LED209 acts as a prodrug, allosterically modifying QseC.

We have previously reported that LED209 significantly attenuates virulence in the human pathogens EHEC,Salmonella, andF. tularensis without inhibiting pathogen growth (4). Compounds that interfere with pathogen virulence represent a novel class of antimicrobial drugs that offer an alternative treatment against microbes that are resistant to current antimicrobials. Using QseC reconstituted in liposomes, we showed biochemically that LED209 acted on the QseC receptor, preventing binding to its signals and consequently its autophosphorylation in response to them (4). To further determine the specificity of LED209 to QseC, we took a genetic approach. We evaluated the expression of key virulence genes after treatment with LED209in vitro in the EHEC wild type (WT) and in isogenicqseC andqseE mutants. The histidine sensor kinases QseC and QseE positively and negatively regulate virulence gene expression in EHEC, respectively, and QseE also responds to epinephrine as a signal (20). TheqseC mutant was refractory to treatment with LED209, whereas theqseE mutant remained responsive to LED209 treatment, further indicating that LED209 acts through QseC (Fig. 1A and B). Importantly, treatment of the WT with LED209 phenocopied aqseC mutation (Fig. 1A). Because QseC phosphorylates three RRs (17), we delved into mapping through which RR LED209-mediated QseC inhibition is decreasing virulence expression in the human pathogen EHEC. We chose to map this circuit in EHEC, because the QseC signaling cascade has been studied in exquisite genetic and biochemical mechanistic detail in this pathogen, with the mode of action of all three response regulators (QseB, KdpE, and QseF) being defined (16–29). As expected, expression of the virulence genes was decreased in a ΔqseBC mutant (seeFig. S1B in the supplemental material), and this mutant did not respond to LED209, similarly to the ΔqseC mutant (Fig. 1A). QseC-dependent expression of the virulence genes (LEE) occurs through the KdpE RR (17,19), and concurrent with previous reports, expression of these genes was decreased in the ΔkdpE mutant, which is also nonresponsive to LED209 and phenocopies the LED209 decreased virulence gene expression in the WT (seeFig. S1), akin to the phenotype of the ΔqseC mutant (Fig. 1A) (11,12,17,30). It is noteworthy that when using animal models, ΔqseC, ΔqseBC, and ΔkdpE mutants are all attenuated for infection (seeFig. S1). QseC homologues are present in more than 25 plant and animal pathogens, suggesting that QseC inhibitors might have broad effectiveness against Gram-negative pathogens (Fig. 1C).

Because of the high conservation of QseC across numerous bacterial species, we sought to better understand the mechanism by which LED209 inhibited QseC, as LED209 represents a potential broad-spectrum therapeutic. LED209 acts as a prodrug, and upon interaction of LED209 with its target QseC within the bacterial cell, it acts as a prodrug by loss of an aniline group to release the warhead of LED209, isothiocyanate OM188 (Fig. 2A). The delivery of OM188 via LED209 avoids the metabolism and/or degradation experienced by OM188 en route to the target. It is worth noting that conversion of LED209 to OM188 via cleavage of the aniline occurs only within the bacterial cell and presumably in close proximity to QseC. This would help account for the remarkable potency of LED209. We cannot detect OM188 or the cleaved aniline in mammalian tissues (data set available from the author). This is a bacterium-specific hydrolysis of the prodrug, which alleviates concerns of host toxicity, which our toxicity studies described below confirm to be negative. Treatment of EHEC with 500 nM of OM188 significantly reduced the expression of the EHECler gene that encodes a transcriptional regulator that activates expression of key virulence genes responsible for the EHEC-mediated hemolytic colitis, theeae gene encoding an essential adhesin to establish gut colonization, and the gene (stx2a) encoding the Shiga toxin that leads to HUS following EHEC infections (Fig. 2B). Additionally, treatment with OM188 also reduced the formation of attaching and effacing (A/E) lesions, a hallmark of EHEC infection and a requisite for gut colonization and diarrheal disease (Fig. 2C and D). Moreover, OM188 also decreasedSalmonella enterica serovar Typhimurium intramacrophage replication (Fig. 2E), which is also a QseC-dependent virulence phenotype inhibited by LED209 (seeFig. 5D).

An azide was added to OM188 to perform click chemistry, generating OM201. We determined that OM201 directly binds to QseC (Fig. 3A and B) and that OM201 is still active to decrease virulence gene expression (Fig. 2C). Mass spectrometry analysis confirmed that OM188 directly binds to two lysine residues (K256 and K427) on QseC (Fig. 3D). Importantly, K256 is conserved in the QseC from allE. coli strains, including EHEC, and fromSalmonella enterica,Shigella species,Citrobacter rodentium,Enterobacter species,Klebsiella pneumoniae,Erwinia species,Pectobacterium atrosepticum, andF. tularensis (Fig. 3E). K427 is conserved only inE. coli andShigella species (Fig. 3E). We generated site-directed mutants of QseC in both K256 and K427, changing K to R to assess the role of these lysines in QseC function (strains RR03 and RR04). Both QseC mutant proteins were expressed normally and folded properly (data not shown). However, neither mutant was able to activate expression of the EHEC virulence genes (Fig. 3F), suggesting that modification of either of these lysines impairs QseC function, akin to the block observed by treatment with LED209 or its warhead OM188 (Fig. 1 and2).

SAR of LED209 shows that minor structural modifications significantly impact its activity.

An extensive SAR of LED209 (seeTable S6 in the supplemental material) was conducted and, for simplicity, the structure of LED209 was divided into four major regions (Fig. 4A). Modifications to the left-hand aniline group (part A) abolished the bioactivity of LED209 (Fig. 4B). Treatment with CF290 resulted in a 2-fold increase ofler, reversing the strong antagonist activity of LED209 into a slight agonist of EHEC virulence. CF287 also lost the antagonist activity seen with LED209 and returned expression levels ofler andstx2a to those seen in the untreated control group. Alterations to part A likely disrupt LED209’s ability to act as a prodrug and release its active component OM188. Modifications to part C, as exemplified in compound CF252, also reversed the antagonist behavior of LED209. In contrast, replacing the benzene ring located in part D with a cyclohexane (CF283) did not modify the antagonist activity compared with the parent LED209. CF283 significantly reduced the expression of bothler andstx2a (Fig. 4B).

A commercially available version of LED209 (Sigma-Aldrich), sold as LED209 hydrate, was devoid of LED209-like antagonist activity against EHEC. Treatment with LED209 hydrate (Sigma LED209) did not alter virulence gene expression levels compared with those of the untreated control (Fig. 4C) and also failed to decrease A/E lesion formation (Fig. 4D and E). The origins of the problems with the Sigma-Aldrich material were further investigated (data set available from the author) and showed that the Sigma-Aldrich LED209 product did not contain LED209 but rather a mixture of CF104 and OM124 (sample B in the data set; all other samples are different batches of LED209 synthesized by us), which are both inactive compounds (seeTable S6 in the supplemental material). To provide consistency for future studies, a large-scale batch of LED209 was manufactured according to Good Laboratory Practice (GLP) regulations, and it works as well as all of our previous LED209 batches to block bacterial virulence expression (Fig. 4F).

Treatment with LED209 protects againstS. Typhimurium andF. tularensis murine infections.

Salmonella enterica serovar Typhimurium carries a homolog of the EHEC QseC (87% similarity). QseC acts globally inS. Typhimurium to regulate key virulence genes required for flagellar motility, invasion of epithelial cells, survival within macrophages, and murine infection, with theqseC mutant being nonvirulent in mice (4,8,31). A single dose of LED209 (20 mg/kg of body weight) or a vehicle control was administered orally to mice 30 min prior to intraperitoneal injection of a lethal dose ofS. Typhimurium. Seventy-two hours after infection, 25% of the untreated mice remained alive, whereas 60% of the LED209-treated group remained alive (Fig. 5A). Twenty-five percent of the untreated mice survived up to 6 days after infection, while 50% of the LED209-treated group survived up to 6 days after infection. To determine if a more rigorous dosing regimen would increase protection againstS. Typhimurium infection, LED209 (20 mg/kg) was administered at 3 h before, at the time of, and 3, 6, 9, and 12 h after intraperitoneal injection of a lethal dose ofS. Typhimurium. At 48 h, only 14% of the untreated mice remained alive, with none surviving beyond 72 h postinfection, whereas 43% of the LED209-treated mice remained alive and survived up to 8 days after infection, when these animals were sacrificed (Fig. 5B). Expression ofsifA, an effector secreted via the type III secretion system SPI-2 and required for vacuolar replication, was significantly decreased when LED209 was administered duringin vitro growth in minimal medium (Fig. 5C). The addition of LED209 also decreased survival ofS. Typhimurium within macrophages by 4 logs (Fig. 5D).

F. tularensis also encodes a homolog of the EHEC QseC (57% similarity) that serves as the sole histidine sensor kinase encoded within theF. tularensis genome. We previously demonstrated that LED209 treatment decreases the expression of severalF. tularensis virulence genes and markedly decreases macrophage survivalin vitro. To explore the dosing requirements to yield protection against a lethal dose of intranasalF. tularensis, a single dose of LED209 was administered orally to mice either 1, 3, 6, 9, or 24 h prior to infection or alternatively 1, 3, 6, 9, or 24 h after infection (Fig. 5E and F). Pretreatment with LED209 prior toF. tularensis infection provided some protection at the 24-h time point, with 25% of the untreated animals surviving by day 7, while 85% of the treated animals were alive. By day 8, 10% of the untreated animals and 50% of the treatment animals were alive. In contrast, a single dose of LED209 administered at 3 or 6 h after infection provided more significant protection compared to that of the untreated mice. On day 7 after infection, 85% of the mice treated at 3 h postinfection, 75% of the mice treated at 6 h postinfection, and 85% of the mice treated 24 h postinfection remained alive, while only 20% of the untreated mice remained alive. On day 8, 85% of the mice treated at 3 h, 60% of the mice treated at 6 h, and 60% of the mice treated at 24 h survived, while only 20% of the untreated animals remained alive. On day 9, 80% of the mice treated at 3 h, 50% of the mice treated at 6 h, and 40% of the mice treated at 24 h survived, while only 10% of the untreated animals were alive. Only 10% of the untreated mice survived 12 days postinfection, whereas 80% and 50% of the mice treated at 3 or 6 h postinfection, respectively, survived. The animals that survived were sacrificed at day 12.

LED209 decreases biofilm formation in EAEC O104:H4 and in several multidrug-resistant clinical isolates of rUTIs.

The addition of 5 nM LED209 significantly reduced biofilm formation by the enteroaggregativeE. coli (EAEC) O104:H4 strain isolated during the German outbreak of 2011 (37% reduction; ***,P = 0.0006) and by the uropathogenicE. coli (UPEC) strain UTI89 (35% reduction; *,P = 0.0486); however, LED209 did not significantly reduce biofilm formation by the UPEC strain CFT073 (8% reduction;P = 0.2682) (Fig. 6A). Additionally, treatment with LED209 significantly reduced the expression ofstx2a in EAEC O104:H4 (Fig. 6B). This EAEC Stx-positive strain was responsible for a diarrhea and uremic syndrome outbreak in Germany in 2011 that led to serious morbidity and mortality rates (32). Multidrug-resistant clinical isolates from patients with recurrent urinary tract infections (rUTI) were collected and isolated from urine and bladder biopsy specimens, and the samples were characterized to determine their probable species. The addition of LED209 significantly reduced biofilm formation by several of the isolates. UPEC isolates 6, 8, and 25 achieved reductions of 43%, 48%, and 43% in biofilm formation, respectively (Fig. 6C and D). Furthermore, LED209 proved efficacious againstPseudomonas andKlebsiella isolates, reducing biofilm formation by 26% in isolate 18 and by 21% in isolate 20 (Fig. 6C and D).

Toxicology and pharmacokinetics of LED209.

Our preliminary toxicology studies were performed by treating mice daily with 5, 20, and 80 mg/kg of LED209 for 14 days. These studies suggested that LED209 did not present any toxicity, with treated and untreated animals having similar body weights, blood chemistry, complete blood counts (CBC), and levels of T4 (as a measure of thyroid function) and no liver toxicity (seeTable S2 toS5 andFig. S2 in the supplemental material). No effects of LED209 in the hERG channel (seeTables S2 toS5 andFig. S2 toS4) were observed as measured by a radioligand binding assay in which 10 µM LED209 was tested in duplicate for its ability to block the interaction of 1.5 nM astemizole to the hERG channel expressed recombinantly in HEK293 cells. Minimal inhibition of binding (4%) was observed. An analysis of LED209 plasma levels after oral gavage revealed that the compound shows good oral bioavailability but nonlinear pharmacokinetics by this route (seeFig. S3). Plasma area under the concentration-time curve (AUC_last, which is defined as the area under the concentration-time curve from time 0 up to the last measurable concentration) increased more than proportionally with doses from 5 to 80 mg/kg, possibly due to inhibition of drug efflux pumps in the gastrointestinal tract or inhibition of drug metabolism enzyme in the gut or liver at high drug concentrations. Exposures at doses above 80 mg/kg increased less than proportionally with dose. We found that at these higher doses, absorption was incomplete, and a significant amount of compound was simply excreted unchanged in feces. In contrast, LED209 pharmacokinetics via the intravenous route from 5 to 20 mg/kg is linear (4). Despite a relatively short terminal half-life of 40 to 100 min in mice, LED209 showed high peak concentrations in plasma at each dose, which may serve as an efficient means of delivering large amounts of compound into bacterial cells for generation of the active OM188 warhead. Although plasma exposures with LED209 were high, limited penetration of the blood-brain barrier was observed (seeFig. S4).

Synthesis of compounds.

LED209 was synthesized as previously described (4). GLP synthesis of 1 kg of LED209 was conducted by Symmetry Biosciences. LED209 hydrate (L8919) was purchased from Sigma.

OM188.

To a stirring, room temperature solution of 4-amino-N-phenyl-benzenesulfonamide (4) (CF103; 506 mg, 2.04 mmol) was dissolved in dry tetrahydrofuran (THF; 10 ml), 1,1′-thiocarbonyldiimidazole (484 mg, 90%, 2.44 mmol). After 4.5 h, the THF was removedin vacuo, and the residual oil was purified on silica gel (20 to 30% EtOAc-hexanes) to give 475 mg (80%) of isothiocyanate OM188 as a white solid.

N-(4-(3-Phenylureido)phenyl)benzenesulfonamide (CF252).

A mixture ofN-(4-aminophenyl)benzenesulfonamide (80 mg, 0.322 mmol), phenylisocyanate (38.4 mg, 35 µl, 1 equivalent), and triethylamine (NEt₃, 65.2 mg, 90 µl, 2 equivalents) in THF (2 ml) was heated at 60°C. After 12 h, all volatiles were removedin vacuo, and the residue was purified by preparative TLC (EtOAc-hexanes, 1:1, 2 elutions) to give CF252 (44 mg, 37%).¹H NMR (300 Mz, CD₃COCD₃) δ 8.78 (br s, 1 H), 8.12 (br s, 2 H), 7.77 to 7.42 (m, 2 H), 7.60 to 7.49 (m, 5 H), 7.42 (d,J = 9.0 Hz, 2 H), 7.26 (t,J = 7.2 Hz, 1 H), 7.09 (d,J = 8.7 Hz, 2 H), 6.99 (t,J = 7.2 Hz, 1 H);¹³C NMR (100 MHz, CD₃COCD₃) δ 152.42, 139.96, 139.87, 137.32, 132.57, 131.71, 128.88(2), 128.63(2), 127.03(2), 122.78(2), 122.05, 119.06(2), 118.52(2).

N-Cyclohexyl-4-(3-phenylureido)benzenesulfonamide (CF283).

Cyclohexylamine (14 mg, 0.142 mmol) and Et₃N (22 µl, 0.155 mmol) were added to a 0°C solution of 4-(3-phenylureido)benzene-1-sulfonyl chloride (40 mg, 0.129 mmol) in dry dichloromethane (2 ml). After being stirred overnight, the mixture was poured into water (10 ml) and extracted with EtOAc (5 ml; three times). The combined organic extracts were washed with brine (5 ml), dried, and concentrated. The residue was purified by PTLC (polar thin-layer chromatography) (CH₂Cl₂-MeOH, 95:5) to give CF283 (40 mg, 83% yield) as a white solid.¹H NMR (300 MHz, CD₃COCD₃) δ 8.66 (br s, 1 H), 8.39 (s, 1 H), 7.79 to 7.69 (m, 4 H), 7.57 to 7.54 (m, 2 H), 7.31 to 7.22 (m, 2 H), 7.03 to 6.98 (m, 1 H), 6.28 (d,J = 7.5 Hz, 1 H), 3.16 to 3.00 (m, 1 H), 1.73 to 1.63 (m, 4 H), 1.29 to 1.14 (m, 6 H);¹³C NMR (100 MHz, CD₃COCD₃) δ 152.12, 143.54, 139.54, 135.27, 128.69(2), 127.79(2), 122.42, 118.72(2), 117.67(2), 52.45, 33.62(2), 25.10(2), 24.66.

4-(3-(2-Methoxyphenyl)ureido)-N-phenylbenzenesulfonamide (CF287).

4-Amino-N-phenylbenzenesulfonamide (248 mg, 1.0 mmol) was added to a solution of triphosgene (312 mg, 1.05 mmol) in dry dioxane (4 ml) at room temperature. The reaction mixture was heated at 70°C for 4 h and then concentratedin vacuo to give crude 4-isocyanato-N-phenylbenzenesulfonamide, which was used without further purification.

2-Methoxyaniline (22 mg, 1 equivalent) was added to a solution of 4-isocyanato-N-phenylbenzenesulfonamide in dry THF (1 ml, 50 mg/1 ml). The mixture was heated at 80°C overnight and then concentratedin vacuo. The residue was purified by PTLC (CH₂Cl₂-MeOH, 9:1) to give CF287 (70 mg, 93% yield) as a white solid.¹H NMR (300 MHz, CD₃COCD₃) δ 9.26 (br s, 1 H), 8.89 (br, 1 H), 8.16 to 8.13 (m, 1 H), 8.07 (br, 1 H), 7.69 to 7.65 (m, 4 H), 7.49 to 7.46 (m, 1 H), 7.28 to 7.23 (m, 5 H), 7.08 to 7.03 (m, 2 H), 2.73 (s, 3 H);¹³C NMR (100 MHz, CD₃COCD₃) δ 151.83, 147.93, 144.24, 138.12, 132.39, 129.01(2), 128.57, 128.29(2), 124.22, 122.27, 120.63(2), 120.61, 118.61, 117.42(2), 110.30, 55.29.

4-(3-(1,3,4-Thiadiazol-2-yl)ureido)-N-phenylbenzenesulfonamide (CF290).

Following the above-given procedure, an equimolar mixture of 4-isocyanato-N-phenylbenzenesulfonamide and 1,3,4-thiadiazol-2-amine in dry THF (1 ml, 50 mg/1 ml) was heated at 80°C overnight and then concentratedin vacuo. Purification of the residue via PTLC gave CF290 (90%) as a white solid.¹H NMR (300 MHz, CD₃SOCD₃) δ 10.2 (s, 1 H), 9.58 (br s, 2 H), 9.04 (br s, 1 H), 7.69 to 7.60 (m, 4 H), 7.23 to 7.18 (m, 2 H), 7.08 to 6.97 (m, 3 H);¹³C NMR (100 MHz, CD₃SOCD₃) δ 160.33, 152.29, 148.03, 142.70, 137.83, 132.80, 129.13(2), 128.12(2), 124.00, 120.09(2), 118.16(2).

Synthesis ofN-(2-hydroxyethyl)-4-(3-phenylthioureido)benzenesulfonamide (CF355).

A mixture ofN-(2-hydroxyethyl)-4-nitrobenzenesulfonamide (530 mg, 2.15 mmol) and 5% WT Pd/C (53 mg) in anhydrous tetrahydrofuran (13 ml) was stirred under hydrogen (1 atm) for 16 h and then diluted with EtOAc (15 ml) and filtered through Celite. The filtrate was concentratedin vacuo and further dried under high vacuum to give 4-amino-N-(2-hydroxyethyl)benzenesulfonamide as a light-brown solid (459 mg, 98%).

Phenylisothiocyanate (180 µl, 1 equivalent) was added to a solution of the above-described crude amine (200 mg, 0.925 mmol) in anhydrous acetonitrile (9 ml). The reaction mixture was stirred at 70°C for 24 h and then concentrated and purified via PTLC (80% EtOAc-hexanes, 2 elutions) to give CF355 (22 mg, 7%) as a white solid (mp of 129.2°C).¹H NMR (400 MHz, CD₃OD) δ 7.83 to 7.78 (m, 2 H), 7.75 to 7.70 (m, 2 H), 7.47 to 7.42 (m, 2 H), 7.40 to 7.34 (m, 2 H), 7.24 to 7.17 (m, 1 H), 3.55 (t,J = 6.0 Hz, 2 H), 2.96 (t,J = 6.0 Hz, 2 H);¹³C NMR (100 MHz, CD₃OD) δ 180.40, 143.37, 138.63, 135.65, 128.52(2), 127.21(2), 125.36, 124.09(2), 122.95(2), 60.40, 44.79.

Quantitative real-time PCR.

Strain information is available inTable S1 in the supplemental material. Overnight cultures were diluted 1:100 and grown aerobically in low-glucose Dulbecco’s modified Eagle medium (DMEM) (Gibco) for EHEC or in N-minimal medium forS. Typhimurium with 5 or 500 nM drug, or with an equivalent volume of dimethyl sulfoxide (DMSO), at 250 rpm to late exponential growth phase (optical density at 600 nm [OD₆₀₀] = 1.0). RNA was extracted from three biological samples using a RiboPure bacterial RNA isolation kit (Ambion) by following the manufacturer’s guidelines. The primers used in the real-time assays were designed using Primer Express v 1.5 (Applied Biosystems) (seeTable S2) and were validated for amplification efficiency and template specificity. Quantitative real-time PCR (qRT-PCR) was performed as previously described (17) in a one-step reaction using an ABI 7500 sequence detection system (Applied Biosystems). Data were collected using the ABI Sequence Detection 1.2 software (Applied Biosystems).

All data were normalized to anrpoA (RNA polymerase subunit A) endogenous control and analyzed using the comparative critical threshold (C_T) method. Virulence gene expression was presented as fold changes over the expression level of WT EHEC orS. Typhimurium grown in the absence of drug. Error bars indicate the standard deviations of the fold change values. The Student unpairedt test was used to determine statistical significance.

Fluorescein actin staining.

Fluorescein actin staining (FAS) assays were performed as described by Knutton et al. (33). Briefly, HeLa cells were grown on coverslips in 12-well culture plates with DMEM supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin-gentamicin (PSG) antibiotic mix at 37°C and 5% CO₂, overnight to 80% confluence. The wells were washed with phosphate-buffered saline (PBS) and replaced with low-glucose DMEM supplemented with 10% FBS. Bacterial cultures were grown statically overnight in LB in the presence of drug or with an equivalent volume of DMSO at 37°C. Overnight bacterial cultures were diluted 1:100 to infect confluent monolayers of HeLa cells for 6 h at 37°C and 5% CO₂ in the presence or absence of drug. After a 6-h infection, the coverslips were washed, fixed, permeabilized, and then treated with fluorescein isothiocyanate (FITC)-labeled phalloidin to visualize actin accumulation and propidium iodide to visualize host nuclei and bacteria. The coverslips were mounted on slides and visualized with a Zeiss Axiovert microscope. To determine the percentage of infected cells, 5 to 7 fields from three separate coverslips (triplicate) were counted and calculated as follows: (cells with pedestals/total cells) × 100. Statistical analyses were performed using the Student unpairedt test.

Mouse survival experiments.

ForS. Typhimurium infections, mice (BALB/c, 7 to 9 weeks old, female) were either treated orally with LED209 (20 mg/kg in 5% DMSO, 23% polyethylene glycol [PEG] 400, 70% sodium bicarbonate [pH 9], 2% Tween 80) at 30 min preinfection only or at 3 h before, at the time of, and 3, 6, 9, and 12 h after infection and then intraperitoneally infected with 10⁶ CFU ofS. Typhimurium strain SL1344. We used 10 mice per treatment, and these experiments were repeated twice to ensure reproducibility. Mice were returned to their cages and monitored daily for signs of morbidity (anorexia, rapid and shallow breathing, scruffy fur, decreased muscle tone, and lethargy) and lethality up to 10 days. The animals that survived were humanely euthanized by CO₂ asphyxiation.

ForF. tularensis infections, mice (C3H/HeN, female) were either treated orally with LED209 (20 mg/kg in 5% DMSO, 23% PEG 400, 70% sodium bicarbonate [pH 9], 2% Tween 80) at 1, 3, 6, 9, or 24 h prior to infection and then intranasally infected with 30 CFU ofF. tularensis strain SCHU S4 or intranasally infected with 30 CFU ofF. tularensis strain SCHU S4 and then treated with LED209 at 1, 3, 6, 9, or 24 h postinfection. All animal work performed with the SCHU S4 strain was conducted in a federally licensed small animal containment level 3. Mice were monitored daily for signs of morbidity and death. At 12 days postinfection, the animals that survived were euthanized by CO₂ asphyxiation.

Macrophage infection.

J774 murine macrophages were infected with opsonizedS. Typhimurium with normal mouse serum at 37°C for 15 min and washed. These macrophages were infected using a multiplicity of infection (MOI) of 100:1 for 30-min bacterium-cell interaction at 37°C and 5% CO₂. These cells were treated with 40 µg/ml of gentamicin for 1 h to kill extracellular bacteria and lysed with 1% Triton-X. Bacteria were diluted and plated in LB plates for CFU determination (34–36). LED209 (5 nM in DMSO) or Om188 (500 nM, 5 nM, or 5 pM) was added to DMEM during the entire incubation.

Biofilm formation inhibition upon LED209 treatment.

Biofilm formation was assessed in experiments as previously described (37) and stained by the crystal violet method. All experiments were performed in triplicate using LB broth. Overnight bacterial cultures grown under static conditions were inoculated into fresh medium in a 1:100 dilution on 96-well cell culture polystyrene plates (Falcon). These plates were incubated at 37°C in a CO₂ atmosphere for 24 h. These conditions were assayed in the absence and presence of 5 nM of LED209 (DMSO), and DMSO in the same quantity was employed under all conditions. All these assays were repeated three times to ensure reproducibility.

Clinical isolates.

Urine and bladder biopsy samples were isolated from clinically relevant bacterial strains obtained from patients at the UTSW hospitals and clinics under the supervision of Philippe Zimmern. Samples were first enriched in LB broth and isolated in MacConkey agar to differentiate Gram-negative bacteria, as well as lactose fermentation, and classified via API 20E kits, according to the manufacturer’s directions (bioMérieux, France).

Construction ofC. rodentium mutants.

Isogenic nonpolarqseC,kdpE, andqseBC mutants were constructed inC. rodentium ICC168 using λ-red mutagenesis as previously described (2).

Mouse infection experiments.

For mouse survival experiment, 10 3.5-week-old male C3H/HeJ mice per group were infected by oral gavage with either 1 × 10⁹ cells of wild-typeC. rodentium ICC168 or the ΔkdpE, ΔqseC, or ΔqseBC mutant; five 3.5-week-old male C3H/HeJ mice were inoculated by oral gavage with 100 µl of PBS. Mouse survival in each group was assessed over the course of 14 days.

For bacterial load and colon weight measurement, as well as for colon pathology, 10 3.5-week-old male C3H/HeJ mice per group were infected with 1 × 10⁹ cells of wild-typeC. rodentium ICC168 or the ΔkdpE, ΔqseC, or ΔqseBC mutant. The infected mice were sacrificed on day 6 postinfection, and their colons were taken. For bacterial load measurement, colons of 5 mice per group were homogenized, and the homogenates were serially diluted and plated on LB agar plates containing 100 µg/ml of nalidixic acid.

Pharmacokinetic analysis.

Female CD-1 mice (6 to 7 weeks old) were purchased from Charles River Laboratories. Animals were allowed to acclimate to the animal facility at UT Southwestern Medical Center for approximately 1 week prior to use, and all animal work was approved by the Institutional Animal Care and Use Committee at UT Southwestern. Mice were treated with 5, 20, 80, or 160 mg/kg LED209 given by oral gavage in 10% DMSO, 10% Cremophor EL, 30% PEG 400, and 50% carbonate buffer (pH 10) either a single time or daily for 14 days. At the indicated time points on either day 1 (single dose) or day 14 (chronic dose), animals were given an inhalation overdose of CO₂, and blood was harvested by cardiac puncture with a sodium citrate-coated needle and syringe containing 50 µl of sodium citrate. Plasma was isolated by centrifugation of the blood for 10 min at 9,300 ×g and frozen at −80°C until analysis. Brains were isolated in an experiment in which animals were dosed at 20 mg/kg, rinsed briefly with PBS to remove surface-adhering blood, weighed, and snap-frozen in liquid nitrogen. Brain homogenates were prepared by homogenizing the brain tissues in a 3-fold volume of PBS (volume of PBS added in milliliters is 3 times the weight of the brain in grams). Total brain homogenate volume was estimated as volume of PBS added plus volume of brain in milliliters. One hundred microliters of plasma or brain homogenate was mixed with 200 μl of acetonitrile containing 0.15% formic acid and 300 ng/ml tolbutamide as an internal standard (IS). The samples were vortexed for 15 s, incubated at room temperature for 10 min, and spun twice at 16,100 ×g for 5 minutes at 4°C in a standard microcentrifuge to pellet precipitated proteins. Supernatant was collected and samples were analyzed on an ABSciex 4000-Qtrap liquid chromatography-tandem mass spectrometer (LC-MS/MS) in positive MRM mode. The MRM transition monitored for LED209 was 384.0 to 94.1 and 271.2 to 91.2 for the tolbutamide internal standard. A Shimadzu Prominence LC system was used with an Agilent ZORBAX Eclipse XDB-C18 column (5 micron packing, 4.6 × 50 mm). Samples were injected and compound eluted using the following chromatography conditions: buffer A, water and 0.1% formic acid; buffer B, MeOH and 0.1% formic acid; flow rate of 1.5 ml/min; gradient conditions of 0 to 1 min 5% B; 1- to 1.5-min gradient to 95% B; 1.5 to 2.5 min 100% B, 2.5- to 2.6-min gradient to 5% B; 2.6 to 3.5 min 5% B. A standard curve was prepared by using blank murine plasma (Bioreclamation, Westbury, NY) spiked with various concentrations of LED209. The lower limit of detection was set at a signal-to-noise ratio of 3:1 when spiked samples and blank plasma were compared. Standard curves were constructed by plotting the analyte-to-internal standard ratio versus the concentration of LED209 in each sample and a quadratic curve fit to the data using the Analyst software program. A value of three times above the signal obtained in the blank plasma or brain homogenate was designated the limit of detection (LOD). The limit of quantitation (LOQ) was defined as the lowest concentration at which back-calculation yielded a concentration within 20% of the theoretical value and above the LOD signal. The LOQ for LED209 was 1 ng/ml. In general, back-calculation of concentrations indicated the curve was accurate to within 15% over 4 logs from 1,000 to 1 ng/ml. Pharmacokinetic parameters for plasma were determined by using the noncompartmental analysis tool on the WinNonlin software package (Pharsight, Mountain View, CA).

In vivo toxicology studies.

Sixteen male and sixteen female CD-1 mice (6 to 7 weeks of age) were purchased from Charles River Laboratories. Animals were allowed to acclimate to the animal facility at UT Southwestern Medical Center for approximately 1 week prior to use, and all animal work was approved by the Institutional Animal Care and Use Committee at UT Southwestern. Animals were placed into one of four groups (vehicle only, 5 mg/kg, 20 mg/kg, or 80 mg/kg) in groups of 4 male and 4 female mice. Animals were dosed daily in the morning by oral gavage with LED209 formulated as 10% DMSO, 10% Cremorphor EL, 30% PEG 400, and 50% carbonate buffer (pH 10) for 14 days. Mice were weighed daily at the same time each day. On day 15, the animals were sacrificed by inhalation overdose of CO₂, blood was collected by cardiac puncture, and organs were isolated and weighed. Whole blood, collected using K₂-EDTA as the anticoagulant, was analyzed by the UT Southwestern Animal Resource Center for complete blood counts. Serum, isolated from whole blood collected in the absence of an anticoagulant, was evaluated by the UTSW Mouse Metabolic Phenotyping Core using the Vitros 250 chemistry analyzer for liver and kidney function. Additionally, a separate sample of serum was shipped to ANI Lytics (Gaithersburg, MD) for measurement of thyroxin (T4) levels as a measure of thyroid function.

hERG radiolabeled ligand binding assay.

LED209 was tested at 10 mM in duplicate in a radiolabeled ligand binding assay by MDS Pharma Services (now Eurofins Panlabs, Taipei, Taiwan) for its ability to block the interaction of 1.5 nM astemizole to the hERG channel expressed recombinantly in HEK293 cells. Minimal inhibition (4%) of binding was observed.

• 1.Rasko DA, Sperandio V.2010. Anti-virulence strategies to combat bacteria-mediated disease. Nat. Rev. Drug Discov.9:117–128. 10.1038/nrd3013. [DOI] [PubMed] [Google Scholar]
• 2.Petri WA, Jr, Miller M, Binder HJ, Levine MM, Dillingham R, Guerrant RL.2008. Enteric infections, diarrhea, and their impact on function and development. J. Clin. Invest.118:1277–1290. 10.1172/JCI34005. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 3.Kostakioti M, Hadjifrangiskou M, Cusumano CK, Hannan TJ, Janetka JW, Hultgren SJ.2012. Distinguishing the contribution of type 1 pili from that of other QseB-misregulated factors when QseC is absent during urinary tract infection. Infect. Immun.80:2826–2834. 10.1128/IAI.00283-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 4.Rasko DA, Moreira CG, Li de R, Reading NC, Ritchie JM, Waldor MK, Williams N, Taussig R, Wei S, Roth M, Hughes DT, Huntley JF, Fina MW, Falck JR, Sperandio V.2008. Targeting QseC signaling and virulence for antibiotic development. Science321:1078–1080. 10.1126/science.1160354. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 5.Moreira CG, Weinshenker D, Sperandio V.2010. QseC mediatesSalmonella enterica serovar Typhimurium virulencein vitro andin vivo. Infect. Immun.78:914–926. 10.1128/IAI.01038-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 6.Kostakioti M, Hadjifrangiskou M, Pinkner JS, Hultgren SJ.2009. QseC-mediated dephosphorylation of QseB is required for expression of genes associated with virulence in uropathogenicEscherichia coli. Mol. Microbiol.73:1020–1031. 10.1111/j.1365-2958.2009.06826.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 7.Weiss DS, Brotcke A, Henry T, Margolis JJ, Chan K, Monack DM.2007.In vivo negative selection screen identifies genes required forFrancisella virulence. Proc. Natl. Acad. Sci. U. S. A.104:6037–6042. 10.1073/pnas.0609675104. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 8.Moreira CG, Sperandio V.2012. Interplay between the QseC and QseE bacterial adrenergic sensor kinases inSalmonella enterica serovar Typhimurium pathogenesis. Infect. Immun.80:4344–4353. 10.1128/IAI.00803-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 9.Unal CM, Singh B, Fleury C, Singh K, Chávez de Paz L, Svensäter G, Riesbeck K.2012. QseC controls biofilm formation of non-typeableHaemophilus influenzae in addition to an AI-2-dependent mechanism. Int. J. Med. Microbiol.302:261–269. 10.1016/j.ijmm.2012.07.013. [DOI] [PubMed] [Google Scholar]
• 10.Khajanchi BK, Kozlova EV, Sha J, Popov VL, Chopra AK.2012. The two-component QseBC signalling system regulatesin vitro andin vivo virulence ofAeromonas hydrophila. Microbiology158:259–271. 10.1099/mic.0.051805-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 11.Kostakioti M, Hadjifrangiskou M, Cusumano CK, Hannan TJ, Janetka JW, Hultgren SJ.2012. Distinguishing the contribution of type 1 pili from that of other QseB-misregulated factors when QseC is absent during urinary tract infection. Infect. Immun.80:2826–2834. 10.1128/IAI.00283-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 12.Hadjifrangiskou M, Kostakioti M, Chen SL, Henderson JP, Greene SE, Hultgren SJ.2011. A central metabolic circuit controlled by QseC in pathogenicEscherichia coli. Mol. Microbiol.80:1516–1529. 10.1111/j.1365-2958.2011.07660.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 13.Novak EA, Shao H, Daep CA, Demuth DR.2010. Autoinducer-2 and QseC control biofilm formation andin vivo virulence ofAggregatibacter actinomycetemcomitans. Infect. Immun.78:2919–2926. 10.1128/IAI.01376-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 14.Mokrievich AN, Kondakova AN, Valade E, Platonov ME, Vakhrameeva GM, Shaikhutdinova RZ, Mironova RI, Blaha D, Bakhteeva IV, Titareva GM, Kravchenko TB, Kombarova TI, Vidal D, Pavlov VM, Lindner B, Dyatlov IA, Knirel YA.2010. Biological properties and structure of the lipopolysaccharide of a vaccine strain ofFrancisella tularensis generated by inactivation of a quorum sensing system geneqseC. Biochemistry75:443–451. 10.1134/S0006297910040073. [DOI] [PubMed] [Google Scholar]
• 15.Wang X, Wang Q, Yang M, Xiao J, Liu Q, Wu H.2011. QseBC controls flagellar motility, fimbrial hemagglutination and intracellular virulence in fish pathogenEdwardsiella tarda. Fish Shellfish Immunol.30:944–953. 10.1016/j.fsi.2011.01.019. [DOI] [PubMed] [Google Scholar]
• 16.Clarke MB, Hughes DT, Zhu C, Boedeker EC, Sperandio V.2006. The QseC sensor kinase: a bacterial adrenergic receptor. Proc. Natl. Acad. Sci. U. S. A.103:10420–10425. 10.1073/pnas.0604343103. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 17.Hughes DT, Clarke MB, Yamamoto K, Rasko DA, Sperandio V.2009. The QseC adrenergic signaling cascade in enterohemorrhagicE. coli (EHEC). PLoS Pathog.5:e1000553. 10.1371/journal.ppat.1000553. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 18.Clarke MB, Sperandio V.2005. Transcriptional regulation offlhDC by QseBC and sigma (FliA) in enterohaemorrhagicEscherichia coli. Mol. Microbiol.57:1734–1749. 10.1111/j.1365-2958.2005.04792.x. [DOI] [PubMed] [Google Scholar]
• 19.Njoroge JW, Nguyen Y, Curtis MM, Moreira CG, Sperandio V.2012. Virulence meets metabolism:cra andkdpE gene regulation in enterohemorrhagicEscherichia coli. mBio3(5):e00280-12. 10.1128/mBio.00280-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 20.Njoroge J, Sperandio V.2012. EnterohemorrhagicEscherichia coli virulence regulation by two bacterial adrenergic kinases, QseC and QseE. Infect. Immun.80:688–703. 10.1128/IAI.05921-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 21.Sperandio V, Torres AG, Kaper JB.2002. Quorum sensingEscherichia coli regulators B and C (QseBC): a novel two-component regulatory system involved in the regulation of flagella and motility by quorum sensing inE. coli. Mol. Microbiol.43:809–821. 10.1046/j.1365-2958.2002.02803.x. [DOI] [PubMed] [Google Scholar]
• 22.Clarke MB, Sperandio V.2005. Transcriptional autoregulation by quorum sensingEscherichia coli regulators B and C (QseBC) in enterohaemorrhagicE. coli (EHEC). Mol. Microbiol.58:441–455. 10.1111/j.1365-2958.2005.04819.x. [DOI] [PubMed] [Google Scholar]
• 23.Reading NC, Torres AG, Kendall MM, Hughes DT, Yamamoto K, Sperandio V.2007. A novel two-component signaling system that activates transcription of an enterohemorrhagicEscherichia coli effector involved in remodeling of host actin. J. Bacteriol.189:2468–2476. 10.1128/JB.01848-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 24.Rasko DA, Moreira CG, Li de R, Reading NC, Ritchie JM, Waldor MK, Williams N, Taussig R, Wei S, Roth M, Hughes DT, Huntley JF, Fina MW, Falck JR, Sperandio V.2008. Targeting QseC signaling and virulence for antibiotic development. Science321:1078–1080. 10.1126/science.1160354. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 25.Reading NC, Rasko DA, Torres AG, Sperandio V.2009. The two-component system QseEF and the membrane protein QseG link adrenergic and stress sensing to bacterial pathogenesis. Proc. Natl. Acad. Sci. U. S. A.106:5889–5894. 10.1073/pnas.0811409106. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 26.Reading NC, Rasko D, Torres AG, Sperandio V.2010. A transcriptome study of the QseEF two-component system and the QseG membrane protein in enterohaemorrhagicEscherichia coli O157:H7. Microbiology156:1167–1175. 10.1099/mic.0.033027-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 27.Pacheco AR, Curtis MM, Ritchie JM, Munera D, Waldor MK, Moreira CG, Sperandio V.2012. Fucose sensing regulates bacterial intestinal colonization. Nature492:113–117. 10.1038/nature11623. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 28.Kendall MM, Gruber CC, Parker CT, Sperandio V.2012. Ethanolamine controls expression of genes encoding components involved in interkingdom signaling and virulence in enterohemorrhagicEscherichia coli O157:H7. mBio3(3):e00050-12.   10.1128/mBio.00050-12.  [DOI] [PMC free article] [PubMed] [Google Scholar]
• 29.Gruber CC, Sperandio V.2014. Posttranscriptional control of microbe-induced rearrangement of host cell actin. mBio5(1):e01025-13. 10.1128/mBio.01025-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 30.Kostakioti M, Hadjifrangiskou M, Pinkner JS, Hultgren SJ.2009. QseC-mediated dephosphorylation of QseB is required for expression of genes associated with virulence in uropathogenicEscherichia coli. Mol. Microbiol.73:1020–1031. 10.1111/j.1365-2958.2009.06826.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 31.Moreira CG, Weinshenker D, Sperandio V.2010. QseC mediatesSalmonella enterica serovar Typhimurium virulencein vitro andin vivo. Infect. Immun.78:914–926. 10.1128/IAI.01038-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 32.Rasko DA, Webster DR, Sahl JW, Bashir A, Boisen N, Scheutz F, Paxinos EE, Sebra R, Chin CS, Iliopoulos D, Klammer A, Peluso P, Lee L, Kislyuk AO, Bullard J, Kasarskis A, Wang S, Eid J, Rank D, Redman JC, Steyert SR, Frimodt-Møller J, Struve C, Petersen AM, Krogfelt KA, Nataro JP, Schadt EE, Waldor MK.2011. Origins of theE. coli strain causing an outbreak of hemolytic-uremic syndrome in Germany. N. Engl. J. Med.365:709–717. 10.1056/NEJMoa1106920. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 33.Knutton S, Baldwin T, Williams PH, McNeish AS.1989. Actin accumulation at sites of bacterial adhesion to tissue culture cells: basis of a new diagnostic test for enteropathogenic and enterohemorrhagicEscherichia coli. Infect. Immun.57:1290–1298. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 34.Detweiler CS, Monack DM, Brodsky IE, Mathew H, Falkow S.2003.virK,somA andrcsC are important for systemicSalmonella enterica serovar Typhimurium infection and cationic peptide resistance. Mol. Microbiol.48:385–400. 10.1046/j.1365-2958.2003.03455.x. [DOI] [PubMed] [Google Scholar]
• 35.Fierer J, Eckmann L, Fang F, Pfeifer C, Finlay BB, Guiney D.1993. Expression of the salmonella virulence plasmid genespvB in cultured macrophages and nonphagocytic cells. Infect. Immun.61:5231–5236. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 36.Pfeifer CG, Marcus SL, Steele-Mortimer O, Knodler LA, Finlay BB.1999.Salmonella Typhimurium virulence genes are induced upon bacterial invasion into phagocytic and nonphagocytic cells. Infect. Immun.67:5690–5698. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 37.Moreira CG, Palmer K, Whiteley M, Sircili MP, Trabulsi LR, Castro AF, Sperandio V.2006. Bundle-forming pili and EspA are involved in biofilm formation by enteropathogenicEscherichia coli. J. Bacteriol.188:3952–3961. 10.1128/JB.00177-06. [DOI] [PMC free article] [PubMed] [Google Scholar]

Supplementary Materials