WO2013006854A2 - Compositions and methods for treating enterobacteriaceae - Google Patents

Abstract

Methods and compounds for treating bacterial infections are provided. Embodiments of the methods and compounds activate CpxRA to treat bacterial infection.

Description

COMPOSITIONS AND METHODS FOR TREATING ENTEROBACTERIACEAE

Cross Reference to Related Applications

This application claims priority to U.S. Provisional Application No. 61/505,307, filed July 7, 2011, the entire disclosure of which is hereby incorporated by reference.

Background

The increasing prevalence of multi-drug (MDR) resistant Gram-negative bacteria is a major public health concern. For instance, the KPC carbapenemase has rendered some strains of Klebsiella pneumoniae resistant to all β-lactams, while the New Delhi metallo (NDM-1) β- lactamase-containing plasmid has rendered some strains of Escherichia coli and K.

pneumoniae pan -resistant. The NDM-1 plasmid raises the specter that common infections, such as urinary tract infections due to E. coli, may soon be caused by organisms that are virtually untreatable.

The traditional approach for antibiotic discovery is to screen libraries of natural or synthetic products for antimicrobial activity in vitro. This strategy has yielded no new classes of antibiotics for Gram-negative bacteria over the past 50 years (Bumann D., Curr. Opin. Microbiol. 2008, 11:387-92; Fischbach & Walsh, Science 2009, 325: 1089-93). Contemporary drug discovery is aimed at identification of inhibitors of novel targets essential for virulence or growth. Attractive targets include bacterial 2-component signal transduction (2CST) systems, which typically contain a sensor kinase (SK) and a response regulator (RR) and have no mammalian homologs. At least a dozen small molecule inhibitors of 2CST systems have antimicrobial activity in vitro (Barrett & Hoch, Antimicrob. Agents Chemother. 1998, 42: 1529-36; Watanabe et al, Adv. Exp. Med. Biol. 2008, 631:229-36). Unfortunately, these compounds have not achieved clinical utility in humans. This may be due to the poor bioavailability of the inhibitors, which target the hydrophobic active site of the SK (Schreiber et al., Curr. Opin. Cell Biol. 2009, 21:325-30).

CpxRA is a 2CST system found in many drug resistant Gram-negative bacteria, including E. coli, K. pneumoniae, Haemophilus ducreyi, and the Salmonella, Shigella, Citrobacter, and Enterobacter species. A major function of CpxRA is to alleviate membrane stress by reducing the flow of protein traffic through the periplasm. Activation of CpxRA dramatically reduces the expression of virulence determinants, consistent with the fact that most virulence factors must traverse the cytoplasmic membrane to reach the cell surface (Raivio, Molecular Microbiol. 2005, 56: 1119-28; Carlsson et al. Infect. Immun. 2007, 75:3913-24; MacRitchie et al., Infect. Immun. 2008, 76: 1465-75; Spinola et al., Infect.

Immun. 2010, 78:3898-904; Vogt & Raivio, FEMS Microbiol. Lett. 2011, 326:2-11).

Uncontrolled activation of CpxRA totally cripples the ability of H. ducreyi to cause disease in human volunteers (Spinola et al., 2010). Uncontrolled activation of CpxRA also abolishes the virulence of Salmonella enterica serovar Typhimurium in a murine infection model

(Humphreys et al., Infect. Immun. 2004, 72:4654-61). Thus, activation of CpxRA impairs the ability of both an extracellular and an intracellular pathogen to infect their natural hosts.

Summary

The present invention includes methods and compounds for treating a bacterial infection. In particular, the methods disclose herein a novel way of treating a bacterial infection. Instead of inhibiting a pathway or protein, embodiments of the disclosed methods activate CpxRA. In activating CpxRA, virulence mechanisms are down regulated, thus enabling a subject's own immune system to clear an infection. In particular, an embodiment includes a method of treating an Enterobacteriaceae infection.

In an embodiment, a method includes treating a bacterial infection comprising activating CpxRA in a subject with said bacterial infection. In an embodiment, a method of treating a bacterial infection includes administering at least one activator of CpxRA, wherein the activator is at least one of nitroaniline, a nitroindole, a quinolone, or a qunioxaline, or derivatives thereof. Such derivatives can be a quinoxaline 1,4-dioxide derivative, a 5- nitroindole derivative, a 4-nitoraniline derivative, or a quinolone derivative. A 5-nitroindole derivative can be 6-nitro-2,3,4,9-tetrahydro-lH-carbazol-l-amine.

An embodiment also includes wherein an activator of CpxRA is formulated in a pharmaceutical composition. A pharmaceutical composition comprising a CpxRA activator can be administered to a subject with a bacterial infection, particularly an Enterobacteriaceae infection.

Another embodiment includes a administering an activator of CpxRA in combination with an antimicrobial. In an embodiment, an antimicrobial is an antibiotic. Brief Description of the Drawings

Fig. 1 shows the CpxRA 2CST. After binding misfolded proteins, CpxP dissociates from CpxA, activating the system. Allosteric drugs may bind anywhere on CpxA, while phosphatase inhibitors would target the active site.

Fig. 2 shows a graph indicating β-galactosidase activity (MU) of the wild type, cpxR and cpxA mutants treated with compound 2 at varying concentrations (μΜ). Values are means + SD of 3 experiments.

Fig. 3 indicates the feasibility of our assay as assessed by determining its screening window coefficient ("Z' factor") using Miller Units of the untreated wild type as a negative control and the untreated cpxA mutant as a positive control.

Detailed Description

Definitions

Administration "in combination with" one or more further therapeutic agents includes simultaneous (concurrent) and consecutive administration in any order.

The term "therapeutic agent" includes any synthetic or naturally occurring

biologically active compound or composition of matter which, when administered to an organism (human or nonhuman animal), induces a desired pharmacologic, immunogenic, and/or physiologic effect by local and/or systemic action. The term therefore encompasses those compounds or chemicals traditionally regarded as drugs, vaccines, and

biopharmaceuticals including molecules such as proteins, peptides, hormones, nucleic acids, gene constructs and the like. More particularly, the term "therapeutic agent" includes compounds or compositions for use in all of the major therapeutic areas including, but not limited to, antimicrobial agents (e.g., antibiotics, antiseptics).

The term "antimicrobial" means a compound or a composition that kills or slows/stops the growth of bacteria.

The term "inhibition" or "inhibiting" refers to a decrease of bacterial associated microorganism formation and/or growth.

"PCR" refers to the technique in which minute amounts of a specific piece of nucleic acid, RNA and/or DNA, are amplified as described in US Patent No. 4,683,195. PCR can be used to amplify specific RNA sequences, specific DNA sequences from total genomic DNA, and cDNA transcribed from total cellular RNA, bacteriophage or plasmid sequences, etc. The term "effective" refers to a sufficient amount of a composition to substantially inhibit the growth, proliferation, or survival of bacteria. The amount will vary for each composition.

The term "mammal" for purposes of treatment or therapy refers to any animal classified as a mammal, including humans, domestic and farm animals, and zoo, sports, or pet animals, such as dogs, horses, cats, cows, and the like. Preferably, the mammal is human.

A "control" is an alternative subject or sample used in an analytical procedure for comparison purposes. A control can be "positive" or "negative". For example, where the purpose of an analytical procedure is to detect a differentially expressed transcript or polypeptide in cells or tissue affected by a disease of concern, it is generally preferable to include a positive control, such as a subject or a sample from a subject exhibiting the desired expression and/or clinical syndrome characteristic of the desired expression, and a negative control, such as a subject or a sample from a subject lacking the desired expression and/or clinical syndrome of that desired expression.

The term "treatment" or "treating" is an approach for obtaining beneficial or desired clinical results. For purposes of this invention, beneficial or desired clinical results include, but are not limited to, alleviation of symptoms, diminishment of extent of disease, stabilized (i.e., not worsening) state of disease, delay or slowing of disease progression, amelioration or palliation of the disease state, and remission (whether partial or total), whether detectable or undetectable. "Treatment" can also mean prolonging survival as compared to expected survival if not receiving treatment. For the purposes herein, "treatment" does not include "preventing" or "prevention" or "prophylaxis" of a microbial infection.

The term "compound" refers to a chemical identified as an activator of CpxRA or a pharmaceutically acceptable salt, solvate, clathrate, hydrate, polymorph or prodrug thereof, and also includes protected derivatives thereof. Compounds may contain one or more chiral centers and/or double bonds and, therefore, exist as stereoisomers, such as double-bond isomers (i.e., geometric isomers), enantiomers, or diastereomers. According to this invention, the chemical structures depicted herein, including the compounds of this invention, encompass all of the corresponding compounds' enantiomers, diastereomers and geometric isomers, that is, both the stereochemically pure form (e.g., geometrically pure,

enantiomerically pure, or diastereomerically pure) and isomeric mixtures (e.g., enantiomeric, diastereomeric and geometric isomeric mixtures). In some cases, one enantiomer,

diastereomer or geometric isomer will possess superior activity or an improved toxicity or kinetic profile compared to other isomers. In those cases, such enantiomers, diastereomers and geometric isomers of compounds of this invention are preferred.

The term "polymorph" refers to solid crystalline forms of a compound or complex thereof. Different polymorphs of the same compound can exhibit different physical, chemical and/or spectroscopic properties. Different physical properties include, but are not limited to stability (e.g., to heat or light), compressibility and density (important in formulation and product manufacturing), and dissolution rates (which can affect bioavailability). Differences in stability can result from changes in chemical reactivity (e.g., differential oxidation, such that a dosage form discolors more rapidly when comprised of one polymorph than when comprised of another polymorph) or mechanical characteristics (e.g., tablets crumble on storage as a kinetically favored polymorph converts to thermodynamic ally more stable polymorph) or both (e.g., tablets of one polymorph are more susceptible to breakdown at high humidity). Different physical properties of polymorphs can affect their processing. For example, one polymorph might be more likely to form solvates or might be more difficult to filter or wash free of impurities than another due to, for example, the shape or size distribution of particles of it.

The term "hydrate" refers to a compound as described herein or a salt thereof, that further includes a stoichiometric or non- stoichiometric amount of water bound by non- covalent intermolecular forces.

The term "clathrate" refers to a compound as described herein or a salt thereof in the form of a crystal lattice that contains spaces (e.g., channels) that have a guest molecule (e.g., a solvent or water) trapped within.

The term "prodrug" refers to a drug molecule of a compound as described herein that is biologically inactive until it is activated by a metabolic process. A prodrug includes a derivative of a compound that can hydrolyze, oxidize, or otherwise react under biological conditions (in vitro or in vivo) to provide a compound of the invention. Prodrugs may become active upon such reaction under biological conditions, or they may have activity in their unreacted forms. Examples of contemplated prodrugs include, but are not limited to, analogs or derivatives of compounds that comprise biohydrolyzable moieties such as biohydrolyzable amides, biohydrolyzable esters, biohydrolyzable carbamates, biohydrolyzable carbonates, biohydrolyzable ureides, and biohydrolyzable phosphate analogues. Other examples of prodrugs include derivatives of compounds that comprise—NO,— N0₂,— ONO, or— ON0₂ moieties. Prodrugs can typically be prepared using well known methods. Antimicrobial Drug Discovery

All currently available antibiotics inhibit a limited number of enzymes that have essential functions for bacterial growth (Bumann, Curr. Opin. Microbiol. 2008;11:387-92; Fischbach & Walsh, Science 2009;325: 1089-93; Silver, IDrugs 2005;8:651-5). The dogma in the field is that only compounds that completely inhibit cell growth will prove to be clinically useful. Despite intensive efforts, few new targets or drug classes have been identified for Gram- negative bacteria in the past 50 years (Bumann, 2008; Becker et al., Nature 2006;440:303-7). Most "new" drugs are modifications of a limited number of existing scaffolds. There is a pressing need to identify novel strategies, targets and classes of drugs (Boucher et al., Clin. Infect. Dis. 2009;48: 1-12, 7, 8; Bumann, 2008; Becker et al., 2006).

2CST systems as antimicrobial targets

Modern drug discovery is aimed at identification of inhibitors of novel targets essential for virulence or growth. Attractive targets include bacterial 2CST systems, which include a sensor kinase (SK) and a response regulator (RR). In response to a stimulus, a SK

autophosphorylates at a histidine residue, and its cognate RR catalyzes the transfer of this phosphate group to itself on an aspartic acid residue. Phosphorylation of the RR leads to a conformational change that causes a response. Important to this proposal, some SKs (e.g. CpxA) also possess phospho-RR phosphatase activity. Also important to this proposal, some RRs (e.g. CpxR) can autophosphorylate using a small molecule phosphodonor, such as acetyl phosphate (Ac-P). The predominant function of most RRs is to bind to DNA and control gene transcription.

Eukaryotic signal transduction systems include kinases and phosphatases that act at tyrosine, serine, or threonine residues (West & Stock, Trends Biochem. Sci. 2001;26:369-76; Gao et al., Trends Biochem. Sci. 2007;32:225-34. ). 2CST systems are excellent

antimicrobial targets because 2CST systems are not found in mammals and perform phosphorelay on different amino acids. At least a dozen inhibitors of essential 2CST systems have been discovered that have antibacterial activity in vitro (Barrett & Hoch. Antimicrob. Agents Chemother. 1998;42: 1529-36; Watanabe et al., Adv. Exp. Med. Biol. 2008;631:229- 36; Schreiber et al., Curr. Opin. Cell Biol. 2009;21:325-30). However, only one compound, closental, is licensed for veterinary use as an antiparasitic agent. The failure to develop these compounds may be due to the poor bioavailability of most inhibitors, which target the hydrophobic active site of the SK.

The CpxRA 2CST

The CpxRA system is highly conserved in the Enterobacteriaceae (Table 1). CpxA is a sensor kinase that spans the cytoplasmic membrane and responds to extracytoplasmic signals by autophosphorylating and donating a phosphoryl group to its cognate RR, CpxR. CpxR is a DNA-binding protein that controls the transcription of at least 100 genes that permit the bacterium to adapt to stress. The CpxRA pathway usually possesses two upstream

components, CpxP and NlpE. CpxP is a periplasmic chaperone that binds CpxA and inhibits its kinase activity. The binding of a misfolded protein to CpxP leads to degradation of both the misfolded protein and CpxP. The absence of CpxP binding activates CpxA kinase activity. By some unknown mechanism, surface adhesion induces the lipoprotein NlpE to activate CpxA. Homologs of CpxRA are also present in H. ducreyi, Haemophilus influenzae, Legionella pneumophila, and other Enterobacteriaceae. Although they lack CpxP and NlpE, CpxRA is functional in these organisms.

Table 1. Homologues of E. coli CpxA, CpxR, CpxP and NlpE in the Enterobacteriaceae and other organisms

In the absence of stress, CpxA acts as a net phosphatase. When wild-type cells are grown in minimal media containing excess carbon sources such as 0.4% glucose, CpxR becomes activated. Glucose-induced activation requires CpxR, Ac-P, and acetylation of lysine 298 of the a subunit of RNAP, but not CpxA. A cpxA deletion (AcpxA) mutant responds more robustly to glucose than does its wild-type parent because the AcpxA mutant lacks phosphatase activity and accumulates activated CpxR. Mutants with cpxA* alleles have constitutively active kinase activity and produce even higher levels of activated CpxR than do AcpxA mutants. Thus, the CpxRA pathway is activated at multiple levels and by multiple signals.

Activation of CpxRA as an antimicrobial strategy

A major function of CpxRA is to alleviate membrane stress by controlling protein traffic across the cytoplasmic membrane. In pathogenic E. coli and Yersinia species, activation of CpxR by a cpxA* allele or by deletion of cpxA reduces the expression of several major virulence determinants, consistent with the fact that such factors are usually exported through the cytoplasmic membrane. These data have led to the concept that activation of CpxRA reduces bacterial virulence and that CpxRA is dispensable for bacterial survival (Raivio, Molecular Microbiol. 2005;56: 1119-28; MacRitchie et al., Infect Immun.

2008;76: 1465-75; Vogt & Raivio; FEMS Microbiol Lett. 2011;326:2-11; Humphreys et al., Infect Immun. 2004:72:4654-61 : l leusipp et al. FEMS Microbiol Lett. 2004;231:227-35). These data suggest that CpxRA would not be a good antimicrobial target, as it is not essential for virulence and its inhibition could increase virulence.

H. ducreyi is an extracellular pathogen that causes chancroid, a genital ulcer disease. We made a H. ducreyi AcpxA mutant, which activates CpxR and down regulates the expression of proteins that mediate resistance to phagocytosis and serum killing. In human volunteers, the AcpxA mutant is the only H. ducreyi mutant tested that is attenuated both for its ability to initiate infection and to cause pustules. In contrast, a H. ducreyi AcpxR mutant is fully virulent in humans. Similarly, both AcpxA and cpxA* mutants of the intracellular pathogen S. Typhimurium are highly attenuated in mice while a AcpxR mutant is fully virulent. Taken together, the data show that uncontrolled activation, not loss, of CpxR cripples the ability of two unrelated pathogens to adapt to the dynamics of the host environment during infection of their natural hosts. The strategy disclosed herein is to activate the CpxRA pathway to down regulate virulence determinants, which would then allow the host's immune system to better clear an infection.

Compounds that activate CpxR could include those that allosterically bind to CpxA and change its bias from a phosphatase to a kinase. While molecules that inhibit CpxA phosphatase activity might target the active site, allosteric effectors can bind anywhere on an enzyme (Fig. 1). Other activators might work through NlpE, CpxP, CpxR, intermediaries of central metabolism (e.g. Ac-P), or by affecting the acetylation of RNAP, which is controlled by YfiQ and CobB. (Fig. 1). Thus, compounds that activate CpxR will not necessarily be hydrophobic or have poor bioavailability.

Compounds that act by a CpxR-dependent mechanism could activate the system at multiple levels (Fig. 1). We will infer mechanism of action by examining the effect of the compounds on the cpxA mutant, as shown in Fig. 2. Compounds that cause higher levels of reporter activity in the cpxA mutant than the untreated cpxA mutant are likely acting on targets downstream of CpxA, such as CpxR, RNAP or Ac-P. Compounds that do not increase reporter activity in the cpxA mutant relative to the untreated cpxA mutant are likely acting on CpxA or upstream of CpxA. Their mechanism of action can be inferred by testing whether they activate cpxP transcription in the nlpE and cpxP mutants.

Compounds that activate CpxR would be excellent drugs if they had effects on the Enterobacteriaceae and other pathogens. The CpxA, CpxR, RNAP, YfiQ and CobB homologs of Shigella, Salmonella, Citrobacter, Klebsiella and Enterobacter are nearly identical to those of E. coli, while those of CpxP and NlpE have less homology. Compounds that act by CpxP- or NlpE-dependent mechanisms are more likely to be species- specific and would have no activity against Haemophilus species and L. pneumophila, which lack these proteins. Thus, potent compounds that act via a CpxA-dependent mechanism or a target downstream of CpxA are likely to have the broadest spectrum and have the highest priority for further characterization.

Methods of Treating Bacterial Infections

Disclosed herein is a method of treating a bacterial infection. The bacterial infection can be a Gram-negative bacterial infection. In an embodiment, a method includes treating an Enterobacteriaceae infection. Enterobacteriaceae include, but are not limited to,

Escherichia, Shigella, Salmonella, Citrobacter, Klebsiella, Proteus, Serratia, Yersinia, and Enterobacter. Infections may also be mixed, wherein multiple strains, species, and/or genuses of bacteria cause an infection. An embodiment of a method of treating a bacterial infection includes activating CpxRA. The bacteria can include any bacteria that comprise CpxRA. An embodiment of a method of treating a bacterial infection includes administering a compound that activates CpxRA. An embodiment of a method of treating a bacterial infection includes administering at least one of a nitroaniline, a nitroindole, a quinolone, a quinoxaline, or a derivative thereof. An embodiment of a method of treating a bacterial infection includes administering at least 6-nitro-2,3,4,9-tetrahydro-lH-carbazol-l -amine to a subject. In an embodiment, an activator of CpxRA can be administered in combination with at least one antimicrobial or antiseptic. These can be triclosan, nitrofurazone, bacitracin, bismuth thiol, bismuth ethanedithiol (BisEDT), ovotransferrin, lactoferrin, sodium usnate, 5- fluorouracil, sodium dodecyl sulfate (SDS), chlorhexidine, benzalkonium chloride, EDTA, silver nanopowder, silver compounds, glucose oxidase, lactose peroxidase, cadexomer iodine, and the like. An antimicrobial can be an antibiotic. An antibiotic can be an aminoglycoside (e.g., neomcycin, tobramycin, gentamicin, amikacin, etc.) a carbapenem (e.g., imipenem), a cephalosporin (e.g., cefalexin, cefprozil, ceftazidime, cefepime, ceftaroline fosamil, ceftobiprole, etc.), a glycopeptide or a lipopeptide, a macrolide (e.g., azithromycin), a monobactam (e.g., aztreonam), a penicillin (e.g., ampicillin, piperacillin), a polymixin (e.g., colistin), a oxazolidinone, a quinolone (e.g., ciprofloxacin, levofloxacin, moxifloxacin, etc.), a rifamycin (e.g., rifampin, rifaximin), a sulfonamide, a sulfone, a tetracycline (doxycycline, minocycline, tetracycline, etc.), metronidazole, linezolid, rifaximin, thiamphenicol, and trimethoprim.

Methods and compositions disclosed herein can be used to treat, for example, respiratory tract infections, acute bacterial otitis media, bacterial pneumonia, urinary tract infections, complicated infections, noncomplicated infections, pyelonephritis, intraabdominal infections, deep-seated abcesses, bacterial sepsis, skin and skin structure infections, soft tissue infections, bone and joint infections, central nervous system infections, bacteremia, wound infections, peritonitis, meningitis, infections after burn, urogenital tract infections, gastro-intestinal tract infections, pelvic inflammatory disease, endocarditis, and other intravascular infections.

Compositions

Administration of a compound may be by any suitable means that is effective for the treatment of a bacterial infection or associated disease. Compounds can be admixed with a suitable carrier substance, and are generally present in an amount of 1-95% by weight of the total weight of a composition. An embodiment includes formulating a nitroaniline, a nitroindole, a quinolone, a quinoxaline, or a derivative thereof as a composition suitable for administration to a subject. In an embodiment, the subject is a mammal. Preferably, the subject is a human.

A composition may be provided in a dosage form that is suitable for oral, parenteral (e.g., intravenous, intramuscular, subcutaneous), rectal, transdermal, nasal, vaginal, inhalant, topical, or ocular administration. Thus, a composition may be in form of, e.g., tablets, capsules, pills, powders, granulates, suspensions, emulsions, solutions, gels including hydrogels, pastes, ointments, creams, plasters, drenches, delivery devices, suppositories, enemas, injectables, implants, sprays, or aerosols. A pharmaceutical composition may be formulated according to conventional pharmaceutical practice (see, e.g., Remington: The Science and Practice of Pharmacy, (21st ed.) ed. D. Troy, 2005, Lippincott Williams & Wilkins, Philadelphia, Pa.).

In addition to the active ingredient (i.e., the compound), solid dosage forms contain a number of additional ingredients referred to as excipients. These excipients include among others diluents, binders, lubricants, glidants and disintegrants.

Diluents are used to impart bulk to the formulation to make a tablet a practical size for compression. Examples of diluents are lactose and cellulose.

Binders are agents used to impart cohesive qualities to the powered material ensuring the tablet will remain intact after compression, as well as improving the free-flowing qualities of the powder. Examples of typical binders are lactose, starch and various sugars.

Lubricants have several functions including preventing the adhesion of the tablets to the compression equipment and improving the flow of the granulation prior to compression or encapsulation. Lubricants are in most cases hydrophobic materials. Excessive use of lubricants can result in a formulation with reduced disintegration and/or delayed dissolution of the drug substance.

Glidants are substances that improve the flow characteristics of the granulation material. Examples of glidants include talc and colloidal silicon dioxide.

Disintegrants are substances or a mixture of substances added to a formulation to facilitate the breakup or disintegration of the solid dosage form after administration.

Materials that serve as disintegrants include starches, clays, celluloses, algins, gums and cross-linked polymers. A group of disintegrants referred to as "super-disintegrants" generally are used at a low level in the solid dosage form, typically 1% to 10% by weight relative to the total weight of the dosage unit. Croscarmelose, crospovidone and sodium starch glycolate represent examples of a cross-linked cellulose, a cross-linked polymer and a cross-linked starch, respectively. Sodium starch glycolate swells seven- to twelve-fold in less than 30 seconds effectively disintegrating the granulations that contain it. EXAMPLES

The traditional approach for developing antibiotics has been to screen libraries of compounds for killing or growth inhibitory activity in vitro. The dogma in the field is that only compounds that target essential enzymes or result in complete inhibition of growth are useful. CpxRA is dispensable for virulence, and AcpxA and cpxA* activation mutants have only slight growth defects in vitro. Thus, CpxRA would be routinely dismissed as a target. Compounds that activate CpxRA should have little effect on bacterial growth in vitro, unless they also have effects on other targets. Thus, these compounds would likely be missed in traditional screenings.

Identification of drugs that target bacterial 2CST has focused exclusively on inhibitors. Identifying compounds that activate a 2CST system is a paradigm shift for the field. We have named this new class of molecules astabiotics (antimicrobial signal transduction activators). Astabiotics (activators) of CpxRA have therapeutic potential for the Enterobacteriaceae and other species. More importantly, if our approach is successful, astabiotics could be sought for different targets in many bacterial pathogens and potentially revolutionize the field.

Example 1: Determination of Z' Factor

E. coli reporter strains have been used for HTS of inhibitors of 2CST systems. When a cpxA mutant is grown in minimal media containing glucose, CpxR is phosphorylated due to the loss of CpxA phosphatase activity and the ability of CpxR to accept phosphate groups from Ac-P. The gene whose transcription is most highly induced by CpxR is cpxP. A series of E. coli K12 strains bearing a chromosomal lacZ reporter downstream of a cpxP promoter were obtained from Drs. Alan Wolfe and Tom Silhavy (Table 2). A 384 well plate assay was developed to detect compounds that activate the reporter in the wild type so that it mimics the cpxA mutant in its level of cpxP transcription.

Table 2. E. coli reporter strains

The cpxR and cpxA mutants and the wild-type E. coli were grown in minimal medium containing 0.4% glucose at 37°C. All strains entered stationary phase after 7 h of growth. The cpxR mutant exhibited less β-galactosidase activity than the wild type in all phases of growth. After 5 h of growth, the cpxA mutant exhibited approximately 10-fold greater cpxP promoter- dependent β-galactosidase activity than the wild type (data not shown). The feasibility of our assay was assessed by determining its screening window coefficient (Kohanski et al., Cell 2008;135:679-90), or "Z' factor", using Miller Units of the untreated wild type as a negative control and the untreated cpxA mutant as a positive control (Fig. 3). An excellent Z' factor is 0.5. The calculated Z' factor herein was 0.4, suggesting that there was sufficient separation in the variability of the values such that the assay distinguishes between the strains.

Example 2: Library Screening

A nonredundant, structurally diverse, drug-like, small molecule library of 35,904 compounds was obtained from ChemDiv and ChemBridge collections and was screened at 10 μΜ against the wild-type strain, E. coli strain PAD282. Each plate contained 352 wells of wild-type PAD282 in the glucose-containing medium treated with individual compounds dissolved in 0.8% DMSO and 16 wells of the untreated wild type and the untreated cpxA mutant in the same medium containing 0.8% DMSO as controls. The OD₄₂₀, OD₅₅₀ and OD₆₀₀ of the wells were measured at time zero and before and after treatment with the All in One β-galactosidase reagent (Thermo Scientific Inc., Rockford, IL) after 5 h of growth. By measuring the OD₄₂o, OD550 and OD₆oo of the wells at the beginning of the assay and before and after the addition of the β-galactosidase reagent, we are able to exclude false positives (compounds that are yellow) and bactericidal antibiotics. The activity of the reporter was calculated in Miller Units. Treated wells with Miller Units greater than the mean plus 3 times the standard deviation of the untreated wild-type wells on its plate were picked as "hits" (Kohanski et al., Cell. 2008;135:679-90).

The final pathway for killing of E. coli by bactericidal antibiotics is the induction of hydroxyl radicals, which is in part mediated by CpxRA activation. The CpxR-activating mutants AcpxA and cpxA* have only slight growth defects in vitro. When interpreting the data, growth of the treated wild type was considered. Molecules that partially or do not inhibit growth likely activate CpxRA directly; those causing complete growth inhibition may activate CpxRA indirectly, as do bactericidal antibiotics. 360 hits were identified in the initial screen; 346 of these were available for rescreening. Eighteen of the 346 compounds activated the reporter in the rescreening assay. Two compounds completely inhibited growth; 16 partially or did not inhibit cell growth. The 16 compounds belonged to 4 structural groups, 3 of which had more than 1 member;

clustering of compounds lends validity to the assay. Seven compounds were nitroaniline or nitroindole derivatives (Table 3), which are predicted to be cell-permeable. Six were analogues of the quinolones norfloxacin and ciprofloxacin (Table 4); none were drugs that are currently in use. These compounds should have desired permeability and other pharmacokinetic properties. Two compounds were quinoxaline 1,4-dioxide derivatives, a class of drug that is used as an antimicrobial agent in animal feeds (Table 5). One compound, 4,7-dimethyl-[l,2,5]oxadiazolo[3,4-d]pyridazine 1,5,6-trioxide, has vasorelaxant and antiplatelet activity. (Table 5).

Table 3. 4-nitroanaline derivatives.

Table 4. Quinolone analogues.

Table 5. Miscellaneous Compounds

Further testing was performed using 110 compounds representing the 4 drug classes; we purchased 80 nitroanilines or nitroindoles, 17 quinolones, 12 quinoxalines, and the vasorelaxant compound. Initial dose response studies were performed using a log scale from 1 nM to 100 μΜ with the wild- type strain and the cpxR reporter strains. As a control, the bactericidal antibiotic, ciprofloxacin was included. Most of the compounds reproducibly activated the reporter in the wild type at concentrations between 10 and 100 μΜ; none activated the reporter in the cpxR mutant, confirming the validity of the screen. As expected, ciprofloxacin activated the reporter in the wild type strain and completely inhibited cell growth (Group C, Table 6). Our top 5 compounds in terms of inducing reporter activity with little or no growth inhibition in the wild type included 2 quinolones and 3 nitroanilines / nitroindoles (Table 6).

Subsequent dose response studies were tested using the wild type, the cpxA, and cpxR mutants with the 5 compounds at 2-fold dilutions ranging from 1.25 to 160 μΜ. Activity of the reporter and viable CFU as a function of drug concentration were measured. Compound 1 optimally activated the reporter in the wild type at 15 μΜ, causing 9000 Miller Units of β- galactosidase activity, and had modest effects on bacterial viability (not shown). At 40 μΜ, compound 2 caused 30,000 Miller Units of β-galactosidase activity in the wild type (Fig. 2) and had no effect on bacterial viability. Neither compound activated the reporter in the cpxR mutant; both activated the reporter in the cpxA mutant (Fig. 2; data not shown), suggesting that they both act on targets downstream of CpxA. Compound 3 only activated the reporter after inducing cell death; compounds 4 and 5 were less potent than 1 and 2 (Table 7). Example 3: Identifying activators of CpxRA from a

structurally diverse compound library

Our initial screen of 36,000 compounds yielded 2 promising compounds that must activate CpxRA through a target downstream of CpxA. Further compounds are tested to identify those that act directly on CpxA, especially those that convert CpxA into a kinase; such compounds may produce the highest levels of activated CpxR. Our initial screen had a hit rate of approximately 1%. Approximately 5% of the hits were confirmed, yielding a true hit rate of 0.05%. Most biochemical screens done by the Chemical Genomics Core Facility have typical hit rates of 0.1-0.2%. We employed live bacteria in the screening process, not a purified enzyme. Thus, it is not surprising that our confirmed hit rate is somewhat lower than that of a typical biochemical screen.

An existing library of 170,000 nonredundant, small molecule compounds selected from the ChemBridge and ChemDiv collections in the Chemical Genomics Core Facility is being screened to identify additional molecules and classes of compounds that activate CpxRA. The compounds in the library obey criteria for good absorption, distribution, metabolism and excretion profiles. Compounds are screened at 10 μΜ. Initial "hits" are rescreened with the wild type; those that completely inhibit growth are excluded. Assuming a confirmed hit rate of 0.05%, approximately 70 additional leads are expected from the remaining 134,000 in the collection. The structures of the hits are analyzed to identify new groups of compounds that activate CpxR-dependent transcription.

Example 4: Determination of compound potency and mechanism of action

To examine whether compounds activate cpxP transcription by a CpxR-dependent mechanism, both strains are treated with each compound in log increments ranging from 1 nM to 100 μΜ. Compounds that activate cpxP transcription in both strains act by a CpxR- independent mechanism and are not brought forward. Based on preliminary data, 95% of the compounds are expected to be CpxR-dependent.

If a compound appears to act through CpxA, testing to determine if compounds inhibit CpxA phosphatase activity or change CpxA to a kinase are performed for their ability to activate the reporter in the wild type grown in the presence or absence of glucose.

Compounds that activate in the absence of glucose favor CpxA kinase activity or those that only activate in the presence of glucose, which is required for the production of Ac-P, likely inhibit its phosphatase activity. If a compound appears to act on a target downstream of CpxA, then the compound is acting directly on CpxR or on other downstream components of the system. Acetylation of lysine 298 of the a subunit of RNAP is required for glucose- induced CpxR-mediated cpxP transcription and is mediated by the acetyltransferase YfiQ and decreased by the deacetylase CobB. Thus, molecules that enhance the activity of YfiQ or inhibit the activity of CobB can increase CpxR-dependent transcription. Reporter strains are available to characterize molecules that act downstream of CpxA.

High priority compounds are tested for their ability to increase the transcription of native cpxP and another CpxR-dependent promoter (degP) by quantitative RT-PCR (qRT- PCR), and also perform qRT-PCR experiments.

Example 5: Preclinical Studies

CpxRA is present in many drug resistant gram-negative bacteria, including E. coli and S. Typhimurium. The CpxA and CpxR proteins of E. coli and S. Typhimurium are virtually identical. There is an established animal model of E. coli urinary tract infection that uses female CBA/J mice and mimics urinary tract infections in women. There is also an established animal model of systemic S. Typhimurium infection that uses male BALB/c mice and mimics typhoid fever. These two infection models were used to test whether these compounds are safe to administer to uninfected female CBA/J mice and male BALB/c mice.

Table 7. CpxRA Activators

Optimal Calculated

Molecular

Structure Category concentration that dose

Weight

activates CpxRA

Compound 1 is l-ethyl-6-fluoro-7-(4-(4-methylbenzoyl)piperazin-l-yl)-4-oxo-l,4- dihydroquinoline-3-carboxylic acid, a norfloxacin derivative (ChemDiv 5706-4098) that is not in clinical use. Compound 2 is 6-nitro-2,3,4,9-tetrahydro-lH-carbazol-l-amine, a 5- nitroindole derivative (ChemBridge 5302860) that is not in clinical use. The optimum concentration of each compound that activates CpxRA in E. coli is listed in the table.

Activation of CpxRA causes some growth inhibition of E. coli. A traditional antibiotic, such as the quinolone ciprofloxacin, causes total growth inhibition in E. coli. For this reason ciprofloxacin was used as a control. Although both compound 1 and compound 2 caused moderate growth inhibition of E. coli, neither totally inhibited growth and therefore are not traditional antibiotics.

Based on their molecular weights, a dose of each compound was calculated in mg/kg that results in an initial plasma concentration that equals the optimal concentration required to activate CpxRA in E. coli. The initial concentration (C ) = (Dose) x fraction absorbed (F) / Volume of Distribution (Vd).

In the screen, the compounds activated the CpxRA system in a phosphate buffer containing 0.8% DMSO. A vehicle containing DMSO can be used for subcutaneous injections.

Example 5A: Toxicity Studies

The toxicity study is designed to examine whether a compound causes morbidity or mortality in uninfected BALB/c male and CBA/J female mice. For the toxicity studies the highest dose (10X the calculated dose) is tested for each compound given subcutaneously. If toxicity is observed at that dose, the dose is reduced by 0.5 log. For example, one mouse of each species is injected with 100 mg/kg of compound 2, and 1 mouse with a vehicle control subcutaneously and observed for acute toxicity for 12 h. If no acute toxicity is observed, 100 mg/kg twice a day for 3 days is administered to the compound treated CBA/J mouse and 2 additional CBA/J mice. The vehicle control is administered to the 2 remaining CBA/J mice in the same schedule (n= 5 CBA/J mice). If no acute toxicity is observed, the compound and the control is administered twice a day for 5 days to BALB/c mice in the same manner (n=5 BALB/c mice). If acute toxicity is observed, the dose is decreased by ½ log to 30 mg/kg and administered to one mouse of each species. If there is no acute toxicity at 30 mg/kg, the compound is administered to a total of 3 mice and the vehicle control to 1 mouse of each species in the regimens described above. If acute toxicity is observed at 30 mg/kg, the dose is decreased by 1 log to 10 mg/kg and administered to 1 mouse of each species. If there is no acute toxicity at this dose, the compound is administered to a total of 2 mice and the vehicle control to 1 mouse of each species. If there is acute toxicity at the 10 mg/kg dose, the compound is judged to be too toxic for further testing. The total number of mice used to test each compound is 10 (5 CBA/J mice; 5 BALB/c mice). For animals who complete a multidose regimen, -50 μΐ of blood will be collected via tail snip 12 hours after the last dose of the compound; serum will be stored for future pK studies. Toxicity testing for compound 2 has been completed. The female CBA /J mice tolerated the 100 mg/kg multidose regimen, while the male BALB/c mice tolerated a 10 mg/kg multidose regimen.

Example 5B: Tests of efficacy in murine models of infection

Two murine models were selected that are accepted surrogates for human diseases and provide contexts in which compounds may be used as treatment for disseminated Gram- negative infections and for urinary tract infections. Testing of candidate drugs in animals infected with S. Typhimurium SL1344 and uropathogenic E. coli CFT073 confirm efficacy.

To demonstrate that compounds activate CpxR in these pathogenic strains, PcpxP'- lacZ chromosomal reporters are introduced into these bacterial strains to confirm that the compounds activate CpxR.

Based on their molecular weights, a dose of each compound is calculated in mg/kg that should result in an initial plasma concentration that equals the optimal concentration determined with the wild-type E. coli reporter strain. The initial concentration (C ) = (Dose) x fraction absorbed (F) / Volume of Distribution (Vd). The compounds are administered in a vehicle containing DMSO subcutaneously, assuming F=100 . The Vd (20-25 ml) is also assumed to correspond to the total body weight of the mice (20-25 g). For both infection models, groups of infected animals will be treated with the vehicle, the calculated dose, and 10 and 0.1 times the calculated dose of each compound. For example, compound 2 has a molecular weight of 231 Da and optimal activity at 40 μΜ; unless the results of toxicity testing indicate otherwise, doses would be 100, 10 and 1 mg/kg.

A. S. Typhimurium challenge model (Purdue University)

S. Typhimurium is an intracellular pathogen that disseminates to mesenteric lymph nodes (MLN), the liver, and spleen and mimics human enteric fever. This model can assess whether treatment with a candidate compound reproduces the S. Typhimurium AcpxA and cpxA* mutant phenotypes. 7

In this model, male BALB/c mice are fed 10' CFU of S. Typhimurium SL1344 and observed for 14 days. Mice develop symptoms on day 2 and begin to die on day 7; the mortality rate approaches 100% in 14 days. To simulate the multiple day treatment regimens given for enteric fever, animals are treated with antibiotics twice a day for 5 days after inoculation. The primary endpoints are survival and mean time to death. Secondary endpoints are CFU/g of tissue in liver, MLN, and spleen in moribund mice and in mice that survive for 14 days. The 4-6 week old male BALB/c mice are used to determine if candidate compounds can reduce the virulence of S. Typhimurium and affect the mean time to death of the mice. Oral infection is accomplished by feeding the mice 10 bacteria with a sterile pipette.

Following the inoculation, mice are monitored for 14 days. The primary endpoint is lethality. The mortality rate approaches 100% in 14 days. To simulate the multiple day treatment regimens usually given for human enteric fever, animals are usually treated with antibiotics twice a day for 5 days after infection is established in this model. Treatment with effective antibiotics usually reduces mortality by 90%.

For these tests, 4 groups of 6 infected mice are treated with the vehicle and 3 doses of the compound over a 2 log scale (e.g., 1 mg/kg, 10 mg/kg, 100 mg/kg) subcutaneously given twice a day over 5 days, 1 day after infection. Blood is obtained on the fifth day of treatment 2 hours post a.m. injection, 8 hours later (prior to p.m. injection), and at the time of sacrifice the following day. Serum is frozen and stored for future pK studies when the compound is proved to be efficacious.

Death or survival for 14 days is the endpoint. All mice are euthanized by cervical dislocation under isoflurane anesthesia. All mice are necropsied and the amount of bacteria is determined in their liver, spleen, and lymph nodes as secondary endpoints.

B. Uropathogenic E. coli challenge model (University of Michigan)

The urinary tract infection model utilizes a UPEC strain isolated from a patient with pyelonephritis and mimics human ascending urinary tract infections in women (Hvidberg et al., Antimicwb. Agents Chemother. 2000, 44: 156-63; Alteri et al., PLoS Pathog.

2009,5:el000586; Allou et al., Antimicwb. Agents Chemother. 2009, 53:4292-7). UPEC is extracellular and also forms intracellular bacterial communities that act as a reservoir for recurrences. This model was chosen due to the emergence of resistant E. coli as a cause of urinary tract infections. In addition, activation of CpxR inhibits the expression of Pap pili, which is essential for the ability of UPEC to cause pyelonephritis. In this model, female CBA/J mice are transurethrally inoculated with approximately 10 CFU of E. coli CFT07, which results in infection of the urine, bladder and kidneys within 24 h of inoculation. Within 3 days, bacteria reach a density of approximately 10⁶ CFU/ml of urine, 10⁴ CFU/g of bladder tissue and 10 CFU/g of kidney tissue. Antibiotics are given twice a day for 3 days after inoculation. This model does not generally result in mortality. The primary endpoints are the quantity of bacteria in the urine, bladder and kidney and the pathology of the bladder and kidney in animals sacrificed 18 h after treatment.

In the UPEC model, 4 to 6 week old female CBA/J mice (Harlan Sprague Dawley) 18-22 gm, are acclimated to the animal facilities for 1 week after receipt from the breeder. During acclimation and throughout the experiment, mice have ad libitum access to water and Purina Lab Chow. Twenty-four hours prior to each experiment, a specimen of spontaneously voided urine is collected in a sterile petri dish from each mouse. Urine is cultured on Luria agar and mice with bacteriuria present >10 CFU/ml (our lower limit of detection) are not used (1-5% of mice examined).

Mice are anesthetized with pentobarbital (0.03ml of a 25 mg/ml solution, IP = 40 mg/kg). Peri-urethral and peri-anal areas are swabbed with 10% povidone iodine solution and excess solution is removed with a sterile swab soaked with sterile PBS. A sterile 25 mm long polyethylene catheter (0.28 mm I.D., 0.61 mm O.D., Clay Adams, Parsippany, NJ) is gently inserted into the bladder through the urethra. A 30 gauge needle attached to a tuberculin syringe containing the microbial suspension (2 x 10 CFU/mouse) is inserted in the catheter lumen and 0.05 ml is infused into the bladder over 30 sec. which is controlled by an infusion pump (Harvard Apparatus). Thermal support will be provided during and after catheterization by using an incandescent heat lamp. The infectious agent is E. coli CFT073 - a urine and blood isolate from a patient with clinical symptoms of acute pyelonephritis. The parent strain is uropathogenic in CBA mice.

The catheter is removed immediately after mouse challenge. Previous determinations have shown, by culture and use of India ink, that the inoculum does not reflux into the ureters by this procedure. After bacterial challenge, mice appear normal; they eat and drink normally; they move about the cage normally; they do not pay any unusual attention to their periurethral area, and their fur is smooth. All distressed and moribund animals are euthanized immediately.

The animals are treated with the compounds 1 day after infection using the 3 day multidose regimen described above. The mice are euthanized 24 hours post-treatment. Prior to euthanasia, urine is collected aseptically from each mouse and cultured. Mice are euthanized by anesthetic overdose (isoflurane) by using a Plexiglas container with anesthetic soaked cotton balls in the bottom. After euthanasia, the bladder and kidneys are removed aseptically and cultured. Microbial counts of urine and homogenates of bladder and kidney after euthanasia are measured.

Claims

1. A method of treating a bacterial infection comprising activating CpxRA in a subject with said bacterial infection.

2. A method of treating a bacterial infection comprising administering at least one activator of CpxRA, wherein the activator is a nitroaniline, a nitroindole, a quinolone, a quinoxaline, or a vasoactive agent; or derivatives thereof.

3. A method of treating a bacterial infection comprising administering a pharmaceutical composition comprising at least one activator of CpxRA, wherein the activator is a nitroaniline, a nitroindole, a quinolone, a quinoxaline, or a vasoactive agent; or derivatives thereof.

4. The method according to claim 2 or claim 3, wherein the activator is a quinoxaline 1,4-dioxide derivative.

5. The method according to claim 2 or claim 3, wherein the activator is a 5 -nitroindole derivative.

6. The method according to claim 5, wherein the 5-nitroindole derivative is 6-nitro- 2,3,4,9-tetrahydro- lH-carbazol- 1-amine.

7. The method according to claim 2 or claim 3, wherein the activator is a 4-nitoraniline derivative.

8. The method according to any one of claims 1 to 7, further comprising administering an antimicrobial.

9. The method according to claim 8, wherein the antimicrobial is an antibiotic.

10. The method according to any one of claims 1 to 9, wherein the bacterial infection is due to bacteria comprising CpxRA system or homolog thereof.

11. The method according to claim 10, wherein the bacterial infection is an

Enterobacteriaceae infection.

12. The method according to claim 10, wherein the bacterial infection is a Klebsiella or Escherichia infection.

13. The method according to claim 12, wherein the Escherichia infection is an E. coli infection.

14. The method according to claim 12, wherein the Klebsiella infection is an K.

pneumoniae infection.