the respective enolate anions.  Figure 13. MCPs used in characterizing HsaD. Table 4. Properties of MCPs Substratea  λmax (nm)  ε (mM-1 cm-1)  Half-life (h)b  pKac  HOPDA  434  25.7  58 d  7.3 d  HOPODA  396  6.8  25  7.5  DSHA  396  3.8  15.4  6.8  a  Structures of the tested substrates are shown in Figure 13. Half-lives were determined in potassium phosphate buffer, (I = 0.1 M) pH 7.5 at 25 ºC. c The pKa values of the respective enolates were determined in unbuffered water. d Seah et al. (107). Standard errors are < 20 %. b  55  3.2 Characterization of HsaC 3.2.1 Expression and purification of HsaC To characterize HsaC from M. tuberculosis H37Rv, the enzyme was heterologously produced in E. coli and anaerobically purified to >99% apparent homogeneity using hydrophobic and anion exchange chromatographies as described in Section 2.4.1. The purified protein displayed a single protein band (Figure 14) with an apparent mass of 33.5 kDa, the expected mass of HsaC. Purified enzyme contained 0.92 equivalents of iron as determined by a Ferene S assay (118). Purified HsaC had a specific activity of 193 U/mg and was obtained in an overall yield of ~60% from raw extracts of recombinant E. coli. After the purified HsaC was flash frozen in liquid nitrogen at 20 mg/mL, the enzyme could be stored at -80 °C for up to 12 months without any significant loss of activity. However, enzyme preparations lost activity rapidly (t½ ~ 60 min) when diluted at room temperature in 20 mM HEPES, 80 mM NaCl, pH 7.5. To stabilize HsaC for kinetic assays, the enzyme was diluted in buffer supplemented with 5% t-butanol, 2 mM dithiothreitol, 0.1 mg/ml bovine serum albumin (BSA) and stored on ice under N2.  56  Figure 14. SDS-PAGE analysis of HsaC purification. Lanes are labelled: M, protein standard (MW indicated at left); 1, soluble extract of E. coli GJ1158 cells expressing HsaC; 2, after Source15 PHE hydrophobic interaction chromatography; and 3, after Source15 PHE hydrophobic interaction and Mono Q anion exchange chromatographies. The arrow indicates the band associated with HsaC enzyme activity. 3.2.2 Substrate specificity of HsaC The substrate specificity of HsaC was evaluated using steady-state kinetics. Due to the oxidative inactivation of both the enzyme and DHSA in air-saturated buffer, these studies were performed using buffer equilibrated with 5% O2 (~60 µM O2) in N2 to obtain better quality data. Under these conditions, HsaC had 90-times greater specificity for the steroid metabolite, DHSA, over DHB, the preferred substrate of BphC (76)(Table 5). To facilitate further kinetic characterization of HsaC, a substrate analogue, DHDS, was designed which incorporated potentially important features of DHSA including the methyl group on the catecholic ring and a saturated 2-carbon bridge between the two ring systems. The specificity  57  of HsaC for DHSA was 10-times greater than for DHDS. The substrate specificity of HsaC for 17-isopropionyl DHSA was not examined due to the lack of purified substrate from biotransformation. Table 5. Steady-state kinetic parameters of HsaC with various catecholic substrates Km  kcat  kcat/Km  (µM)  (s-1)  (µM-1s-1)  DHB  10 (2)  1.6 (0.1)  0.16 (0.02)  1,060 (10)  DHDS  4.8 (0.6)  6.7 (0.2)  1.4 (0.1)  2,300 (300)  DHSA  1.1 (0.2)  15.9 (0.6)  15 (2)  1,900 (200)  2’,6’-diCl DHB  6.7 (0.7)  6.6 (0.3) x 10-3  0.99 (0.04) x 10-3  30 (10)  4-Cl DHDS  3.2 (0.3)  277 (9) x 10-3  85 (6) x 10-3  44 (4)  Compounda  a  Partition Ratio  Structures of the tested substrates are shown in Figure 10.  3.2.3 Inactivation of HsaC by chlorinated catechols As extradiol dioxygenases are subject to oxidative inactivation during catalytic turnover (88), The susceptibility of HsaC to inactivation during the steady-state cleavage of each of the catecholic substrates was investigated using the partition ratio, the amount of substrate consumed per mole of enzyme inactivated. The partition ratios of HsaC for DHDS and DSHA were more than 2 orders of magnitude less than what has been reported for other extradiol dioxygenases for their preferred substrates (88,123). As for BphC, HsaC was more susceptible to inactivation by poorer substrates compared to DHB and DHSA, respectively (Table 5). In addition, 2’,6’-diCl DHB, a PCB metabolite that potently inhibits (Kic = 7 ± 1  58  nM) and oxidatively inactivates BphC (89), and 4-Cl DHDS, a chlorinated substrate analogue, were cleaved very slowly by HsaC and with partition ratios <50. While the Km values of HsaC for 2’,6’-diCl DHB and 4-Cl DHDS were ~1,000-fold greater than that of BphC, these compounds illustrate the potential of substituted catechols to inhibit and inactivate the mycobacterial dioxygenase. 3.2.4 Oxidative inactivation of HsaC The steady-state utilization of O2 by HsaC was evaluated in the presence of DHDS due to the availability of this compound. It is expected that the reactivity of the enzyme with O2 will be very similar in the presence of DHDS and DHSA as the two compounds have similarly substituted catecholic rings. The apparent KmO2 of HsaC was 90 ± 20 µM, 13-fold lower than that of BphC (1280 ± 70 µM) (124) and nearly 3-times lower than the concentration of O2 in air-saturated buffer. Nevertheless, the specificity of HsaC for O2 is only 5-times less than that of BphC (0.20 ± 0.01 µM-1 s-1 vs. 1.0 ± 0.1 µM-1 s-1). 3.3 High-throughput screening (HTS) for inhibitors of HsaC 3.3.1 HTS activity assay In the pursuit of a sensitive, robust activity assay for the high-throughput screening of HsaC for small molecule inhibitors, an assay was developed using DHB as a reporter substrate. DHB was chosen because it was commercially available and the strong absorption signal of its MCP product, HOPDA.  59  To optimize the assay, a number of parameters were tested including: HsaC concentration (25 to 100 nM), DHB concentration (0 to 200 µM), buffer conditions (20 mM HEPES, 80 mM NaCl at pH 7 and pH 8) and several additives known to stabilize HsaC activity (0.1 mg/mL BSA, 10% t-butanol, and 1 mM DTT). Buffer conditions at pH 8 increased the signal-to-noise ratio due to the higher extinction coefficient of the enolate product. Of the tested additives, 0.1 mg/mL BSA improved HsaC stability without inhibiting the rate of DHB cleavage. To minimize bias in identifying different types of inhibitors (competitive and uncompetitive), a concentration of substrate equal to the Km was used in the assay (125). The Km of HsaC for DHB under the assay conditions was approximately 10 µM. Overall, the optimized assay contained 50 nM HsaC, 0.1 mg/mL BSA, 10 µM DHB, 20 mM HEPES, 80 mM NaCl, pH 8.0. This conditions allowed single-point reads of absorbance 8 min after reaction initiation to approximate initial velocity (Figure 15). Assays were performed in a final volume of 50 µL in a 384-well plate format.  60  Figure 15. Representative progress curves for the positive and negative controls in the HsaC screening assay. The optimized assay in 384-well format contained 50 nM HsaC, 0.1 mg/mL BSA, 10 µM DHB in 20 mM HEPES, 80 mM NaCl pH 8.0. Absorbance changes at 434 nm in positive control wells, containing HsaC and DHB, and in negative control wells, containing DHB alone, are indicated by circles and triangles, respectively. The quality of HTS assays are evaluated using a statistical parameter called the Z’ factor (121). This value reflects both the dynamic range between the positive and negative controls and the variability of their means. A Z’-factor of 0.5 or greater is indicative of a good quality assay with a well-defined hit window. The mean of the Z’-factors of the primary screens of HsaC was > 0.6 indicating that the assay used to detect HsaC activity was robust and well adapted to HTS. 3.3.2 Actives identified from HTS The KD2 and DiverSet libraries were screened for inhibitors due to their high quality, known pharmacological properties, and diversity of natural and chemically synthesized compounds. KD2 library consists of 4,761 natural and chemically synthesized compounds with known pharmaceutical activities. The DiverSet library consists of 49,940 chemically synthesized, drug-like compounds and represents the broadest pharmacophore coverage possible. In the DiverSet library, 150 compounds had >50 % inhibitory activity against HsaC and these were chosen for further investigation. Compounds identified from a primary and counter screens are referred to as “actives”. A further 19 compounds with structural similarities to some of the top 30 actives were also included in the analysis. Finally, 5 compounds that did not inhibit HsaC were included as negative controls. These 174  61  compounds were subject to a replicate of the primary screen (pre-incubated with DHB) and a counter screen (pre-incubated with HsaC) to ensure reproducibility and robustness of the assay. Of these, 27 compounds inhibited HsaC activity >50% in the replicated primary screens and the counter screen (Figure 16). The majority of these actives were polycyclic compounds. Three of these were polycyclic sulfonamides (Figure 17). One of the latter, 4chloro-N-methyl-N-(4-[(4-methylpiperidin-1-yl)carbonyl]phenyl)  benzene-1-sulfonamide  (PPB sulfonamide, Figure 17C) was chosen for further investigation due to its strongest inhibition of HsaC compared to other polycyclic sulfonamides. Forty-nine compounds from the KD2 library inhibited HsaC >50%. Strikingly, six of the top eleven actives that inhibited HsaC >80% were gedunins, a family of compounds that share a five-ringed steroidal structure. Due to the large number of structurally similar actives from the KD2 library that inhibited HsaC > 50 %, a total of fifteen compounds out of the forty-nine, representing various structural classes of molecules, including three gedunins, were selected for replicate testing and counter screening. Of these, six reproducibly inhibited HsaC >50% (Figure 16).  62  DiverSet library  KD2 library  49,940  4,761  Primary screen  Primary screen  Compounds that have >50 % inhibitory activity against HsaC  Compounds that have >50 % inhibitory activity against HsaC  150  49 Repeated primary screen Counter screen  Grouped structures  Compounds that have >50 % inhibitory activity against HsaC  Representative structures  27  15  IC50 determination  Compounds that yielded IC50 values in the range of concentrations tested 9  Repeated primary screen Counter screen Compounds that have >50 % inhibitory activity against HsaC 6  IC50 determination  Compounds that yielded IC50 values in the range of concentrations tested 5  Figure 16. Flow-chart of the HsaC high-throughput screening of the DiverSet and KD2 libraries. Numbers in the boxes are the number of compounds that passed each screening step.  63  Figure 17. Structures of polycyclic sulfonamides that were top actives identified from HsaC screening of the DiverSet library. Compounds A, B and C have the same R-group shown on the right. Compounds A and B are N-(4-(azepane-1-carbonyl)phenyl)-4-chloro-Nmethylbenzenesulfonamide and 4-chloro-N-(4-(4-ethylpiperazine-1-carbonyl)phenyl)-Nmethylbenzenesulfonamide, respectively. Compound C is 4-chloro-N-methyl-N-(4-[(4methylpiperidin-1-yl)carbonyl]phenyl) benzene-1-sulfonamide, referred to as PPB sulfonamide. . The IC50 values were determined for the 6 actives from the KD2 library and the 27 from the DiverSet library. The dose-response curves of 14 of the 33 tested compounds exhibited the sigmoidal semilogarithmic curve associated with desirable inhibitors and yielded IC50 values of 0.7-4.3 µM (Table 6). Compounds that have been confirmed as inhibitors by dose-response curves are referred to as “hits”. A selected six IC50 curves are shown in Figure 18. A dose-response curve for 4Cl-DHDS yielded an IC50 value of 5.3 ± 0.5 µM (Figure 19). A dose-response curve for 2’6’-diCl DHB could not be determined due to the lack of inhibition at the highest concentration of compound tested.  64  Figure 18. Dose-response curves for six hits from the HsaC screening campaign of the KD2 and DiverSet libraries. A. 6,7-diCl 5,8-dihydroquinolinequione, B. gedunin, C. (R,R) cisdiethyl tetrahydro-2,8-chrysenedione, D. chloranil, E. mexicanolide, and F. PPB sulfonamide. IC50 values were calculated using nonlinear regression analysis of these curves. IC50 values are reported in Table 6. The corresponding structure of each compound is shown on each curve.  65  Table 6. IC50 values of hits from the HsaC screening campaign. CDRD compound ID  a  Compound  IC50 (µM)  SL-000-305-E09  6,7-dichloro-5,8-dihydroquinoline-5,8-dione  0.68 ± 0.03  SL-000-1221-C03  Gedunin  0.8 ± 0.2  SL-000-1247-M08  (R,R) cis-diethyl tetrahydro 2,8-chrysenediol  0.9 ± 0.3  Dihydrogedunin  1.1 ± 0.6  SL-000-1236-D03  Chloranil  1.3 ± 0.1  SL-000-1221-C15  Mexicanolide  1.7 ± 0.4  SL-000-779-D05  PPB sulfonamide  3.0 ± 0.8  4-((3-bromo-4-((2-fluorophenyl)methoxy)-5methoxyphenyl)methyl)amino phenol  3.05 ± 0.8  2-(((4-(dimethylamino)phenyl)amino)methyl)phenol  3.2 ± 0.6  N/A  3.5 ± 0.5  1-(2,4-dimethylphenyl)-N,5-dimethyl-N-phenyl-1H1,2,3-triazole-4-carboxamide  3.6 ± 0.4  N/A  3.7 ± 0.4  c  N/A  3.8 ± 0.4  SL-000-1741-F09  N/A  4.1 ± 1  SL-000-1221-A19  c  SL-000-2067-C09  b  SL-000-1997-D11  b  SL-000-775-H06  c  SL-000-681-F08 SL-000-2029-B09 SL-000-537-B02  c  a  Unique identification number assigned to each compound by the Center for Drug Research and Development. b Denotes compounds that did not reconfirm with newly purchased stock. c Denotes compounds that were not chosen for retesting.  Figure 19. The dose-response curve for 4-Cl DHDS. IC50 value of 5.3 ± 0.5 µM was calculated from the data using nonlinear regression analysis. The structure of 4-Cl DHDS is shown in the insert.  66  3.3.3 Inhibitor studies of HsaC PPB sulfonamide from the DiverSet library and gedunin from the KD2 library were identified as potent inhibitors of HsaC and were further characterized by steady-state kinetics. These compounds represent the two largest classes of molecules among the top actives identified by the screening campaign. Both compounds were re-confirmed as inhibitors using newly purchased and freshly made stock solutions using the counter screen format as previously described. To investigate their mechanisms of inhibition, PPB sulfonamide and gedunin were each preincubated with HsaC under aerobic or anaerobic conditions. Incubation of either compound at concentrations up to 10 µM for 1 hour did not affect the enzyme’s activity more than incubation with a buffer control. It was therefore concluded that the principle mode of inhibition was reversible. This was characterized further using steady-state kinetic experiments at pH 8 in which DHB was the catecholic substrate. Competitive, uncompetitive and mixed inhibition models were fit to the data. The best fit was obtained using the competitive model. The calculated Kic values for PPB sulfonamide and gedunin were 450 ± 50 nM and 80 ± 10 nM, respectively (Table 7, Figure 20A,B). Table 7. Competitive inhibition parameters of two reversible inhibitors of HsaC Compounda  Km (µM)  Kic (µM)  PPB Sulfonamide  3.1 ± 0.3  0.45 ± 0.05  Gedunin  2.3 ± 0.2  0.08 ± 0.01  a  The structures of PPB sulfonamide and gedunin can be found in Figure 18 F and B, respectively.  67  A  B  Figure 20. Dixon plots of steady-state analysis of PPB sulfonamide (A) and gedunin (B), two reversible inhibitors of HsaC. The rate of reaction was determined using 2 to 100 µM DHB. The calculated Kic values for PPB sulfonamide and gedunin were 450 ± 50 nM and 80 ± 10 nM, respectively. The concentration of HsaC (50 nM) was the same for both assays. Reactions were performed in 20 mM HEPES, 80 mM NaCl pH 8.0 buffer at 25 ˚C. Parameters were evaluated by fitting the competitive inhibition equation (Eq. 5) to the data using LEONORA. These best fit parameters were used to calculate the shown plotted lines.  68  3.4 Characterization of HsaD 3.4.1 Expression and purification of HsaD Recombinant HsaD was produced in E. coli and purified using anion exchange chromatography and ammonium sulphate precipitation as described in Section 2.4.2. Preparations of purified protein displayed a major protein band (Figure 21) with an apparent mass of 31.9 kDa, the expected mass of HsaD. A total of 120 mg of purified HsaD with a specific activity of 96 U/mg was obtained from 4 L of E. coli culture.  Figure 21. SDS-PAGE analysis of HsaD purification. Lanes are labelled: M, broad range protein standard (molecular weights indicated on left); 1, soluble extract of E. coli GJ1158 cells expressing HsaD; 2, after Source Q anion exchange chromatography; and 3, after anion exchange and ammonium sulfate precipitation. The arrow indicates the band associated with HsaD.  69  3.4.2 Substrate specificity of HsaD Steady-state kinetic studies were conducted to assess the substrate specificity of HsaD. Over the range of substrate concentrations studied, HsaD displayed Michaelis-Menten kinetics with all three MCPs. Consistent with the enzyme’s role in cholesterol degradation, HsaD had the highest specificity (kcat/Km) for DSHA, which was ~30-times higher than for HOPODA but only ~4-times higher than for HOPDA (Table 8). Although HsaD had a higher specificity for DSHA over HOPODA, the enzyme turned over these substrates with similar kcat values. Table 8. Steady-state kinetic parameters of HsaD with HOPDA, HOPODA and DSHA. Substratea  Km (µM)  kcat (s-1)  kcat/Km (mM-1s-1)  HOPDA  8 (1)  0.066 (0.005)  9 (1)  HOPODA  310 (70)  0.33 (0.06)  1.06 (0.06)  DSHA  17 (2)  0.55 (0.02)  33 (3)  a  Structures of the tested substrates are shown in Figure 13. Experiments were performed using potassium phosphate buffer (I = 0.1 M), pH 7.5 at 25 ºC. Values in parentheses indicate standard deviations. 3.4.3 Transient kinetic studies of HsaD To compare the catalytic mechanism of HsaD to other MCP hydrolases, HsaD was studied using a stopped-flow reaction analyzer equipped with a photodiode array detector. Studies could only be performed using HOPODA and HOPDA due to the scarcity of DSHA. Under single turnover conditions, absorption bands were observed at 434 nm for HOPDA (Figure 22) and 330 nm and 396 nm for HOPODA (Figure 23). Importantly, no significant 70  amounts of E:Sred were observed. In the BphD hydrolysis of HOPDA, this intermediate absorbs maximally at ~492 nm. The decrease in absorbance at 434 nm observed in the turnover of HOPDA could be fit to a double exponential equation with an initial decrease (1/τ1 = 7 ± 2 s-1) corresponding to ~3 % loss in the HOPDA absorbance, then a slower decrease (1/τ2 = 0.0489 ± 0.0006 s-1) corresponding to the kcat value (Table 8). Similarly, the decrease in absorbance at 396 nm observed in the turnover of HOPODA could be fit to a single exponential fit with 1/τ = 0.12 ± 0.04 s-1. This value was comparable to the corresponding kcat value (Table 8).  Figure 22. Time-resolved spectra of the hydrolysis of HOPDA by HsaD. The reaction contained 20 µM HsaD and 2 µM HOPDA in potassium phosphate, pH 7.5 (I = 0.1 M) at 25 ˚C. The black line indicates the spectrum of HOPDA and the coloured lines are spectra recorded on a log time-scale for a total of 100 s.  71  Figure 23. Time-resolved spectra from single turnover of HOPODA by HsaD. The reaction contained 16 µM HsaD and 8 µM HOPODA in potassium phosphate, pH 7.5 (I = 0.1 M) at 25 ˚C. The black line indicates the spectrum of HOPODA alone and the coloured lines are spectra recorded on a log time-scale for a total of 10 s. Decay beyond 10 s was not examined.  3.4.4 Characteristics of S114A in solution The mechanism of HsaD was further investigated using the catalytically impaired S114A variant of HsaD. The corresponding S112A variant of BphD forms a stable complex with HOPDA whose spectrum is similar to that of the E:Sred (102). S114A HsaD was prepared as described in section 2.2.2 and 2.4.2. Addition of a 60-fold excess of S114A to a 5 µM solution of DSHA produced a species with an absorbance maximum of 456 nm (Figure 24A), 60 nm red-shifted with respect to that of the DSHA enolate tautomer. This species had a half-life of 3 h (Figure 25), similar to that of the BphD S112A:HOPDA complex (102). Based on this, the upper limit of the kcat value of S114A for DSHA was 8.5 (± 0.9) x 10-5 s-1,  72  6500-fold lower than that of the WT enzyme. This slow turnover facilitated further study of the S114A:DSHA complex. Titration of the DSHA with S114A yielded a dissociation constant (Kd) of 51 ± 2 µM (Figure 24B). In comparison, titration of a 2 µM solution of HOPDA with the variant enzyme resulted in a spectral shift of ~17 nm in the presence of a 10-fold excess of S114A, most of the HOPDA (~90%) was present as the free enolate (Figure 26). Overall, the S114A:DSHA complex contributes to the spectroscopic characterization of the red-shifted intermediate.  Figure 24. Characterization of the S114A:DSHA complex by UV-visible absorption spectroscopy. (A) Spectra of S114A:DSHA and DSHA in solution are shown by dotted and solid lines, respectively. The complex absorbs maximally at 456 nm. (B) The curve represents the fit of the binding equation to ΔA456 observed upon titrating DSHA with S114A (Kd = 51 ± 2 µM). Experiments were performed using potassium phosphate, pH 7.5 (I = 0.1 M) at 25 °C. Reprinted with permission from Lack, N.A., Yam, K.C., Lowe, E.D., Horsman, G.P., Owen, R.L., Sim, E., and Eltis L.D. Characterization of a carbon-carbon bond hydrolase from Mycobacterium tuberculosis involved in cholesterol metabolism (2010) Journal of Biological Chemistry 285, 434-443.  73  Figure 25. Decay of S114A:DSHA complex at 456 nm. The complex was formed by mixing 14 µM S114A and 6 µM DSHA in potassium phosphate buffer (I = 0.1 M, pH 7.5) at 25°C. The solid line represents a single-exponential fit to the data.  Figure 26. Time-resolved spectra of HOPDA binding by HsaD S114A. The reaction contained 20 µM S114A and 2 µM HOPDA in potassium phosphate, pH 7.5 (I = 0.1 M) at 25 ˚C. The black line indicates the spectrum of HOPDA alone and the coloured lines are spectra recorded on a log time-scale for a total of 10 s. Subsequent decay was not detected on the time scale examined (<1 min). . 74  3.5 In vitro growth of BCG on cholesterol To assess the role of HsaC and HsaD in cholesterol metabolism, wild-type BCG and BCG deletion mutants of hsaC and hsaD were grown in 7H9 media supplemented with either no additional carbon, 10 mM acetate, 0.5 mM cholesterol or both 10 mM acetate plus 0.5 mM cholesterol. Carbon-sources in 7H9 minimal media include citrate and glutamate that could support limited growth in the absence of additional carbon, referred to herein as background growth. Importantly, the growth of both mutants on cholesterol was severely impaired. Whether measured by CFU or by protein concentration, growth was lower than that on acetate alone or background levels. This suggests that a toxic metabolite accumulated in each of these mutants. Likewise, the CFUs and protein levels further indicate that both the hsaC and hsaD mutants grew similarly on acetate or on 7H9 medium with no additional carbon source. From protein assays, wild-type BCG grew to similar levels in the presence of acetate or cholesterol, and this growth was significantly higher than background growth. A combination of both acetate and cholesterol appeared to double the rate of wild-type growth and the overall protein yield (Figure 27A). For reasons that are unclear, this increase in protein concentration did not correlate to an increase in CFU’s (Figure 27B). An explanation for this observed discrepancy might be due to cell clumping that can occur during serial dilution before plating on 7H10 agar. The inclusion of 0.05% Tween-80 or 0.5% Tyloxapol in the dilution medium could prevent bacterial clumping (67). However, the extent of clumping does not appear to be consistent between wild-type and mutant strains, since trends in CFUs of the mutants correlate with those from the protein assays. The hsaC and hsaD mutants may be less susceptible to clumping if these mutations resulted in altered cell wall  75  morphology. Incubation of BCG cells with DHB was previously reported to significantly diminished levels of complex lipids, including various mycolic acids (77). Nevertheless, both protein concentration and CFUs suggest that the ΔhsaD mutant grew as well on acetate and cholesterol as on acetate alone, while cholesterol slowed the growth of the ΔhsaC mutant in medium containing acetate. In addition to the attenuated growth phenotype of ΔhsaC strain in cholesterol containing media, cultures containing cholesterol developed a pink colour, starting at day 10, and became purple-brown over time. There was no significant colour change in cholesterolgrown cultures of the ΔhsaD mutant. This notable pink colour, also observed in the ΔhsaC mutant of RHA1 and Mtb (18,116), is characteristic of o-benzoquinones (oxidized catechols) and their condensation products.  76  Figure 27. Growth of wild-type M. bovis BCG, ΔhsaC, and ΔhsaD mutants on acetate, cholesterol or both acetate and cholesterol. Growth is monitored by milligrams of protein per milliliter of culture (A) and colony forming units (CFU’s) per milliliter of culture (B). BCG strains were grown on 7H9 media (containing 0.5 % Tyloxapol) with no added carbon source, 10 mM acetate, 0.5 mM cholesterol or 10 mM acetate plus 0.5 mM cholesterol. Values represent mean of triplicates. Error bars represent the standard error.  77  CHAPTER 4: DISCUSSION 4.1 The role of HsaC and HsaD in cholesterol catabolism The genetic and enzymological evidence presented in this thesis establishes that HsaC and HsaD catalyze essential reactions in the aerobic catabolism of cholesterol by actinomycetes. More specifically, HsaC catalyzes the extradiol cleavage of a cholesterol metabolite in which ring A is catecholic and HsaD catalyzes the subsequent hydrolysis of the resulting MCP. Firstly, the hsaC and hsaD deletion strains of BCG prevented growth on cholesterol but not acetate. This result mirrors those obtained for ΔhsaC mutants of R. jostii RHA1 (18) and Mtb H37Rv (116) (Figure 28A), respectively. Secondly, incubation of ΔhsaC of RHA1 on cholesterol resulted in two catecholic metabolites identified by GC-MS: DHSA and 17-isopropionyl-DHSA. These have the same core structure, with a catecholic ring A and intact bicycloalkanone C and D rings, differing only in the C17 substituent. Consistent with the accumulation of these catechols in the ΔhsaC mutants of RHA1, Mtb (Figure 28B) and BCG, cultures of these strains developed a pink colour when incubated in the presence of cholesterol. Such coloration develops from the exposure of catechols to O2, arising from the mixture of oxidation and polymerization products that result. Finally, of the tested substrates, HsaC and HsaD displayed highest specificity (kcat/Km) for DHSA and its MCP, DSHA, respectively. The growth phenotypes of the BCG knockouts further indicate that the accumulated steroid-derived catechol and MCP are toxic. The growth of the ΔhsaC and ΔhsaD mutants on cholesterol was severely attenuated and these strains exhibited levels of growth lower than  78  growth on acetate alone or minimal media. Consistent with these results, an ΔhsaC mutant of Mtb also lost viability in the presence of cholesterol, displaying a ten-fold decrease in colony-forming units (CFUs) over a 14-day growth experiment compared to the wild-type H37Rv while the mutant growth on glycerol was not impaired (Figure 28A) (116). Interestingly, the presence of acetate in the cholesterol-containing media could better rescue the ΔhsaD BCG mutant compared with the ΔhsaC mutant. This result suggests that the MCPs that accumulate in the ΔhsaD mutant are less toxic than the catechols that accumulate in the ΔhsaC mutant. Regardless, the toxicity of the corresponding biphenyl metabolites have been known for many years; knockouts of bphC and bphD in C. testosteroni showed an increase in percentage of cells containing two sets of chromosomes, evidence for inhibition of cell separation and reduction in cell viability (126). The cytotoxicity of these compounds can arise from three mechanisms: (a) redox cycling between quinones and catechols, which generates superoxide radicals and other reactive oxygen species (ROS) and hence causes oxidative stress; (b) depletion of essential cellular antioxidants such as glutathione and mycothiol which upsets the cellular redox balance; and (c) formation of DNA adducts which can lead to mutations during cell replication (127,128). The mechanism by which MCPs exert their toxicity is not known.  79  Figure 28. Growth of a ΔhsaC mutant of M. tuberculosis on cholesterol and in mice A. Growth of H37Rv strains in minimal media containing 0.1% (v/v) glycerol, 0.8% (v/v) isopropanol (solvent control), 0.02% (w/v) cholesterol, or no added carbon source. The plotted values represent the means of triplicates, with error bars indicating standard deviation. B. Accumulation of a colored metabolite during cholesterol utilization by the ΔhsaC mutant. C. Survival of SCID mice after intravenous infection with 105 CFU of wildtype H37Rv, the ΔhsaC mutant or the complemented ΔhsaC mutant, respectively (n = 10 mice per group). Taken from Yam, K.C., D’Angelo, I., Kalscheuer, R., Zhu, H., Wang, J.X., Snieckus, V., Ly, L., Converse, P.J., Jacobs, W.R., Jr., Strynadka, N. and Eltis L.D. Studies of a ring-cleaving dioxygenase illuminate the role of cholesterol metabolism in the pathogenesis of Mycobacterium tuberculosis (2009) Plos Pathogens e1000344. (116)  80  4.2 Physiological substrates of Hsa enzymes may be CoA thioesters Although HsaC and HsaD had the highest specificity (kcat/Km) for DHSA and DSHA, respectively, among the tested substrates, the current data indicate that these are not the enzymes’ physiological substrates. Firstly, the specificity of HsaC for DHSA (15 ± 2 µM-1 s1  ) and HsaD for DSHA (0.033 ± 0.003 µM-1 s-1) are significantly lower than those of other  extradiol dioxygenases and MCP hydrolases for their respective physiological substrates. For example, the homologues from the Bph pathway, BphC and BphD, have specificity constants of 62 ± 8 µM-1 s-1 and 22 ± 1 µM-1 s-1 for DHB and HOPDA, respectively (89,129). The relatively low specificity constants of HsaC and HsaD are consistent with what has been reported for other characterized cholesterol ring A/B-degrading enzymes in Mtb. Thus, the kcat/Km for KshAB (130) and HsaAB (21) are 300- and 350-fold less, respectively, than biphenyl dioxygenase of Burkholderia xenovorans LB400 (131) and p-hydroxyphenylacetate hydroxylase (HPAH) of Pseudomonas aeruginosa (132). Secondly, HsaC is unusually susceptible to oxidative inactivation during the catalytic turnover of DHSA. For example, the partition ratios of HsaC for DHSA was 50-fold less than that of BphC and its preferred substrate, DHB (124) and over 700-fold less than XylE for catechol (133). The mechanismbased inactivation of HsaC is likely similar to the O2-dependent inactivation of BphC during the turnover of poor substrates. EPR and absorption spectroscopy studies demonstrated that in the latter enzyme, inactivation arises mainly from the dissociation of superoxide from the enzyme:ligand:O2 ternary complex with concomitant oxidation of the active site Fe(II) to Fe(III) (88).  81  In considering the identity of the physiological substrates of HsaC and HsaD, one possibility may be substrates with a side-chain at C-17. This conjecture arises from the suggestion that side chain degradation and ring cleavage are thought to occur concurrently in at least some actinomycetes (21,134,135). In Mtb, side chain degradation is initiated by oxidation of C27 to a carboxylic acid by a cytochrome P450 (25,136). The 8-carbon branched alkyl side chain is then shortened successively by 3, 2, and 3 carbons. The extent to which side chain degradation must occur before ring cleavage begins is not known. Regardless of its length, the partially degraded side chain metabolites are predicted to exist as CoA-thioesters during the degradation process. Thus, the physiological substrates of HsaC and HsaD may possess a 3-, 5- or 8-carbon side chain at C17 (Figure 29A), or a CoA thioester thereof (Figure 29B), depending on how many rounds of β-oxidation had occurred.  Figure 29. Side chains of potential cholesterol metabolites. A.CoA-thioester metabolites generated during cholesterol catabolism are predicted to have an 8-, 5- or 3-carbon side chain at C17. For simplicity, only ring D and part of ring C are shown. B. The chemical structure of coenzyme A.  82  The occurrence of 17-isopropionyl-DHSA in the culture supernatant of cholesterolgrown ΔhsaC RHA1 suggests that ring and side chain degradation occur concurrently to at least some extent during cholesterol catabolism in this strain. Similarly, a ΔhsaA mutant of RHA1 accumulated HSA with an isopropionate moiety at C17, consistent with partial ring degradation before complete side chain degradation (21). These 17-isopropionyl metabolites were not tested as substrates for HsaC and HsaA due to a lack of purified material. However, recent experiments with 17-isopropionyl 4-AD indicate that longer chain metabolites are better substrates for KshAB (J. K. Capyk and L. D. Eltis, unpublished). Currently, there is no direct evidence for the occurrence of cholesterol metabolites as CoA-thioesters. Moreover, the substrates of ring-degrading oxygenases in bacteria are usually not CoA-thioesters. Interestingly, CoA-thioesters are metabolites in the aerobic catabolism of phenylacetate by the Paa pathway, described in RHA1 and other bacteria (137). Indeed, an aromatic acyl-CoA is the substrate of the pathway’s multi-component oxygenase, PaaGHIJK (138,139). 4.3 Insights into substrate specificity of HsaC and HsaD from their structures Crystal structures of HsaC were obtained in its substrate-free form and in complex with DHSA at resolutions of 2.0 and 2.2 Å, respectively, in collaboration with Natalie Strynadka (UBC) (116). The asymmetric unit of the crystals contains two well-ordered molecules. Crystallographic four-fold symmetry of the two molecules in the asymmetric unit indicates the enzyme is octameric, like BphC. Crystallographic statistics are summarized in Table 9 (Appendix). The overall two-domain fold of HsaC consists of four copies of βαβββ motif analogous to other members of the type I extradiol dixoygenase family, with the structure most closely resembling that of BphC (79) (Figure 30). The active site is located  83  within the central cavity of the slightly larger C-terminal domain, with the catalytically essential mononuclear Fe2+ ligated by His145, His215 and Glu266. In the resting state enzyme, two solvent molecules (wat1 and wat2) complete the coordination sphere such that the metal ion's coordination geometry is square pyramidal. The catechol moiety of DHSA in the active site exhibits asymmetric bidentate binding to Fe(II) (Figure 31); the longer Fepromixal hydroxyl (O4) bond compared to the Fe-distal hydroxyl (O3) bond is consistent with observations in BphC (88) and homoprotocatechuate 2,3-dioxygenase (HPCD) (140), indicating that the catechol is bound as a monoanion. However, the substrate-binding pocket of HsaC is larger than that of BphC, presumably to accommodate the bulkier bicycloalkanone rings of steroid metabolites.  Figure 30. Structure of HsaC:DHSA complex. The Cα traces of the structurally similar Nand C-terminal domains are colored in silver and dark green, respectively. As in other twodomain type I extradiol dioxygenases, the active site is located in the C-terminal domain. The iron ion is colored orange. The C and O atoms of the bound DHSA are grey and red,  84  respectively. Taken from Yam, K.C., D’Angelo, I., Kalscheuer, R., Zhu, H., Wang, J.X., Snieckus, V., Ly, L., Converse, P.J., Jacobs, W.R., Jr., Strynadka, N. and Eltis L.D. Studies of a ring-cleaving dioxygenase illuminate the role of cholesterol metabolism in the pathogenesis of Mycobacterium tuberculosis (2009) Plos Pathogens e1000344. (116).  Figure 31. DHSA in the HsaC active site. The (2Fo – Fc) electron density (blue, contour level = 1σ) was calculated without ligand to remove bias. HsaC carbons and DHSA carbons are coloured green and yellow, respectively. Oxygen, nitrogen and iron atoms are coloured red, blue and orange, respectively. Taken from Yam, K.C., D’Angelo, I., Kalscheuer, R., Zhu, H., Wang, J.X., Snieckus, V., Ly, L., Converse, P.J., Jacobs, W.R., Jr., Strynadka, N. and Eltis L.D. Studies of a ring-cleaving dioxygenase illuminate the role of cholesterol metabolism in the pathogenesis of Mycobacterium tuberculosis (2009) Plos Pathogens e1000344. (116). To characterize the interactions of HsaD with DSHA, crystallization studies were undertaken using a catalytically impaired S114A variant in collaboration with Edith Sim (Oxford) (141). Crystal structures were obtained of S114A in complex with DSHA at 1.9 Å resolution. The statistics of the data collection and refinement for these structures are summarized in Table 10 (Appendix). Briefly, the HsaD structure comprises a core α/β hydrolase domain with a four-helix lid domain (Figure 32). The core domain is composed of  85  three anti-parallel β-sheets followed by 5 parallel strands, surrounded by 5 α-helices. The active site, situated between the core and lid domains, is the location of the well-conserved catalytic triad: Ser114 (replaced by Ala), His 269 and Asp 241.  Figure 32. Secondary structure and B-factor of the S114A variant of HsaD. The secondary structure of S114A was color-coded according to Cα B-factor from blue (lowest B-factor: 6.0 Å2) to red (highest B-factor: 67.4 Å2). Reprinted with permission from Lack, N.A., Yam, K.C., Lowe, E.D., Horsman, G.P., Owen, R.L., Sim, E., and Eltis L.D. Characterization of a carbon-carbon bond hydrolase from Mycobacterium tuberculosis involved in cholesterol metabolism (2010) Journal of Biological Chemistry 285, 434443.(141) In the HsaD S114A:DSHA complex, the ligand is held in a strained non-planar orientation in the active site (Figure 33), similar to that observed for HOPDA binding in BphD S112A (102). The dienoate was bound to the P subsite lined with Asn54, Asn113, Arg192 and Trp270. The “oxyanion hole” formed by backbone amides of Gly45 and Leu115, hydrogen bonds with the C6-oxo group of DSHA to help to position the substrate in close proximity with the active site residue 114 and increase the ligand’s reactivity to ketonization and eventual cleavage. The bicycloalkanone moiety of DSHA is located within the NP 86  subsite, a hydrophobic region formed by Phe212, Met208, Leu115 and Val155. In general, the NP subsite of HsaD is at least twice as a large as the corresponding cavity in BphD LB400 and four-fold larger than the cavities in MhpC and BphD RHA1, presumably to fit the larger cholesterol metabolite DSHA (142).  Figure 33. Overlay of the MCP substrates HOPDA, HOPODA and DSHA in complex with the S114A variant of HsaD. The protein (white) is shown in with the side chains involved in ligand binding shown as sticks. The ligands are represented as sticks and are colored pink (HOPDA), cyan (HOPODA) and purple (DSHA). In all structures N, O and S atoms colored blue, red and yellow respectively. Reprinted with permission from Lack, N.A., Yam, K.C., Lowe, E.D., Horsman, G.P., Owen, R.L., Sim, E., and Eltis L.D. Characterization of a carbon-carbon bond hydrolase from Mycobacterium tuberculosis involved in cholesterol metabolism (2010) Journal of Biological Chemistry 285, 434-443. A striking feature of the HsaC:DHSA and HsaD S114A:DSHA structures is that the C17 keto group is orientated towards the solvent, away from the binding cavity (Figure 34 A,B). This orientation essentially alleviates any restriction on the length of the side chain at C17. Further inspection of the structures reveals that there are no positively charged amino acid residues that could interact with the carboxylate of a C17 isopropionyl substrate. By  87  contrast, the HsaC:DHSA and S114A:DSHA structures have patches of positively charged residues on their surfaces within 20 Å of the C17 keto that could stabilize the negatively charged phosphate groups of a CoA moiety at the predicted distance. Such residues include Arg173 and Arg191 in HsaC and Lys156, Lys160 and Arg206 in HsaD. Overall, the structural data indicate that these enzymes could accommodate substrates possessing a long CoA-thioester side chain. Nevertheless, biochemical evidence is needed to support the hypothesis that the substrates of the ring degradation enzymes are CoA-thioesters.  Figure 34. Cavity of HsaC and HsaD S114A can accommodate substrates with longer side chains. Cartoon images illustrating a portion of the binding pocket in HsaC:DHSA (A) and HsaD S114A:DSHA (B) showing space available for a substrate with extended side chain and CoA-thioester at C17, indicated by an asterisk. DHSA and DSHA are coloured yellow and cyan, respectively. Surfaces are coloured by electrostatic potential; red and blue represent negative and positive charges, respectively.  88  4.4 Inhibitors of HsaC Two classes of HsaC inhibitors were identified in this study: mechanism-based and competitive. The former include 2’,6’-diCl DHB and 4-Cl DHDS which inactivated HsaC during catalytic turnover. Nevertheless, their respective modes of action likely differ, reflecting steric and electronic considerations. 2’6’-DiCl DHB is an example of a mechanism-based inhibitor of HsaC and likely acts in the same manner as in BphC. As previously demonstrated, 2’,6’-diCl DHB strongly inhibits BphC (Kic = 7 ± 1 nM) due to partial occlusion of the likely O2-binding site by one of the chloro substituents (89). The O2binding site, defined by Val148, Phe187 and Ala198 is conserved in HsaC (Val147, Phe192 and Ala203) and catecholic substrates bind in the same orientation in the two enzymes relative to this site. 2’,6’-DiCl DHB does not inhibit HsaC as effectively as BphC, likely due to the poorer fit of the dichlorinated phenyl ring into the active site. In contrast, 4-Cl DHDS likely inactivates HsaC through a combination of steric and electronic factors. The electronwithdrawing chlorine at C4 on the catecholic ring of 4-Cl DHDS may sterically hinder binding and reduce electron transfer from the catecholic substrate to the Fe ion. This basis of inactivation has been reported in a range of extradiol enzymes, including BphC (88) and human 3-hydroxy-anthranilate-3,4-dioxygenase (HAD) (143). Inactivation studies of BphC with 3-chloro and 3-methyl catechols (88) suggests that catechols with electron-withdrawing substituents were effective mechanism based inactivators for the reasons mentioned above. Additionally, structural studies of 4-Cl 3-hydroxyanthranilate (ClHAA) bound to HAD, an enzyme essential to the biosynthesis of quinolinate from tryptophan, suggests that the chloro  89  substituent at C4 inhibits catalysis by causing unfavourable steric interactions and preventing correct positioning of the substrate in the active site (143). 4-Chloro-N-methyl-N-(4-[(4-methylpiperidin-1-yl) carbonyl] phenyl) benzene-1sulfonamide (PPB sulfonamide) and gedunin (Figure 18F and B, respectively), identified using a high-throughput screen, are competitive inhibitors with submicromolar Kics, consistent with tight binding (Table 7). Polycyclic sulfonamides are synthetic compounds grouped according to their shared sulfonamide linker between cyclic moieties. There is no published biological study on polycylic sulfonamides as therapeutics. Gedunin and its derivatives are a group of tetranortriterpenoids consisting of a 4,4,8-trimethyl-17furanylsteroidal skeleton, bearing several oxygenated functionalities. Isolated from the Indian neem tree (Azadirachta indica) and Bangladeshi mangrove tree (Xylocarpus granatum), gedunin displays anti-malarial (144), antifungal (145) and anti-cancer (146,147) properties. The molecular basis of the anti-cancer activities of gedunins lies in their binding to the Hsp90 folding machinery(146). Anti-tubercular activity by this group of compounds has not been reported. As with all the top hits identified from the HTS campaign, PPB sulfonamide and gedunin share multiple-ring structures that bear some resemblance to the substrates of HsaC. Thus, gedunin is structurally related to steroids. Similarly, PPB sulfonamide has three rings and a two-atom linker similar to DHSA. Many polycyclic compounds are inhibitors of enzymes whose substrates are steroids. Polycyclic pyridine amide derivatives, structurally similar to the polycyclic sulfonamides, selectively inhibit 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) which catalyzes the interconversion of inactive cortisone to active  90  cortisol, the principal glucocorticoid in humans (148). Interestingly, a sulfonylmethyl derivative (Figure 35B) potently inhibits 11β-HSD1 (IC50 of 35 nM), despite a marked departure from the sterol structure of the cortisone substrate (Figure 35A). A crystal structure of 11β-HSD1 complexed with a related pyridine amide compound indicates that the inhibitor binds in the same active site pocket as cortisone and exhibits at least two binding conformations (148). Although the binding mode of PPB sulfonamide and gedunin by HsaC is not known, the competitive mode of inhibition indicates that these compounds bind in the same active site as DHSA. The mode of binding of gedunin to HsaC was analyzed in collaboration with Natalie Strynadka’s lab (UBC) using an automatic docking program, Autodock Vina (149), with standard settings (K.C. Yam, R.J. Gruninger, L. Worrall, N. Strynadka and L.D. Eltis, unpublished). The docking procedure yielded nine possible binding conformations of gedunin in the HsaC active site, which were ranked by minimized binding energies (kcal/mol). The conformation with the lowest binding energy (-8.1 kcal/mol) was also the conformation that corresponded best with the overlay of DHSA (Figure 36). In this conformation, the furan ring of gedunin is closest to the Fe ion (within <5 Å), and the steroid rings B and C make extensive interactions with hydrophobic residues Leu190, Leu205, Val214 and Phe294. These same residues contact the bicycloalkanone moiety of DHSA (116). Notably, hydrogen bonds (<3 Å) are predicted between oxo group on the D-ring and the furanyl oxygen of gedunin with His215 and His257, respectively. The docking provided additional support for gedunin as a competitive inhibitor of HsaC since binding of this molecule is predicted to occupy the same space as DHSA. It is not clear if gedunin could  91  bind in multiple conformations, however, crystal structures of HsaC:gedunin complexes could identify the molecular basis of inhibition and provide a good starting point for structural optimization. Considering the similarity of gedunin to cholesterol and other substrates of the cholesterol ring-cleaving enzymes, gedunin may strongly inhibit other cholesterol catabolic enzymes, particularly those such as KshAB whose substrates are structurally more similar to gedunin.  Figure 35. Compounds bound by human 11β-HSD1. Cortisone (A) is transformed by 11βHSD1, which activates to cortisol, the principal glucocorticoid in humans. The enzyme is strongly inhibited by a sulfonylmethyl analogue of pyridine amide (B).  92  Figure 36. Lowest binding energy conformation of gedunin in HsaC active site by automated docking. Gedunin (A) and DHSA (B) are shown in the active site of HsaC, in the same orientation. The overlay of gedunin and DHSA (C) is shown without HsaC residues. Gedunin and DHSA are coloured yellow and cyan, respectively. Carbons of the HsaC residues are coloured green. Oxygen, nitrogen and iron atoms are coloured red, blue and orange, respectively. Although gedunins have not been reported to have anti-tubercular effects, several natural and synthetic steroids have been investigated as potential anti-tubercular drugs. Extracts of Morinda citrifolia (Rubiaceae), a traditional remedy for tuberculosis and respiratory infection in the Philippines contained a number of steroids (150). While most steroids extracted exhibited mild to moderate anti-tubercular activity, a 2:1 mixture of steroids 1 and 2 (Figure 37) isolated from this plant was more potent (MIC < 2 µg/mL). A  93  steroid enone (Figure 37, 3) isolated from Ruprechtia triflora, an Argentinian plant, exhibited a similar MIC value of 2 µg/mL (151). The mechanism by which these steroids inhibit Mtb growth is unknown, but it is possible that they target one or more cholesterol catabolic enzymes. A 6-azasteroid (Figure 37, 4) was found to be a tight binding competitive inhibitor (IC50 = 0.5 µM and Ki = 100 nM) of 3β-HSD, the dehydrogenase that catalyzes the first step of cholesterol ring degradation (152). An example of a non-steroid as a good inhibitor of an enzyme in a steroid pathway includes a 1,2,4-triazole-3-thione (Figure 38) that has been identified as an inhibitor of an arylamine N-acetyltransferase (NAT), in the cholesterol catabolic gene cluster (153). This compound mimics the effects of a Δnat strain: inhibiting growth of BCG and Mtb with an MIC of <10 µg/mL and modifying the lipid profile of these cells. The HsaC inhibitor studies along with the studies mentioned above describe the first characterized inhibitors of cholesterol pathway enzymes and lead the way to further develop novel therapies targeting the cholesterol pathway. So far, MIC values of the most potent steroid inhibitors tested are at least 10-times greater in comparison to rifampicin, a commonly used TB therapeutic, indicating these steroid compounds are overall less potent. With some modifications to improve potency, steroids and polycyclic sulfonamides may be useful scaffolds for the design of new anti-tuberculosis agents.  94  Figure 37. Steroid skeleton of potential sterol anti-tuberculars from plant extracts. Adapted from (150-152)  Figure 38. The 1,2,4-triazole-3-thione compound, an inhibitor of NAT and a potential antitubercular (153). 4.5 HsaD mechanism The spectroscopic and structural data of the S114A:DSHA complex establish that the mechanism of HsaD is typical of MCP hydrolases. As presented in the section 1.4.2.2, MCP hydrolases utilize a nucleophilic mechanism in which the enolate substrate is first tautomerized to a keto tautomer (Figure 39). During this tautomerization, an ESred intermediate is formed. The catalytic serine acts as the nucleophile to generate a covalent acyl-enzyme intermediate (Figure 39). Deacylation of this intermediate appears to be ratelimiting. The 60 nm red-shift associated with the ESred species in S114A:DSHA observed by UV-Vis spectroscopy is similar in magnitude to that observed in the S112A:HOPDA  95  complex for BphD (72 nm) (93,154). Inspection of the crystal structure of the S114A:DSHA complex indicates that this substrate is held in a similar non-planar orientation in the active site as HOPDA in BphD. The identity of ESred remains unclear. The S114A:DSHA structural data are consistent with both the 2-oxo, 6-oxo and the 2-oxo, 6-oxido tautomers (Figure 40, 2 and 3): both refined without significant deviation from planarity of the C3-C4 double bond. This is consistent with structural studies of DxnB2 S105A:5,8-diF HOPDA, an MCP hydrolase in the dibenzofuran degradation pathway of Sphingomonas wittichii RW1 (Figure 41) (S. Bhowmik, K.C. Yam, J.T. Bolin and L.D. Eltis, in preparation). Unfortunately, it is not possible to distinguish between these tautomers from the crystallographic data alone. However, the UV-Vis spectra of the S114A:DSHA is most consistent with the 2-oxo, 6oxido tautomer as it is fully conjugated so would absorb in the visible range. By contrast, the ketonized HOPDA is predicted to result in a blue shift (maximum absorbance of < 250 nm) based on the spectroscopic signatures of its component moieties, acetophenone and 2-keto-3enoate (104). As has been previously noted, the red-shift observed in the turnover of DSHA and HOPDA by HsaD and BphD, respectively, is similar to the 60 nm red-shift of the Lphotointermediate of the retinal chromophore of bacteriorhodopsin (155). The latter has been explained on the basis of a 40˚ twist about double bonds of the L-photointermediate, a distortion of the chromophore that has recently been observed crystallographically (156). Based on this study, the enolate tautomer (Figure 40, 1) was ruled out since refinement of this tautomer introduced a 67˚ deviation from planarity of the torsion angle about the C4-C5 double bond (Table 11, Appendix). Previous structural studies of the BphD S112A:HOPDA  96  complex also rejected the possibility of a bound enolate based on torsional angle restraints (102). In considering the 2-oxo, 6-oxido stereoisomer, some strain would be required to obtain a shift of 60 nm associated with the ESred species in S114A:DSHA. The similar electronic absorption spectra of S114A:DSHA and other E:Sred species formed in various homologues indicate the likely occurrence of a strained 2-oxo, 6-oxido species as a catalytic intermediate and imply that the serine residue plays a key role in tautomerization of the bound MCP, perhaps by protonating C5.  E:Sred Gly45 N H  Leu115 HO  O  NH  O  N Leu115  CO2  N  Trp270 R  O  R  Trp270 O  Asn113 H OH  O  CO2  O  O  Ser114  Arg192  Gly45  Arg192  Ser114  NHis269  OH  Asn113  H NHis269  DOHNAA  2-oxo 6-oxido intermediate -OOC  OH  DSHA  Gly45 O  Gly45 Leu115  Leu115  N H  NH  Arg192  NH  Trp270  R  O  H  O  OH R  O O  Ser114  CO2  O  NH  O  Gly45 H NHis269  Leu115 NH  Asn113  H NHis269  Ser114  N H  O  H O  O  H  CO2  R O Ser114  NHis269  OH  HHD  Acyl enzyme  Figure 39. The proposed mechanism of HsaD. The identity of E:Sred is likely a 2-oxo, 6oxido species.  97  Figure 40. Possible conformations of DSHA in the HsaD S114A active site. MCP ligands were refined as enolate (1) and 2-oxo, 6-oxo (2), E:Sk, tautomers based on the electron density map of the ligands. A 2-oxo, 6-oxido stereoisomer (3), is consistent with the electron density and the absorption spectrum of S114A:DSHA. The bicycloalkanone rings of DSHA are represented by “R”.  Figure 41. Different tautomers of 5,8-diF HOPDA fit to the electron density. The structures of 2-oxo, 6-oxo (left) and 2-oxo, 6-oxido (right) are compatible with the torsion angle restraints and the electron density of the ligand in the DxnB2 S105A:5,8-diF HOPDA crystal structure. Lack of ESred in HOPODA and HOPDA turnover suggests that formation of the redshifted intermediate is rate-limiting for these poor substrates. Stopped-flow data presented herein demonstrate little to no formation of a red-shifted intermediate during turnover of HOPDA and HOPODA by HsaD, but the decays at 434 nm and 396 nm, respectively, corresponded to their respective kcat values. This result is consistent with an absence of the red-shifted intermediate in a BphD S112A:3-Cl HOPDA complex (109) and during turnover of 3-Cl HOPDA by DxnB2 (S. Bhowmik, K.C. Yam, J.T. Bolin and L.D. Eltis, in 98  preparation). In these two cases, the additional steric bulk at C3 was found to stabilize an alternate binding planar mode, termed ESe, in which the substrate is planar; a crystal structure of BphD:3-Cl HOPDA revealed a non-productive binding mode where formation of ESred and tautomerization are impaired (109).	
  While the present structure of S114A:HOPDA and S114A HOPODA complexes do not show that ligands are held in the planar binding mode to the same extent, the rate-limiting step for the HsaD turnover of HOPDA and HOPODA is likely tautomerization and formation of ESred. 4.6 Role of cholesterol metabolism in Mtb pathogenicity 4.6.1 Cholesterol catabolism may be essential for Mtb pathogenicity As part of the current study, the survival of ΔhsaC and ΔhsaD mutants of Mtb was studied in infection models by our collaborators at Albert Einstein College of Medicine. Both mutants displayed an attenuated phenotype in severe combined immunodeficient (SCID) mice. Animals intravenously infected with 105 CFUs of the ΔhsaC mutant survived ~50% longer than mice infected with wild-type Mtb H37Rv while complementation of ΔhsaC restored the virulent phenotype (Figure 28C) (116). Similarly, SCID mice infected with 104 CFUs of the ΔhsaD mutant survived almost 100% longer than wild-type infected animals (R. Kalsheuer and W. Jacobs Jr, unpublished data, Figure 42). Although a stronger growth defect of the cholesterol-grown ΔhsaC BCG mutant versus that of the ΔhsaD mutant can be attributed to greater toxicity of the accumulated catecholic metabolites, it is clearly not the only factor mediating attenuation in the Mtb knockouts. The unexpected weaker knockout phenotype of the ΔhsaC mutant can be attributed to experimental differences in CFU dosage.  99  Survival of ΔhsaC mutant is also moderately attenuated in the guinea pig model of infection. Overall, animals aerosol-infected with ~102 CFUs of the ΔhsaC mutant had lower bacterial loads in lung at 8 weeks, representing the chronic phase of infection, and slower dissemination to the spleen compared to the animals infected with the wild-type (Figure 43A). Granulomas that were observed in the lungs of guinea pigs infected with the mutant were smaller and less necrotic (Figure 43B, C), however, infection continued to persist. The attenuation of the ΔhsaC mutant in the animal models may be explained by the accumulation and toxicity of oxidized catechols and quinones mentioned in the previous sections. However, it is unknown whether the metabolite that accumulates in the ΔhsaD mutant may exhibit toxicity to the same extent in the guinea pig model of infection. The precise role and importance of cholesterol catabolism in the pathogenesis of Mtb remains unclear. Some phenotypes of knockouts of cholesterol catabolic enzymes suggest that these enzymes are essential for Mtb pathogenesis. Thus, ΔkshA and ΔkshB mutants did not kill SCID mice and were rapidly killed by immunocompetent mice (71). The authors suggest that AD and ADD accumulation are not cytotoxic to cells, but may be transformed into immunomodulatory steroids that mediate the initial stages of infection. However, most of the evidence is not as clear-cut. The toxicity of a steroid-derived metabolite has been observed in an Δigr and a Δcyp125 mutant in strain CDC1551 (24,157). In each of these cases, toxicity in the presence of cholesterol was attributed to the accumulation of cholest-4en-3-one. Aerosol infection of mice with the Δigr mutant delayed growth and dissemination to the spleen and reduced lung pathology, but no change in persistence was observed (158). For example, 3β-HSD catalyzes the first committed step in ring degradation and is absolutely  100  required for in vitro growth of Mtb on cholesterol (28). However, the virulence phenotype of a Δ3βhsd strain was similar to that of the wild-type in activated macrophages and in a guinea pig model (72). In addition, fadA5, tentatively annotated as encoding a side-chain cleaving enzyme is required for cholesterol metabolism, but is not required for growth of Mtb in mice (67). These results are in line with those obtained for the ΔhsaC and ΔhsaD mutants, in which the phenotype in SCID mice is due to toxicity. Indeed, it is likely that Mtb cocatabolizes multiple carbon sources as demonstrated by recent metabolomic studies (159). In fact, in the restrictive environment of a macrophage, host lipids and other long-chain fatty acids are abundant; persistence of Mtb in macrophages and mouse models requires isocitrate lyase (ICL), the key enzyme in the glyoxylate shunt required for fatty acid catabolism (160,161). Genomic differences in various Mtb strains may translate into varying control of cholesterol metabolic regulation between these strains. Discovery of a compensatory cholesterol 27-hydroxylase, CYP142, in Δcyp125 of H37Rv but not in CDC1551 nor in BCG (26), suggests that mutations affecting the cholesterol pathway may occur more frequently than previously believed. The hypervirulent member of the W-Beijing lineage, HN878, or an extensively drug resistant (XDR) strain may exhibit tighter metabolic regulation of the cholesterol pathway, compared to the H37Rv standard lab strain. Therefore, it is possible that an HsaC knockout in a different lab strain or clinical strain, may exhibit a stronger phenotype due to a stronger dependence on the cholesterol pathway during infection.  101  Figure 42. Survival of SCID mice after intravenous infection with 104 CFUs wild-type H37Rv, the ΔhsaD mutant or the complemented ΔhsaD mutant, respectively. (n = 8 mice per group).  102  Figure 43. Growth of the ΔhsaC mutant in the guinea pig model of tuberculosis. A. Growth kinetics in the lung and spleen of guinea pigs aerosol-infected with H37Rv, ΔhsaC mutant, or the complemented ΔhsaC mutant (n = 15 guinea pigs per group). Asterisks indicate significant (*p,0.05) or highly significant (**p,0.01) differences found between animals infected with the wild-type and ΔhsaC mutant strain. B. Gross pathology of guinea pig lungs infected with wild-type, mutant, and complemented strains at week 4 and week 8. C. Histopathological appearance of same lung specimens as those depicted in B. Taken from Yam, K.C., D’Angelo, I., Kalscheuer, R., Zhu, H., Wang, J.X., Snieckus, V., Ly, L., 103  Converse, P.J., Jacobs, W.R., Jr., Strynadka, N. and Eltis L.D. Studies of a ring-cleaving dioxygenase illuminate the role of cholesterol metabolism in the pathogenesis of Mycobacterium tuberculosis (2009) Plos Pathogens e1000344. (116) 4.6.2 Cholesterol metabolism may modulate the host’s immune response In addition to an energy source and as building blocks for virulence lipids, cholesterol metabolites are implicated in modulating the host’s adaptive immune response. In general, the resistance against infectious extracellular microorganisms is often associated with a Th2 humoral response in which B-cells are activated and antibodies specific to the pathogen are produced (162). In contrast, an effective defense against intracellular organisms such as viruses and Mtb require a Th1 response, which is dominated by cell-mediated forms of immunity characterized by cellular cytolytic activity and production of cytokines IFN-γ and TNF (162). However, Mtb successfully infects and persists in macrophages due to an ineffective Th2 immune response recruited against tubercle bacilli rather than a protective Th1 immune response. As mentioned in the introduction, antiglucocorticoids (sex steroids) such as dihydroepiandrosterone (DHEA) are protective in a mouse TB model whereas, the Th1 immune response is sensitive to inactivation by glucocorticoids (GC) (62). Six mycobacterial genes of eukaryotic origin are thought to be specific for steroid metabolism and are hypothesized to drive the balance to higher levels of GC steroids though the inactivation of host sex steroids, thus causing a sustained Th2 immune response and prolonged infection (163). One of these genes identified from this study, Rv3548, a putative 17β-hydroxyestradiol dehydrogenase, is part of the cholesterol catabolic cluster controlled by KstR2, a transcriptional repressor.  104  Interestingly, regulation of genes involved in cholesterol catabolism and β-oxidation appear to vary with the changing phagocytic environment; these genes are upregulated to a greater extent after infection of dendritic cells compared to macrophages (164). While both cell types are central to anti-mycobacterial immunity, they serve distinct roles during the infection process. Alveolar macrophages play a scavenger role, engulfing foreign inhaled particles, however, they are poor activators of naive T cells. Dendritic cells, arguably the most appropriate human cell model to study Mtb infection, have a much wider role: they are able to recognize and phagocytose pathogens at infection sites, secrete cytokines, then process and present antigens to native lymphocytes. Thus, the differential metabolism of cholesterol by mycobacteria in dendritic cells compared to macrophages may implicate cholesterol metabolites in modulating the host’s adaptive immune response. The mechanism by which Mtb influences the host immune response is unknown, but one can speculate that it may occur via interaction of the cholesterol metabolites with the vitamin D receptor (VDR) expressed in almost all immune cells. It has been known since the mid-1900s that activated vitamin D, 1,25(OH)2D3, induces anti-mycobacterial activity in monocytes and macrophages (165). More recent studies describe the induction of cathelicidin, an antimicrobial peptide, by 1,25(OH)2D3. Cathelicidin is required for antimicrobial activity against intracellular Mtb (166-168). Given the structural similarity of ring B-cleaved metabolites to vitamin D analogues (Figure 44), metabolites generated from the cholesterol degradation pathway by the pathogen may shift the balance from a Th1 to Th2 immune response in a VDR-mediated fashion. More specifically, sterol substrates of HsaAB and HsaC possessing an intact side chain are potential ligands of the VDR and may  105  act as an immune modulator. Further research is required to understand the mechanism of the antimicrobial effect of vitamin D regarding Mtb infection and persistence.  106  Figure 44. Structure of 1,25(OH)2D3 (calcitriol) and of selected analogues with immunoregulatory properties. Reprinted from Vol 8, C. Mathieu and L. Adorini, The coming of age of 1,25-dihydroxyvitamin D3 analogs as immunomodulatory agents, 174-179 (2002) with permission from Elsevier (169).  107  4.7 Future directions This thesis describes the characterization of two cholesterol catabolic enzymes in Mtb H37Rv, a bacterial pathogen.  Using molecular genetic, biochemical and structural  approaches, HsaC and HsaD were identified as an extradiol dioxygenase and a MCP hydrolase, respectively, in the cholesterol catabolic pathway. There remains some uncertainly about the nature of the physiological substrates of these enzymes given the low specificity of HsaC and HsaD for cholesterol metabolites, DHSA and DSHA, respectively, and the unusual oxidative susceptibility of the former enzyme. As mentioned in section 4.2, physiological substrates of the cholesterol catabolic pathway enzymes may be CoA-thioesters. Steady-state kinetics of HsaC and HsaD with CoA-thioesters of DHSA and DSHA, respectively, would provide insight into the identity of the true physiological substrates of these enzymes. The questions of whether and why cholesterol metabolism is essential for Mtb virulence needs to be resolved. Immunological studies could be performed with macrophages and dendritic cells infected with ΔhsaC or ΔhsaD mutants versus wild-type strains. To investigate the host immune response upon infection by these mutants, profiles of secreted cytokines could be analyzed by enzyme-linked immunosorbant assay (ELISA) as described by Jang et. al. (170). Analysis of specific cytokines such as IFN-γ, IL-2 and TNF-α versus IL-4 and IL-5 may indicate the extent of induction of Th1 versus Th2 immunity, respectively. In this experiment, the use of hsaC and hsaD knockouts in more virulent Mtb strains should be explored to determine the role of cholesterol metabolism in virulence. As demonstrated by the contrasting growth phenotypes of Δcyp125 in H37Rv and CDC1551, genomic differences among various Mtb strains may be correlated to dependence upon cholesterol catabolism.  108  Although the precise role of cholesterol catabolism in Mtb pathogenesis remains to be established, several mechanism-based and competitive inhibitors of HsaC were identified in this study.  2’6’-DiCl DHB and 4-Cl DHDS likely inhibit HsaC based on steric and  electronic considerations. In contrast, development of a high-throughput HsaC assay has led to the identification of PPB sulfonamide and gedunin, two competitive inhibitors with low Kic values (<500 nM). Interestingly, the chemical structure of PPB sulfonamide and gedunin differ significantly from the catecholic cholesterol metabolites identified from cholesterolgrown ΔhsaC RHA1 cells. In fact, gedunin and polycyclic sulfonamides may also be inhibitors of upstream ring-cleaving enzymes of the pathway, such as KstD, HsaAB and KshAB, given their similarity to substrates of these enzymes. Assays of KstD, HsaAB and KshAB have been developed; therefore it would be straightforward to test inhibition of these enzymes by these small molecules. Previous studies of the inhibition of BphD with chlorinated HOPDAs formed during PCB degradation indicate the potential of HsaD inhibitors to be developed as TB therapeutics. As mentioned in the introduction, 3-Cl and 4-Cl HOPDAs competitively inhibit BphD displaying Kic values in the low micromolar range (108). Presumably, appropriately substituted derivatives of DSHA would competitively inhibit HsaD. A colorimetric HTS assay, similar that the one developed for HsaC, can be developed for HsaD to screen for small molecule inhibitors. The assay would be based on the decrease at 434 nm associated with HsaD-catalyzed hydrolysis of HOPDA. While the exact molecular basis of inhibition of HsaC by gedunins and polycyclic sulfonamides remains unknown, the inhibition studies presented here indicate that these  109  compounds competitively prevent binding of catecholic substrates. To determine if gedunin is bacteriostatic or bacteriocidal against mycobacteria, growth studies of BCG and Mtb in the presence of gedunin should be performed; the minimum inhibitory concentration (MIC) could be determined from these inhibition studies. Mtb strains that could be tested for gedunin susceptibility include the CDC1551 strain, a MDR or XDR strain. In addition to potentially increased dependence on cholesterol metabolism, the drug-resistant strains are often less fit and less transmissible (171). In parallel, toxicity of this compound to macrophages will be evaluated to confirm that the inhibitory effect is specific for bacterial cells. Given that gedunin was the tightest binding inhibitor identified from the HTS campaign and this compound could be docked in the HsaC active site, crystallographic studies of HsaC in complex with gedunin have been initiated. These structural studies will provide insight into the structure-activity relationship (SAR) and lead to the design of small molecules with improved selectivity and potency. 	
    110  BIBLIOGRAPHY 1.  Volkman, J. K. (2003) Appl Microbiol Biotechnol 60, 495-506  2.  Pearson, A., Budin, M., and Brocks, J. J. (2003) Proc Natl Acad Sci U S A 100, 15352-15357  3.  Kannenberg, E. L., and Poralla, K. (1999) Naturwissenschaften 86, 168-176  4.  Yeagle, P. L. (1985) Biochim Biophys Acta 822, 267-287  5.  Spector, A. A., and Yorek, M. A. (1985) J Lipid Res 26, 1015-1035  6.  Cooper, R. A. (1978) J Supramol Struct 8, 413-430  7.  Waksman, S. A., and Henrici, A. T. (1943) J Bacteriol 46, 337-341  8.  Turfitt, G. E. (1944) Biochem J 38, 492-496  9.  Dodson, R. M., and Muir, R. D. (1961) Journal of the American Chemical Society 83, 4627-&  10.  Kieslich, K. (1985) J Basic Microbiol 25, 461-474  11.  Sih, C. J., Tai, H. H., and Tsong, Y. Y. (1967) J Am Chem Soc 89, 1957-1958  12.  Owen, R. W., Mason, A. N., and Bilton, R. F. (1983) J Lipid Res 24, 1500-1511  13.  Marsheck, W. J., Kraychy, S., and Muir, R. D. (1972) Appl Microbiol 23, 72-77  14.  Horinouchi, M., Hayashi, T., Yamamoto, T., and Kudo, T. (2003) Appl Environ Microbiol 69, 4421-4430  15.  van der Geize, R., Hessels, G. I., van Gerwen, R., Vrijbloed, J. W., van der Meijden, P., and Dijkhuizen, L. (2000) Appl Environ Microbiol 66, 2029-2036  16.  van der Geize, R., and Dijkhuizen, L. (2004) Curr Opin Microbiol 7, 255-261  17.  McLeod, M. P., Warren, R. L., Hsiao, W. W., Araki, N., Myhre, M., Fernandes, C., Miyazawa, D., Wong, W., Lillquist, A. L., Wang, D., Dosanjh, M., Hara, H., 111  Petrescu, A., Morin, R. D., Yang, G., Stott, J. M., Schein, J. E., Shin, H., Smailus, D., Siddiqui, A. S., Marra, M. A., Jones, S. J., Holt, R., Brinkman, F. S., Miyauchi, K., Fukuda, M., Davies, J. E., Mohn, W. W., and Eltis, L. D. (2006) Proc Natl Acad Sci U S A 103, 15582-15587 18.  van der Geize, R., Yam, K., Heuser, T., Wilbrink, M. H., Hara, H., Anderton, M. C., Sim, E., Dijkhuizen, L., Davies, J. E., Mohn, W. W., and Eltis, L. D. (2007) Proc Natl Acad Sci U S A 104, 1947-1952  19.  Pandey, A. K., and Sassetti, C. M. (2008) Proc Natl Acad Sci U S A 105, 4376-4380  20.  Mohn, W. W., van der Geize, R., Stewart, G. R., Okamoto, S., Liu, J., Dijkhuizen, L., and Eltis, L. D. (2008) J Biol Chem  21.  Dresen, C., Lin, L. Y., D'Angelo, I., Tocheva, E. I., Strynadka, N., and Eltis, L. D. (2010) J Biol Chem 285, 22264-22275  22.  Szentirmai, A. (1990) J Ind Microbiol Biotechnol 6, 101-115  23.  Sih, C. J., Tai, H. H., Tsong, Y. Y., Lee, S. S., and Coombe, R. G. (1968) Biochemistry 7, 808-818  24.  Ouellet, H., Guan, S., Johnston, J. B., Chow, E. D., Kells, P. M., Burlingame, A. L., Cox, J. S., Podust, L. M., and de Montellano, P. R. (2010) Mol Microbiol 77, 730-742  25.  Capyk, J. K., Kalscheuer, R., Stewart, G. R., Liu, J., Kwon, H., Zhao, R., Okamoto, S., Jacobs, W. R., Jr., Eltis, L. D., and Mohn, W. W. (2009) J Biol Chem 284, 3553435542  26.  Driscoll, M. D., McLean, K. J., Levy, C., Mast, N., Pikuleva, I. A., Lafite, P., Rigby, S. E., Leys, D., and Munro, A. W. (2010) J Biol Chem 285, 38270-38282  27.  Horinouchi, S., Ishizuka, H., and Beppu, T. (1991) Appl Environ Microbiol 57, 13861393  28.  Yang, X., Dubnau, E., Smith, I., and Sampson, N. S. (2007) Biochemistry 46, 90589067  29.  Yue, Q. K., Kass, I. J., Sampson, N. S., and Vrielink, A. (1999) Biochemistry 38, 4277-4286  112  30.  Sampson, N. S., and Vrielink, A. (2003) Acc Chem Res 36, 713-722  31.  Lee, S. S., and Sih, C. J. (1967) Biochemistry 6, 1395-&  32.  Miclo, A., and Germain, P. (1992) App Microbiol Biotechnol 36, 456-460  33.  Horinouchi, M., Hayashi, T., and Kudo, T. (2010) J Steroid Biochem Mol Biol  34.  Laskin, A. I., Fried, J., Meyers, C. D. L., and Grabowich, P. (1964) Journal of Medicinal Chemistry 7, 406-&  35.  Garnier, T., Eiglmeier, K., Camus, J. C., Medina, N., Mansoor, H., Pryor, M., Duthoy, S., Grondin, S., Lacroix, C., Monsempe, C., Simon, S., Harris, B., Atkin, R., Doggett, J., Mayes, R., Keating, L., Wheeler, P. R., Parkhill, J., Barrell, B. G., Cole, S. T., Gordon, S. V., and Hewinson, R. G. (2003) Proc Natl Acad Sci U S A 100, 7877-7882  36.  Hermon-Taylor, J. (2009) Gut Pathog 1, 15  37.  Weinstock, D. M., and Brown, A. E. (2002) Clin Infect Dis 34, 1379-1385  38.  Navas, J., Gonzalez-Zorn, B., Ladron, N., Garrido, P., and Vazquez-Boland, J. A. (2001) J Bacteriol 183, 4796-4805  39.  de Chastellier, C., and Thilo, L. (2006) Cell Microbiol 8, 242-256  40.  Brennan, P. J., and Nikaido, H. (1995) Annu Rev Biochem 64, 29-63  41.  Barry III, C. E., Lee, R. E., Mdluli, K., Sampson, A. E., Schroeder, B. G., Slayden, R. A., and Yuan, Y. (1998) Progress in Lipid Research 37, 143-179  42.  Dye, C., Bassili, A., Bierrenbach, A. L., Broekmans, J. F., Chadha, V. K., Glaziou, P., Gopi, P. G., Hosseini, M., Kim, S. J., Manissero, D., Onozaki, I., Rieder, H. L., Scheele, S., van Leth, F., van der Werf, M., and Williams, B. G. (2008) Lancet Infect Dis 8, 233-243  43.  Wright, A., Bai, G., Barrera, L., Boulahbal, F., Martín-Casabona, N., Gilpin, C., Drobniewski, F., Havelková, M., Lepe, R., Lumb, R., Metchock, B., Portaels, F., Rodrigues, M., Rüsch-Gerdes, S., Van Deun, A., Vincent, V., Leimane, V., Riekstina, V., Skenders, G., Holtz, T., Pratt, R., Laserson, K., Wells, C., Cegielski, P., and Shah, N. S. (2006) Emergence of Mycobacterium tuberculosis with Extensive Resistance to  113  Second-Line Drugs --- Worldwide, 2000-2004. . in M.M.W.R. Morb. Mortal. Wkly. Rep. , Centers for Disease Control and Prevention, Atlanta, GA 44.  Mason, M., and Mendoza, M. (2009) First Case of Highly Drug-Resistant TB Found in US. in The Associated Press, Lantana, FL  45.  Cole, S. T., Brosch, R., Parkhill, J., Garnier, T., Churcher, C., Harris, D., Gordon, S. V., Eiglmeier, K., Gas, S., Barry, C. E., 3rd, Tekaia, F., Badcock, K., Basham, D., Brown, D., Chillingworth, T., Connor, R., Davies, R., Devlin, K., Feltwell, T., Gentles, S., Hamlin, N., Holroyd, S., Hornsby, T., Jagels, K., Krogh, A., McLean, J., Moule, S., Murphy, L., Oliver, K., Osborne, J., Quail, M. A., Rajandream, M. A., Rogers, J., Rutter, S., Seeger, K., Skelton, J., Squares, R., Squares, S., Sulston, J. E., Taylor, K., Whitehead, S., and Barrell, B. G. (1998) Nature 393, 537-544  46.  Fleischmann, R. D., Alland, D., Eisen, J. A., Carpenter, L., White, O., Peterson, J., DeBoy, R., Dodson, R., Gwinn, M., Haft, D., Hickey, E., Kolonay, J. F., Nelson, W. C., Umayam, L. A., Ermolaeva, M., Salzberg, S. L., Delcher, A., Utterback, T., Weidman, J., Khouri, H., Gill, J., Mikula, A., Bishai, W., Jacobs Jr, W. R., Jr., Venter, J. C., and Fraser, C. M. (2002) Journal of bacteriology 184, 5479-5490  47.  Pym, A. S., Brodin, P., Brosch, R., Huerre, M., and Cole, S. T. (2002) Molecular Microbiology 46, 709-717  48.  Reyrat, J. M., and Kahn, D. (2001) Trends Microbiol 9, 472-474  49.  Stinear, T. P., Seemann, T., Harrison, P. F., Jenkin, G. A., Davies, J. K., Johnson, P. D., Abdellah, Z., Arrowsmith, C., Chillingworth, T., Churcher, C., Clarke, K., Cronin, A., Davis, P., Goodhead, I., Holroyd, N., Jagels, K., Lord, A., Moule, S., Mungall, K., Norbertczak, H., Quail, M. A., Rabbinowitsch, E., Walker, D., White, B., Whitehead, S., Small, P. L., Brosch, R., Ramakrishnan, L., Fischbach, M. A., Parkhill, J., and Cole, S. T. (2008) Genome Res 18, 729-741  50.  Altaf, M., Miller, C. H., Bellows, D. S., and O'Toole, R. (2010) Tuberculosis 90, 333337  51.  Via, L. E., Deretic, D., Ulmer, R. J., Hibler, N. S., Huber, L. A., and Deretic, V. (1997) J Biol Chem 272, 13326-13331  52.  Russell, D. G. (2007) Nat Rev Microbiol 5, 39-47  114  53.  Kim, M. J., Wainwright, H. C., Locketz, M., Bekker, L. G., Walther, G. B., Dittrich, C., Visser, A., Wang, W., Hsu, F. F., Wiehart, U., Tsenova, L., Kaplan, G., and Russell, D. G. (2010) EMBO Mol Med 2, 258-274  54.  Hunter, R. L., Olsen, M., Jagannath, C., and Actor, J. K. (2006) Am J Pathol 168, 1249-1261  55.  Gatfield, J., and Pieters, J. (2000) Science 288, 1647-1651  56.  Munoz, S., Rivas-Santiago, B., and Enciso, J. A. (2009) Scand J Immunol 70, 256263  57.  Peyron, P., Bordier, C., N'Diaye, E. N., and Maridonneau-Parini, I. (2000) J Immunol 165, 5186-5191  58.  Vasanji, A., Ghosh, P. K., Graham, L. M., Eppell, S. J., and Fox, P. L. (2004) Dev Cell 6, 29-41  59.  Pagel, W., and Pagel, M. (1925) Virchows Arch Path Anat 256, 629-640  60.  Yang, X., Nesbitt, N. M., Dubnau, E., Smith, I., and Sampson, N. S. (2009) Biochemistry 48, 3819-3821  61.  Martens, G. W., Arikan, M. C., Lee, J., Ren, F., Vallerskog, T., and Kornfeld, H. (2008) Infect Immun 76, 3464-3472  62.  Hernandez-Pando, R., De La Luz Streber, M., Orozco, H., Arriaga, K., Pavon, L., AlNakhli, S. A., and Rook, G. A. (1998) Immunology 95, 234-241  63.  Chen, W., Chen, G., Head, D. L., Mangelsdorf, D. J., and Russell, D. W. (2007) Cell Metab 5, 73-79  64.  Javitt, N. B. (2002) J Lipid Res 43, 665-670  65.  Hansson, M., Ellis, E., Hunt, M. C., Schmitz, G., and Babiker, A. (2003) Biochim Biophys Acta 1593, 283-289  66.  DuSell, C. D., Umetani, M., Shaul, P. W., Mangelsdorf, D. J., and McDonnell, D. P. (2008) Mol Endocrinol 22, 65-77  115  67.  Nesbitt, N. M., Yang, X., Fontan, P., Kolesnikova, I., Smith, I., Sampson, N. S., and Dubnau, E. (2010) Infect Immun 78, 275-282  68.  Schnappinger, D., Ehrt, S., Voskuil, M. I., Liu, Y., Mangan, J. A., Monahan, I. M., Dolganov, G., Efron, B., Butcher, P. D., Nathan, C., and Schoolnik, G. K. (2003) J Exp Med 198, 693-704  69.  Rengarajan, J., Bloom, B. R., and Rubin, E. J. (2005) Proc Natl Acad Sci U S A 102, 8327-8332  70.  Sassetti, C. M., Boyd, D. H., and Rubin, E. J. (2003) Mol Microbiol 48, 77-84  71.  Hu, Y., van der Geize, R., Besra, G. S., Gurcha, S. S., Liu, A., Rohde, M., Singh, M., and Coates, A. (2010) Mol Microbiol 75, 107-121  72.  Yang, X., Gao, J., Smith, I., Dubnau, E., and Sampson, N. S. (2011) J Bacteriol  73.  Vaillancourt, F. H., Bolin, J. T., and Eltis, L. D. (2006) Crit Rev Biochem Mol Biol 41, 241-267  74.  Gerlt, J. A., and Babbitt, P. C. (2001) Annu Rev Biochem 70, 209-246  75.  Armstrong, R. N. (2000) Biochemistry 39, 13625-13632  76.  Vaillancourt, F. H., Haro, M. A., Drouin, N. M., Karim, Z., Maaroufi, H., and Eltis, L. D. (2003) J Bacteriol 185, 1253-1260  77.  Anderton, M. C., Bhakta, S., Besra, G. S., Jeavons, P., Eltis, L. D., and Sim, E. (2006) Mol Microbiol 59, 181-192  78.  Uragami, Y., Senda, T., Sugimoto, K., Sato, N., Nagarajan, V., Masai, E., Fukuda, M., and Mitsui, Y. (2001) J Inorg Biochem 83, 269-279  79.  Han, S., Eltis, L. D., Timmis, K. N., Muchmore, S. W., and Bolin, J. T. (1995) Science 270, 976-980  80.  Senda, T., Sugiyama, K., Narita, H., Yamamoto, T., Kimbara, K., Fukuda, M., Sato, M., Yano, K., and Mitsui, Y. (1996) Journal of Molecular Biology 255, 735-752  81.  Bugg, T. D., and Ramaswamy, S. (2008) Curr Opin Chem Biol 12, 134-140  116  82.  Kovaleva, E. G., Neibergall, M. B., Chakrabarty, S., and Lipscomb, J. D. (2007) Acc Chem Res 40, 475-483  83.  Spence, E. L., Langley, G. J., and Bugg, T. D. H. (1996) Journal of the American Chemical Society 118, 8336-8343  84.  Winfield, C. J., Al-Mahrizy, Z., Gravestock, M., and Bugg, T. D. H. (2000) J Chem Soc Perk T 1, 3277-3289  85.  Sanvoisin, J., Langley, G. J., and Bugg, T. D. H. (1995) Journal of the American Chemical Society 117, 7836-7837  86.  Kovaleva, E. G., and Lipscomb, J. D. (2007) Science 316, 453-457  87.  Groce, S. L., Miller-Rodeberg, M. A., and Lipscomb, J. D. (2004) Biochemistry 43, 15141-15153  88.  Vaillancourt, F. H., Labbe, G., Drouin, N. M., Fortin, P. D., and Eltis, L. D. (2002) J Biol Chem 277, 2019-2027  89.  Dai, S., Vaillancourt, F. H., Maaroufi, H., Drouin, N. M., Neau, D. B., Snieckus, V., Bolin, J. T., and Eltis, L. D. (2002) Nat Struct Biol 9, 934-939  90.  Li, C., Hassler, M., and Bugg, T. D. (2008) Chembiochem 9, 71-76  91.  Catelani, D., Colombi, A., Sorlini, C., and Treccani, V. (1973) Biochemical Journal 134, 1063-1066  92.  Dunn, G., Montgomery, M. G., Mohammed, F., Coker, A., Cooper, J. B., Robertson, T., Garcia, J. L., Bugg, T. D., and Wood, S. P. (2005) J Mol Biol 346, 253-265  93.  Horsman, G. P., Ke, J., Dai, S., Seah, S. Y., Bolin, J. T., and Eltis, L. D. (2006) Biochemistry 45, 11071-11086  94.  Ollis, D. L., Cheah, E., Cygler, M., Dijkstra, B., Frolow, F., Franken, S. M., Harel, M., Remington, S. J., Silman, I., and Schrag, J. (1992) Protein Engineering 5, 197211  95.  Holmquist, M. (2000) Curr Protein Pept Sci 1, 209-235  117  96.  Nardini, M., Ridder, I. S., Rozeboom, H. J., Kalk, K. H., Rink, R., Janssen, D. B., and Dijkstra, B. W. (1999) J Biol Chem 274, 14579-14586  97.  Franken, S. M., Rozeboom, H. J., Kalk, K. H., and Dijkstra, B. W. (1991) EMBO J 10, 1297-1302  98.  Fulop, V., Bocskei, Z., and Polgar, L. (1998) Cell 94, 161-170  99.  Bugg, T. D. (2004) Bioorg Chem 32, 367-375  100.  Fleming, S. M., Robertson, T. A., Langley, G. J., and Bugg, T. D. (2000) Biochemistry 39, 1522-1531  101.  Li, J. J., Li, C., Blindauer, C. A., and Bugg, T. D. (2006) Biochemistry 45, 1246112469  102.  Horsman, G. P., Bhowmik, S., Seah, S. Y., Kumar, P., Bolin, J. T., and Eltis, L. D. (2007) J Biol Chem 282, 19894-19904  103.  Lam, W. W., and Bugg, T. D. (1997) Biochemistry 36, 12242-12251  104.  Williams, D. H., and Fleming, I. (1995) Spectroscopic methods in organic chemistry, 5th ed., McGraw-Hill, London  105.  Furukawa, K., Tomizuka, N., and Kamibayashi, A. (1979) Appl Environ Microbiol 38, 301-310  106.  Seeger, M., Timmis, K. N., and Hofer, B. (1995) Appl Environ Microbiol 61, 26542658  107.  Seah, S. Y., Labbe, G., Nerdinger, S., Johnson, M. R., Snieckus, V., and Eltis, L. D. (2000) J Biol Chem 275, 15701-15708  108.  Seah, S. Y., Labbe, G., Kaschabek, S. R., Reifenrath, F., Reineke, W., and Eltis, L. D. (2001) J Bacteriol 183, 1511-1516  109.  Bhowmik, S., Horsman, G. P., Bolin, J. T., and Eltis, L. D. (2007) J Biol Chem 282, 36377-36385  110.  Nerdinger, S., Kendall, C., Cai, X., Marchart, R., Riebel, P., Johnson, M. R., Yin, C. F., Henaff, N., Eltis, L. D., and Snieckus, V. (2007) J Org Chem 72, 5960-5967  118  111.  Seto, M., Kimbara, K., Shimura, M., Hatta, T., Fukuda, M., and Yano, K. (1995) Appl Environ Microbiol 61, 3353-3358  112.  Albarran, G., Boggess, W., Rassolov, V., and Schuler, R. H. (2010) J Phys Chem A 114, 7470-7478  113.  Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K. (2000) Current protocols in molecular biology  114.  Tabor, S., and Richardson, C. C. (1985) Proc Natl Acad Sci U S A 82, 1074-1078  115.  Dente, L., Cesareni, G., and Cortese, R. (1983) Nucleic Acids Res 11, 1645-1655  116.  Yam, K. C., D'Angelo, I., Kalscheuer, R., Zhu, H., Wang, J. X., Snieckus, V., Ly, L. H., Converse, P. J., Jacobs, W. R., Jr., Strynadka, N., and Eltis, L. D. (2009) PLoS Pathog 5, e1000344  117.  Bradford, M. M. (1976) Anal Biochem 72, 248-254  118.  Haigler, B. E., and Gibson, D. T. (1990) J Bacteriol 172, 457-464  119.  Cornish-Bowden, A. (1995) Analysis of enzyme kinetic data, Oxford University Press, New York  120.  Erman, J. E., and Vitello, L. B. (1980) J Biol Chem 255, 6224-6227  121.  Zhang, J. H., Chung, T. D., and Oldenburg, K. R. (1999) J Biomol Screen 4, 67-73  122.  Haberland, M. E., and Reynolds, J. A. (1973) Proceedings of the National Academy of Sciences of the United States of America 70, 2313-2316  123.  Cerdan, P., Rekik, M., and Harayama, S. (1995) Eur J Biochem 229, 113-118  124.  Vaillancourt, F. H., Han, S., Fortin, P. D., Bolin, J. T., and Eltis, L. D. (1998) J Biol Chem 273, 34887-34895  125.  Copeland, R. A. (2003) Anal Biochem 320, 1-12  126.  Hiraoka, Y., Yamada, T., Tone, K., Futaesaku, Y., and Kimbara, K. (2002) Appl Environ Microbiol 68, 5104-5112  119  127.  Monks, T. J., Hanzlik, R.P., Cohen, G.M., Ross, D., Graham, D.G. (1992) Toxicol. Appl. Pharmacol. 112, 14  128.  O'Brien, P. J. (1991) Chem Biol Interact 80, 1-41  129.  Seah, S. Y., Ke, J., Denis, G., Horsman, G. P., Fortin, P. D., Whiting, C. J., and Eltis, L. D. (2007) J Bacteriol 189, 4038-4045  130.  Capyk, J. K., D'Angelo, I., Strynadka, N. C., and Eltis, L. D. (2009) J Biol Chem 284, 9937-9946  131.  Gomez-Gil, L., Kumar, P., Barriault, D., Bolin, J. T., Sylvestre, M., and Eltis, L. D. (2007) J Bacteriol 189, 5705-5715  132.  Ballou, D. P., Entsch, B., and Cole, L. J. (2005) Biochemical and biophysical research communications 338, 590-598  133.  Cerdan, P., Wasserfallen, A., Rekik, M., Timmis, K. N., and Harayama, S. (1994) J Bacteriol 176, 6074-6081  134.  Sih, C. J., and Wang, K. C. (1965) J Am Chem Soc 87, 1387-1388  135.  Whitmarsh, J. M. (1964) Biochem. J. 90, 23p-24p  136.  Rosloniec, K. Z., Wilbrink, M. H., Capyk, J. K., Mohn, W. W., Ostendorf, M., van der Geize, R., Dijkhuizen, L., and Eltis, L. D. (2009) Mol Microbiol 74, 1031-1043  137.  Yam, K., Van der Geize, R., and Eltis, L. D. (2010) Catabolism of Aromatic Compounds and Steroids by Rhodococcus. in Biology of Rhodococcus (Alvarez, H. M. ed.), 1 Ed., Springer-Verlag Berlin Heidelberg, Heidelberg, Germany. pp 133-170  138.  Olivera, E. R., Minambres, B., Garcia, B., Muniz, C., Moreno, M. A., Ferrandez, A., Diaz, E., Garcia, J. L., and Luengo, J. M. (1998) Proc Natl Acad Sci U S A 95, 64196424  139.  Navarro-Llorens, J. M., Patrauchan, M. A., Stewart, G. R., Davies, J. E., Eltis, L. D., and Mohn, W. W. (2005) J Bacteriol 187, 4497-4504  140.  Vetting, M. W., Wackett, L. P., Que, L., Jr., Lipscomb, J. D., and Ohlendorf, D. H. (2004) J Bacteriol 186, 1945-1958  120  141.  Lack, N. A., Yam, K. C., Lowe, E. D., Horsman, G. P., Owen, R. L., Sim, E., and Eltis, L. D. (2010) J Biol Chem 285, 434-443  142.  Lack, N., Lowe, E. D., Liu, J., Eltis, L. D., Noble, M. E., Sim, E., and Westwood, I. M. (2008) Acta Crystallogr Sect F Struct Biol Cryst Commun 64, 2-7  143.  Zhang, Y., Colabroy, K. L., Begley, T. P., and Ealick, S. E. (2005) Biochemistry 44, 7632-7643  144.  Lee, S. E., Kim, M. R., Kim, J. H., Takeoka, G. R., Kim, T. W., and Park, B. S. (2008) Phytomedicine 15, 533-535  145.  Sundarasivarao, B., Nazma, and Madhusudhanarao, J. (1977) Curr Sci India 46, 714716  146.  Brandt, G. E., Schmidt, M. D., Prisinzano, T. E., and Blagg, B. S. (2008) J Med Chem 51, 6495-6502  147.  Uddin, S. J., Nahar, L., Shilpi, J. A., Shoeb, M., Borkowski, T., Gibbons, S., Middleton, M., Byres, M., and Sarker, S. D. (2007) Phytother Res 21, 757-761  148.  Wang, H., Ruan, Z., Li, J. J., Simpkins, L. M., Smirk, R. A., Wu, S. C., Hutchins, R. D., Nirschl, D. S., Van Kirk, K., Cooper, C. B., Sutton, J. C., Ma, Z., Golla, R., Seethala, R., Salyan, M. E., Nayeem, A., Krystek, S. R., Jr., Sheriff, S., Camac, D. M., Morin, P. E., Carpenter, B., Robl, J. A., Zahler, R., Gordon, D. A., and Hamann, L. G. (2008) Bioorg Med Chem Lett 18, 3168-3172  149.  Trott, O., and Olson, A. J. (2010) J Comput Chem 31, 455-461  150.  Saludes, J. P., Garson, M. J., Franzblau, S. G., and Aguinaldo, A. M. (2002) Phytother Res 16, 683-685  151.  Woldemichael, G. M., Franzblau, S. G., Zhang, F., Wang, Y., and Timmermann, B. N. (2003) Planta Med 69, 628-631  152.  Thomas, S. T., Yang, X., and Sampson, N. S. (2011) Bioorganic & Medicinal Chemistry Letters 21, 2216-2219  153.  Westwood, I. M., Bhakta, S., Russell, A. J., Fullam, E., Anderton, M. C., Kawamura, A., Mulvaney, A. W., Vickers, R. J., Bhowruth, V., Besra, G. S., Lalvani, A., Davies, S. G., and Sim, E. (2010) Protein Cell 1, 82-95  121  154.  Nandhagopal, N., Yamada, A., Hatta, T., Masai, E., Fukuda, M., Mitsui, Y., and Senda, T. (2001) J Mol Biol 309, 1139-1151  155.  Hu, J. G., Sun, B. Q., Petkova, A. T., Griffin, R. G., and Herzfeld, J. (1997) Biochemistry 36, 9316-9322  156.  Lanyi, J. K., and Schobert, B. (2007) J Mol Biol 365, 1379-1392  157.  Chang, J. C., Miner, M. D., Pandey, A. K., Gill, W. P., Harik, N. S., Sassetti, C. M., and Sherman, D. R. (2009) Journal of bacteriology 191, 5232-5239  158.  Chang, J. C., Harik, N. S., Liao, R. P., and Sherman, D. R. (2007) J Infect Dis 196, 788-795  159.  de Carvalho, L. P., Fischer, S. M., Marrero, J., Nathan, C., Ehrt, S., and Rhee, K. Y. (2010) Chem Biol 17, 1122-1131  160.  Bentrup, K. H. Z., Miczak, A., Swenson, D. L., and Russell, D. G. (1999) Journal of Bacteriology 181, 7161-7167  161.  McKinney, J. D., Honer zu Bentrup, K., Munoz-Elias, E. J., Miczak, A., Chen, B., Chan, W. T., Swenson, D., Sacchettini, J. C., Jacobs, W. R., Jr., and Russell, D. G. (2000) Nature 406, 735-738  162.  Constant, S. L., and Bottomly, K. (1997) Annu Rev Immunol 15, 297-322  163.  Gamieldien, J., Ptitsyn, A., and Hide, W. (2002) Trends Genet 18, 5-8  164.  Tailleux, L., Waddell, S. J., Pelizzola, M., Mortellaro, A., Withers, M., Tanne, A., Castagnoli, P. R., Gicquel, B., Stoker, N. G., Butcher, P. D., Foti, M., and Neyrolles, O. (2008) PLoS One 3, e1403  165.  Rook, G. A., Steele, J., Fraher, L., Barker, S., Karmali, R., O'Riordan, J., and Stanford, J. (1986) Immunology 57, 159-163  166.  Liu, P. T., Stenger, S., Li, H., Wenzel, L., Tan, B. H., Krutzik, S. R., Ochoa, M. T., Schauber, J., Wu, K., Meinken, C., Kamen, D. L., Wagner, M., Bals, R., Steinmeyer, A., Zugel, U., Gallo, R. L., Eisenberg, D., Hewison, M., Hollis, B. W., Adams, J. S., Bloom, B. R., and Modlin, R. L. (2006) Science 311, 1770-1773  122  167.  Liu, P. T., Stenger, S., Tang, D. H., and Modlin, R. L. (2007) J Immunol 179, 20602063  168.  Selvaraj, P., Prabhu Anand, S., Harishankar, M., and Alagarasu, K. (2009) J Clin Immunol 29, 470-478  169.  Mathieu, C., and Adorini, L. (2002) Trends Mol Med 8, 174-179  170.  Jang, S., Uematsu, S., Akira, S., and Salgame, P. (2004) Journal of immunology 173, 3392-3397  171.  Daley, C. L. (2002) Am J Respir Crit Care Med 165, 742-743  123  APPENDIX: CRYSTALLOGRAPHIC DATA Table 9. Crystallographic properties, X-ray diffraction data, and refinement statistics for HsaC. Diffraction Data  HsaC (substrate free)  X-ray source  Cu-Kα  ALS 8.2.2  Wavelength (Å)  1.542  1.000  Space group  P4212  P4212  Unit cell (Å)  a=b=123.7, c=106.7  a=b=124.3, c=106.3  Resolution (Å)  2.0  2.2  Highest shell (Å)  2.0-2.2  2.32-2.20  Total observations  791,882 (109,166)  384,012 (55,996)  Unique reflections  56,454 (8,105)  42,904 (6,156)  I / σI  28.9 (7.1)  26.0 (6.1)  9.1 (44.1)  8.1 (38.9)  99.9 (99.9)  100 (100)  Resolution range (Å)  20-2.0  20-2.2  No. Reflections  53,536  40,622  22/18  26/19  Total  5,364  5,224  Protein  4,677  4,692  Solvent  712  485  Protein  19.4  25.1  Fe  14.8  20.3  DHSA  -  45.5  Bond lengths (Å)  0.015  0.024  Bond angles (deg)  1.67  2.42  Rsym(%)  ‡  Completeness (%)  HsaC:DHSA  Refined Model  Rfree/Rfactor (%)  †  No. atoms:  2  Mean B values (Å ):  R.m.s. deviations:  ‡  Rsym = ΣhΣi I(hkl) − I(hkl) /ΣhΣiI(hkl). Rwork = Σ||Fobs | − | Fcalc||/Σ | Fobs |. Rfree is the Rwork value for 5% of the reflections excluded from the refinement. Data for the highest resolution shell are given in parentheses. †  124  Taken from Yam, K.C., D’Angelo, I., Kalscheuer, R., Zhu, H., Wang, J.X., Snieckus, V., Ly, L., Converse, P.J., Jacobs, W.R., Jr., Strynadka, N. and Eltis L.D. Studies of a ring-cleaving dioxygenase illuminate the role of cholesterol metabolism in the pathogenesis of Mycobacterium tuberculosis (2009) Plos Pathogens e1000344. (116).  125  Table 10. Summary of data-collection and refinement statistics for S114A and S114A complexes of HsaD. Reprinted with permission from Lack, N.A., Yam, K.C., Lowe, E.D., Horsman, G.P., Owen, R.L., Sim, E., and Eltis L.D. Characterization of a carbon-carbon bond hydrolase from Mycobacterium tuberculosis involved in cholesterol metabolism (2010) Journal of Biological Chemistry 285, 434-443.(141)  126  Table 11. Torsional angles of the refined 2-enol and 2-oxo, 6-oxo tautomers of HOPDA, HOPODA and DSHA in the S114A variant of HsaD. 2-Enol  2-Oxo, 6-oxo  C2-C3  C3-C4  C4-C5  C2-C3  C3-C4  C4-C5  Bonding  Double  Single  Double  Single  Double  Single  HOPDAchainB  174  175  113  177  177  116  HOPDAchainA  164  168  129  149  173  139  HOPODA  173  165  103  176  165  98  DSHA  159  167  113  158  175  97  Reprinted with permission from Lack, N.A., Yam, K.C., Lowe, E.D., Horsman, G.P., Owen, R.L., Sim, E., and Eltis L.D. Characterization of a carbon-carbon bond hydrolase from Mycobacterium tuberculosis involved in cholesterol metabolism (2010) Journal of Biological Chemistry 285, 434-443.(141)  127  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0072179/manifest

Comment

Related Items