The chemical biology of new drugs in the development for tuberculosis

Download The chemical biology of new drugs in the development for tuberculosis

Post on 26-Jun-2016

215 views

Category:

Documents

2 download

Embed Size (px)

TRANSCRIPT

<ul><li><p>Available online at www.sciencedirect.com</p><p>g</p><p>nrifampicin alone [3 ,4]. Isoniazid and rifampicin arerapidly bactericidal agents, while pyrazinamide isincluded to treat a presumed population of nonreplicatingorganisms [5].</p><p>In the early 1980s, significant clinical resistance to iso-niazid began to be noted and 15 years later, combinedresistance to isoniazid and rifampicin similarly began toincrease [6,7]. MDR is defined as resistance to these lattertwo rapidly bactericidal agents [8,9]. Strains that have</p><p>assessed for enhanced activity. Unfortunately, none ofthis chemistry has appeared outside of the patent litera-ture where it is sketchy and incomplete limiting theamount of available structureactivity information (USPatent 2007/0249667). A smaller series of similarly sub-stituted quinolines have subsequently been reported byanother group but these have much more limited potency[19]. R207910 was selected as the lead compound inthe series after a mouse infection study showed it wasthe only compound in a series of three tested to have</p><p>Current Opinion in Chemical Biology 2010, 14:456466 www.sciencedirect.comThe chemical biology of new drutuberculosisClifton E Barry 3rd1,2 and John S Bla</p><p>With the worldwide emergence of multidrug-resistant (MDR)</p><p>and extensively drug-resistant (XDR) strains of Mycobacterium</p><p>tuberculosis (Mtb), there are serious concerns about the</p><p>continued ability to contain this disease. We discuss the most</p><p>promising new drugs in late-stage development that might be</p><p>useful in treating MDR and XDR forms of the disease. These</p><p>agents have novel mechanisms of action that are not targeted</p><p>by the standard drugs used presently to treat susceptible</p><p>strains.</p><p>Addresses1 Tuberculosis Research Section, Laboratory of Clinical Infectious</p><p>Disease, National Institute of Allergy and Infectious Diseases, National</p><p>Institute of Health, Bethesda, MD 20892, United States2 Department of Biochemistry, Albert Einstein College of Medicine, 1300</p><p>Morris Park Avenue, Bronx, NY 10461, United States</p><p>Corresponding authors: Blanchard, John S (blanchar@aecom.yu.edu)</p><p>Current Opinion in Chemical Biology 2010, 14:456466</p><p>This review comes from a themed issue on</p><p>Next Generation Therapeutics</p><p>Edited by Michael Gelb and Adrian Whitty</p><p>Available online 7th May 2010</p><p>1367-5931/$ see front matter</p><p># 2010 Elsevier Ltd. All rights reserved.</p><p>DOI 10.1016/j.cbpa.2010.04.008</p><p>IntroductionAlthough effective chemotherapy for tuberculosis hasbeen in place for over 50 years, it was clear from the firstclinical trials that monotherapy with any agent led to thedevelopment of resistance and clinical failure in two tofive months [1]. Therefore, tuberculosis chemotherapyhas involved the administration of multiple drugs sincethe late 1960s [2]. At present, the most widely used short-course therapy involves taking isoniazid, pyrazinamide,and rifampicin (Figure 1) for two months, followed by anadditional four months of treatment with isoniazid and</p><p>s in the development for</p><p>chard1,2</p><p>acquired additional resistance to one of the three mostpotent injectables (second-generation aminoglycosidessuch as kanamycin and amikacin, and cyclic polypeptidessuch as capreomycin) and a fluoroquinolone are deemedextensively drug-resistant (XDR) [10,11]. This combi-nation of resistances is associated with particularly pooroutcomes in patients because of the relative lack ofpotency of the remaining second-line agents [12,13]. Ina report published in 2006, an outbreak of XDR-TB wasdescribed in KwaZulu Natal, South Africa where thestrain was responsible for 100% mortality of 54 infectedindividuals with an average time from diagnosis to deathof 16 days [14].</p><p>In the last 10 years, there has been resurgence in interestin identifying new compounds that are effective againstMtb. Much of this interest has been within academic andgovernmental laboratories, as opposed to the biotechnol-ogy or large pharmaceutical industries. Because of spacelimitations, we will be unable to discuss many promisingnew compounds in preclinical development (many ofwhich have been reviewed recently [1517]), and haveelected instead to discuss those compounds for whichclinical trials (Phases 1 and 2) have or will begin shortly insome detail. Several of these compounds are similar tothose used against other bacterial infections; however,some appear to be quite specific for Mtb.</p><p>Diarylquinolines (R207910)In early 2005, a report from a Johnson and Johnson groupin Europe appeared describing the activity of a new classof diarylquinolines against Mtb [18]. Using a whole-cellassay against a surrogate organism (a fast-growing sapro-phytic cousin of Mtb called Mycobacterium smegmatis) a hitwas identified that was a structurally unique, highlysubstituted quinoline that also showed activity againstMtb. Diarylquinolines are distinct both in structure andmechanism from fluoroquinolones (which inhibit type IItopoisomerases such as DNA gyrase) and quinolines suchas mefloquine. A series of analogs were prepared and</p></li><li><p>New Drugs in Development for Tuberculosis Barry and Blanchard 457Figure 1significant activity. As shown in Figure 2, R207910 is asingle enantiomer of a compound with two chiral centerswith the carbon bearing the phenyl substituent of the Rconfiguration and with the carbon bearing the hydroxylgroup of the S configuration.</p><p>R207910 has extraordinary activity against both drug-susceptible and drug-resistant strains of M. tuberculosis,exhibiting MIC (Minimum Inhibitory Concentration)values of 30120 ng/ml, equal to or lower than isoniazidand rifampicin. The compound exhibits a rapid bacteri-cidal activity in vitro, causing losses of 3 log orders of CFU(Colony Forming Units)/ml in 12 days. Surprisingly, thecompound exhibits a very narrow spectrum of activity andloses activity against even closely related actinomycetessuch as Corynebacteria, and it is also inactive againstother Gram-positive and Gram-negative pathogens.Among the mycobacteria, R207910 shows consistentactivity against many species, including many medicallyimportant opportunistic pathogens [20]. Recently it haseven been shown to have activity against Mycobacteriumleprae, the causative agent of leprosy, in a mouse model ofthe disease [21].</p><p>The target of R207910was identified by selecting resistantmutants in both Mtb and M. smegmatis and subsequently</p><p>Drugs currently used in modern short-course chemotherapy of</p><p>Tuberculosis.</p><p>www.sciencedirect.comapplying whole genome resequencing to identify candi-date polymorphisms associated with resistance. Mutationscommon to the organisms that had acquired resistancewere found in the atpE gene, encoding the membrane-bound subunit of the F0 ATP synthase complex. In anMtbmutant, a single point mutation that generated an A63Psubstitution was observed, while in one M. smegmatismutant, a point mutation generated a D32V substitution.Subsequent investigations of large numbers of in vitrogenerated mutants have confirmed a role for AtpE inresistance to R207910, and the few atypical mycobacteriathat show intrinsic resistance to R207910 have naturallyoccurring polymorphisms at A63 [20,22]. Similarly, theeukaryotic mitochondrial ATP synthase has a methionineat position 63, a fact that may explain the remarkableselectivity (&gt;20 000-fold) of R207910 forMtb versus mam-mals [23]. On the basis of a homologymodel built from theNMR-derived structure of the c-subunit from E. coli, bothA63 andD32 arewithin themembrane-spanningportion ofthe antiparallel a-helices (Figure 2). Computational dock-ing studies based on this model structure suggest thatimportant contacts are possible with R186, E61, and F65but do not convincingly explain the observed mutations atA63 andD32, nor do they produce a convincing ranking ofthe stereoisomeric versions of this compound [24]. Theintrinsically low resolution ofNMRstructures amplifiedbyuncertainties in deriving homology models from suchstructures limit the confidence in the predicted bindingsite and contacts of R207910 with this protein.</p><p>A recent study looking at the development of resistanceto R207910 in nearly 100 spontaneously resistant mutantsobtained from seven different clinical isolates foundmutation of atpE only partially accounted for the resist-ance observed. In 38 of these isolates no mutations werefound anywhere in the ATP synthase operon, suggestingthe possibility that R207910 has additional, as yetunknown, targets [22]. However, inhibition of ATP syn-thesis clearly plays a major role in the mechanism ofR207910. Treatment of whole cells with the drug reducesATP concentrations and the drug inhibits ATP synthesiseven in isolated vesicles. Finally, coupling of a derivativeof the drug (Figure 2) to an affinity column effectivelybound the a and b subunits of ATP synthase fromsolubilizedmembrane proteins ofM. smegmatis (but unfor-tunately not the c subunit which is highly hydrophobic)[25]. These results are not overly surprising given thenuances of drug-mediated bacterial cell death that arebeing revealed for even simple antibiotics like amino-glycoside inhibitors of protein synthesis [26,27].</p><p>One of the major challenges for tuberculosis drug de-velopment is the ability of a subpopulation of the organ-ism to exist in a nonreplicating, but viable form. Several</p><p>models of this physiological state have been developed,but the most commonly used is the Wayne model [28].Although assumed to be an obligate aerobe, when</p><p>Current Opinion in Chemical Biology 2010, 14:456466</p></li><li><p>458 Next Generation TherapeuticsFigure 2deprived of oxygen, M. tuberculosis can enter into a meta-bolic state termed persistence. In this state, the organ-isms are insensitive to, or substantially less sensitive tocommon drugs used to treat the disease, including iso-niazid and rifampicin. In this state, intracellular ATPconcentrations decrease to approximately 20% of thosein actively growing cells. Administration of R207910 tocultures of anaerobic cells causes a dose-dependentreduction of ATP levels and a corresponding decreasein CFUs recoverable after readmission of oxygen [29].These results have led to the conclusion that ATPhomeostasis is critical in nonreplicating cells and validate</p><p>Homology model of the Mtb c subunit of ATP synthase. Position 28 (left, sphe</p><p>63 (right sphere) is the Ala residue mutated to a Pro in the Mtb strain resistan</p><p>Mark Girvin, Department of Biochemistry, Albert Einstein College of Medicin</p><p>Current Opinion in Chemical Biology 2010, 14:456466the ATP synthase as a valuable target in this importantsubpopulation [30].</p><p>R207910 has performed exceptionally well in variousmouse models of TB [3135] and initial Phase 1 safetyand Phase 2 efficacy studies continue to look very prom-ising [18,36,37]. The most compelling results so far arefrom a Phase 2, randomized, controlled clinical trail inSouth Africa of patients diagnosed withMDR-TB. In thisstudy, patients were treated with either R207910 or aplacebo in combination with five other second-line anti-tuberculosis agents. In the placebo control group fewer</p><p>res) is the aspartate residue equivalent to D32 in M. smegmatis. Position</p><p>t to TMC207. This figure was generously provided by Josh Goldman and</p><p>e.</p><p>www.sciencedirect.com</p></li><li><p>to unravel some of the biochemistry surroundinganaerobic metabolism. Because the primary mechanismof resistance to this agent involved the loss of reductioncapacity in resistant organisms, the major cofactorinvolved in this reductive process, the deazaflavin F420was quickly identified, as was F420-dependent mycobac-terial glucose-6-phosphate dehydrogenase (G6PDH), itsmajor redox partner. Careful examination of PA-824-resistant mutants also allowed the identification of themajor biosynthetic genes involved in F420 biosynthesis[46,47]. What became clear ultimately was that F420 wasessential for reduction, and Fdg (the G6PDH) was essen-tial for reducing F420 at the expense of G6P, but thatthere had to be amissing enzyme that catalyzed the actualreduction.</p><p>Because the major class of mutants that were obtainedalways mapped to the biosynthetic genes for F420 or theFdg protein whether selected in vivo (from treated mice)or in vitro searching for the missing enzyme requiredscreening through many mutants and quickly ruling out</p><p>New Drugs in Development for Tuberculosis Barry and Blanchard 459than 10% of patients had converted their sputum fromculture-positive to culture-negative after two months,while 50% of the patient group treated with R207910converted in the same period.</p><p>R207910 represents one of the most promising of the newdrug candidates for the treatment of tuberculosis butthere remains a critical issue. The drug is rapidly metab-olized by cytochrome P-450 isoform 3A4, an isoform thatis strongly upregulated with the use of rifampicin, per-haps the single most important agent used for first linechemotherapy [33]. In healthy volunteers receiving bothdrugs the level of R207910 was reduced by 50%. Althoughadditional drugdrug interaction studies are certainlynecessary it remains to be seen whether R207910 willbe able to be used as a component of first line therapy fordrug-susceptible disease. In retrospect it is surprising thata compound with such a liability was selected as acandidate since screening for metabolic instability typi-cally occurs very early in drug development. Unfortu-nately, the complexity of an integral membrane targetwith little structural information, added to the chemicalcomplexity of the scaffold that requires chiral HPLCresolution at the last step, makes it very difficult toaddress this liability from an informed medicinal chem-istry perspective.</p><p>Nitroimidazoles (PA-824 and OPC67683)Metronidazole is perhaps the most well-known antibioticof the nitroimidazole class. It is effective againstanaerobic bacteria, protozoa and hypoxically adapted,nonreplicatingMtb [38]. In 1989, a group fromCiba-GeigyIndia reported the powerful antitubercular activity of aseries of bicyclic nitroimidazo[2,1-b]oxazoles. The mostpotent of these, CGI 17341, inhibited both drug-suscept-ible andmultidrug-resistant strains ofMtb and was specifi-cally active against Mtb with significantly less activeagainst M. avium, M. intracellulare, and M. fortuitum[39]. In preliminary tests in a mouse model of infection,the compound exhibited a dose-dependent increase inmean survival time. However, further development of thecompound was halted, presumably because of concernsrelating to mutagenicity.</p><p>Ten years later, a related compound, PA-824, featuring asix-membered oxazine rather than the five-memberedoxazole, was reported [40]. From more than 300 six-substituted 6,7-dihydro-5H-imidazo[2,1-b][1,3]oxazines,PA-824 was selected as a lead molecule (Figure 3). Muchof the initial excitement around the series was because ofthe fact that this compound, like its distant progenitormetronidazole, had activity against hypoxic bacilli, anobservation interpreted as indicating potential for short-ening of the overall duration of treatment. PA-824,</p><p>though was selected based on one characteristic activity in the mouse model o...</p></li></ul>