Discovery of benzothiazoles as antimycobacterial agents: Synthesis, structure-activity relationships and binding studies with Mycobacterium tuberculosis decaprenylphosphoryl-ß-D-ribose 2'-oxidase.

We report the discovery of benzothiazoles, a novel anti-mycobacterial series, identified from a whole cell based screening campaign. Benzothiazoles exert their bactericidal activity against Mycobacterium tuberculosis (Mtb) through potent inhibition of decaprenylphosphoryl-ß-d-ribose 2'-oxidase (DprE1), the key enzyme involved in arabinogalactan synthesis. Specific target linkage and mode of binding were established using co-crystallization and protein mass spectrometry studies. Most importantly, the current study provides insights on the utilization of systematic medicinal chemistry approaches to mitigate safety liabilities while improving potency during progression from an initial genotoxic hit, the benzothiazole N-oxides (BTOs) to the lead-like AMES negative, crowded benzothiazoles (cBTs). These findings offer opportunities for development of safe clinical candidates against tuberculosis. The design strategy adopted could find potential application in discovery of safe drugs in other therapy areas too.

Optimization of pyrrolamides as mycobacterial GyrB ATPase inhibitors: structure-activity relationship and in vivo efficacy in a mouse model of tuberculosis.

Moxifloxacin has shown excellent activity against drug-sensitive as well as drug-resistant tuberculosis (TB), thus confirming DNA gyrase as a clinically validated target for discovering novel anti-TB agents. We have identified novel inhibitors in the pyrrolamide class which kill Mycobacterium tuberculosis through inhibition of ATPase activity catalyzed by the GyrB domain of DNA gyrase. A homology model of the M. tuberculosis H37Rv GyrB domain was used for deciphering the structure-activity relationship and binding interactions of inhibitors with mycobacterial GyrB enzyme. Proposed binding interactions were later confirmed through cocrystal structure studies with the Mycobacterium smegmatis GyrB ATPase domain. The most potent compound in this series inhibited supercoiling activity of DNA gyrase with a 50% inhibitory concentration (IC50) of <5 nM, an MIC of 0.03 µg/ml against M. tuberculosis H37Rv, and an MIC90 of <0.25 µg/ml against 99 drug-resistant clinical isolates of M. tuberculosis. The frequency of isolating spontaneous resistant mutants was ∼10(-6) to 10(-8), and the point mutation mapped to the M. tuberculosis GyrB domain (Ser208 Ala), thus confirming its mode of action. The best compound tested for in vivo efficacy in the mouse model showed a 1.1-log reduction in lung CFU in the acute model and a 0.7-log reduction in the chronic model. This class of GyrB inhibitors could be developed as novel anti-TB agents.

Alanine racemase mutants of Mycobacterium tuberculosis require D-alanine for growth and are defective for survival in macrophages and mice.

Alanine racemase (Alr) is an essential enzyme in most bacteria; however, some species (e.g. Listeria monocytogenes) can utilize d-amino acid transaminase (Dat) to generate d-alanine, which renders Alr non-essential. In addition to the conflicting reports on gene knockout of alr in Mycobacterium smegmatis, a recent study concluded that depletion of Alr does not affect the growth of M. smegmatis. In order to get an unambiguous answer on the essentiality of Alr in Mycobacterium tuberculosis and validate it as a drug target in vitro and in vivo, we have inactivated the alr gene of M. tuberculosis and found that it was not possible to generate an alr knockout in the absence of a complementing gene copy or d-alanine in the growth medium. The growth kinetics of the alr mutant revealed that M. tuberculosis requires very low amounts of d-alanine (5-10 µg ml(-1)) for optimum growth. Survival kinetics of the mutant in the absence of d-alanine indicated that depletion of this amino acid results in rapid loss of viability. The alr mutant was found to be defective for growth in macrophages. Analysis of phenotype in mice suggested that non-availability of d-alanine in mice leads to clearance of bacteria followed by stabilization of bacterial number in lungs and spleen. Additionally, reversal of d-cycloserine inhibition in the presence of d-alanine in M. tuberculosis suggested that Alr is the primary target of d-cycloserine. Thus, Alr of M. tuberculosis is a valid drug target and inhibition of Alr alone should result in loss of viability in vitro and in vivo.

Essentiality and functional analysis of type I and type III pantothenate kinases of Mycobacterium tuberculosis.

Pantothenate kinase, an essential enzyme in bacteria and eukaryotes, is involved in catalysing the first step of conversion of pantothenate to coenzyme A (CoA). Three isoforms (type I, II and III) of this enzyme have been reported from various organisms, which can be differentiated from each other on the basis of their biochemical and structural characteristics. Though most bacteria carry only one of the isoforms of pantothenate kinases, some of them possess two isoforms. The physiological relevance of the presence of two types of isozymes in a single organism is not clear. Mycobacterium tuberculosis, an intracellular pathogen, possesses two isoforms of pantothenate kinases (CoaA and CoaX) belonging to type I and III. In order to determine which pantothenate kinase is essential in mycobacteria, we performed gene inactivation of coaA and coaX of M. tuberculosis individually. It was found that coaA could only be inactivated in the presence of an extra copy of the gene, while coaX could be inactivated in the wild-type cells, proving that CoaA is the essential pantothenate kinase in M. tuberculosis. Additionally, the coaA gene of M. tuberculosis was able to complement a temperature-sensitive coaA mutant of Escherichia coli at a non-permissive temperature while coaX could not. The coaX deletion mutant showed no growth defects in vitro, in macrophages or in mice. Taken together, our data suggest that CoaX, which is essential in Bacillus anthracis and thus had been suggested to be a drug target in this organism, might not be a valid target in M. tuberculosis. We have established that the type I isoform, CoaA, is the essential pantothenate kinase in M. tuberculosis and thus can be explored as a drug target.

Inactivation of the ilvB1 gene in Mycobacterium tuberculosis leads to branched-chain amino acid auxotrophy and attenuation of virulence in mice.

Acetohydroxyacid synthase (AHAS) is the first enzyme in the branched-chain amino acid biosynthesis pathway in bacteria. Bioinformatics analysis revealed that the Mycobacterium tuberculosis genome contains four genes (ilvB1, ilvB2, ilvG and ilvX) coding for the large catalytic subunit of AHAS, whereas only one gene (ilvN or ilvH) coding for the smaller regulatory subunit of this enzyme was found. In order to understand the physiological role of AHAS in survival of the organism in vitro and in vivo, we inactivated the ilvB1 gene of M. tuberculosis. The mutant strain was found to be auxotrophic for all of the three branched-chain amino acids (isoleucine, leucine and valine), when grown with either C(6) or C(2) carbon sources, suggesting that the ilvB1 gene product is the major AHAS in M. tuberculosis. Depletion of these branched chain amino acids in the medium led to loss of viability of the DeltailvB1 strain in vitro, resulting in a 4-log reduction in colony-forming units after 10 days. Survival kinetics of the mutant strain cultured in macrophages maintained with sub-optimal concentrations of the branched-chain amino acids did not show any loss of viability, indicating either that the intracellular environment was rich in these amino acids or that the other AHAS catalytic subunits were functional under these conditions. Furthermore, the growth kinetics of the DeltailvB1 strain in mice indicated that although this mutant strain showed defective growth in vivo, it could persist in the infected mice for a long time, and therefore could be a potential vaccine candidate.

PKCalpha regulates phosphorylation and enzymatic activity of cPLA2 in vitro and in activated human monocytes.

Phospholipases A(2) (PLA(2)) are potent regulators of the inflammatory response. We have observed that Group IV cPLA(2) activity is required for the production of superoxide anion (O(2)(-)) in human monocytes [Li Q., Cathcart M.K. J. Biol. Chem. 272 (4) (1997) 2404-2411.]. We have previously identified PKCalpha as a kinase pathway required for monocyte O(2)(-) production [Li Q., Cathcart M.K. J. Biol. Chem. 269 (26) (1994) 17508-17515.]. We therefore investigated the potential interaction between PKCalpha and cPLA(2) by evaluating the requirement for specific PKC isoenzymes in the process of activating cPLA(2) enzymatic activity and protein phosphorylation upon monocyte activation. We first showed that general PKC inhibitors and antisense oligodeoxyribonucleotides (ODN) to the cPKC group of PKC enzymes inhibited cPLA(2) activity. To distinguish between PKCalpha and PKCbeta isoenzymes in regulating cPLA(2) protein phosphorylation and enzymatic activity, we employed our previously characterized PKCalpha or PKCbeta isoenzyme-specific antisense ODN [Li Q., Subbulakshmi V., Fields A.P., Murray, N.R., Cathcart M.K., J. Biol. Chem. 274 (6) (1999) 3764-3771]. Suppression of PKCalpha expression, but not PKCbeta expression, inhibited cPLA(2) protein phosphorylation and enzymatic activity. Additional studies ruled out a contribution by Erk1/2 to cPLA(2) phosphorylation and activation. We also found that cPLA(2) co-immunoprecipitated with PKCalpha and vice versa. In vitro studies demonstrated that PKCalpha could directly phosphorylate cPLA(2).and enhance enzymatic activity. Finally, we showed that addition of arachidonic acid restored the production of O(2)(-) in monocytes defective in either PKCalpha or cPLA(2) expression. Taken together, our data suggest that PKCalpha, but not PKCbeta, is the predominant cPKC isoenzyme required for cPLA(2) protein phosphorylation and maximal induction of cPLA(2) enzymatic activity upon activation of human monocytes. Our data also support the concept that the requirements for PKCalpha and cPLA(2) in O(2)(-) generation are solely due to their seminal role in generating arachidonic acid.

Protein kinase C delta is required for p47phox phosphorylation and translocation in activated human monocytes.

Our laboratory is interested in understanding the regulation of NADPH oxidase activity in human monocyte/macrophages. Protein kinase C (PKC) is reported to be involved in regulating the phosphorylation of NADPH oxidase components in human neutrophils; however, the regulatory roles of specific isoforms of PKC in phosphorylating particular oxidase components have not been determined. In this study calphostin C, an inhibitor for both novel PKC (including PKCdelta, -epsilon, -theta;, and -eta) and conventional PKC (including PKCalpha and -beta), inhibited both phosphorylation and translocation of p47phox, an essential component of the monocyte NADPH oxidase. In contrast, GF109203X, a selective inhibitor of classical PKC and PKCepsilon, did not affect the phosphorylation or translocation of p47phox, suggesting that PKCdelta, -theta;, or -eta is required. Furthermore, rottlerin (at doses that inhibit PKCdelta activity) inhibited the phosphorylation and translocation of p47phox. Rottlerin also inhibited O2 production at similar doses. In addition to pharmacological inhibitors, PKCdelta-specific antisense oligodeoxyribonucleotides were used. PKCdelta antisense oligodeoxyribonucleotides inhibited the phosphorylation and translocation of p47phox in activated human monocytes. We also show, using the recombinant p47phox-GST fusion protein, that p47phox can serve as a substrate for PKCdelta in vitro. Furthermore, lysate-derived PKCdelta from activated monocytes phosphorylated p47phox in a rottlerin-sensitive manner. Together, these data suggest that PKCdelta plays a pivotal role in stimulating monocyte NADPH oxidase activity through its regulation of the phosphorylation and translocation of p47phox.