Probe Project Purpose

Uropathogenic Escherichia coli (UPEC) is the leading cause of community-acquired urinary tract infections (UTIs). Over 100 million UTIs occur annually throughout the world, including more than 7 million cases in U.S. adolescents and adults (14). UTIs in younger children are associated with greater risk of morbidity and mortality than in older children and adults. Antimicrobial resistance among UPEC is on the rise (10, 15–18), driving efforts to discover vulnerable targets in the molecular pathogenesis of infection.

During UTI, UPEC lives in intracellular and extracellular locales. UPEC adheres to the apical bladder epithelium and invades into it (19–21). Within the bladder epithelium, UPEC typically reproduces in a biofilm-like state called intracellular bacterial communities (IBC; (5)). After maturation of IBCs, UPEC disperses away from the IBC and exits the infected cells. Extracellular UPEC must then re-adhere, initiating the invasion and intracellular reproduction phases again. Past studies have revealed bacteria encased in the IBC within a complex matrix of fibrous protein assemblies and polysaccharides (5). Our prior studies have also shown that disruption of the IBC pathway aborts experimental UTI, highlighting the importance of this intracellular lifecycle (22–25). A detailed study of urine samples from women with acute UTI demonstrated IBC in shed bladder epithelial cell, showing that the pathway is conserved in humans (26).

Investigators have found that bacterial encapsulation is an important UPEC virulence factor (4–6), and experiments show that the K capsule contributes to multiple aspects of pathogenesis, including IBC formation. K capsules, also called K antigens, are enveloping structures composed of high-molecular-weight polysaccharides. Among UPEC, the K antigens K1, K2, K5, K30, and K92 are thought to be most prevalent (27). Capsules are well-established virulence factors for a variety of pathogens that are thought to protect the cell from opsonophagocytosis and complement-mediated killing (reviewed in (28, 29)). While they did not study the effects of K antigen from UPEC, Llobet et al. recently demonstrated that the highly acidic polysaccharide capsules of diverse organisms including Klebsiella pneumoniae, Pseudomonas aeruginosa, and Streptococcus pneumoniae interact strongly with APs, acting as “sponges” to sequester and neutralize the APs (30).

Of the different K types, the Group 2 and Group 3 capsules are most prevalent among UPEC isolates, with K1 and K5 being leading types. Although the capsules have different compositions, they are regulated, synthesized, assembled, and exported by functionally homologous factors, leading us to hypothesize that we can develop small molecular inhibitors of K-type encapsulation that target the most medically important K capsule types. Furthermore, the medically important infectious agents Campylobacter jejuni, Haemophilus influenzae, Neisseria meningitides, and Salmonella typhimurium among others, use homologous components in the biogenesis of their capsules. The K1 capsule type is closely associated with pathogenic isolates; not only is it the leading type in UTI, but it also accounts for much of the extra-urinary tract complications. Animals studies of E. coli K1 sepsis demonstrated that injection of a K1 capsule degrading enzyme abrogates infection (7). However, the enzyme treatment is immunogenic; accordingly, chemical inhibition may prove to be a superior approach.

There are currently no therapeutics that specifically inhibit the formation of any bacterial capsule, and this is a novel strategy for preventing or decreasing the prevalence of chronic or re-occurring urinary tract infections. New insights into the roles of K1 capsules in UPEC virulence during UTI make capsules an attractive target for therapeutic intervention. Antimicrobial resistance among UPEC is on the rise (11, 12, 15, 31, 32), and the discovery of novel small molecules that can act as probes or lead compounds for the investigation and treatment of UTI will add to the arsenal of compounds available for single or combination therapies.

Prior Art

Through the efforts of the Seed lab toward developing assay and screening techniques for inhibitors of bacterial capsule biogenesis, a library of 2,195 compounds obtained from the Developmental Therapeutics Program at the National Cancer Institute was tested. In the K1F phage lysis 96-well plate format assay, 35 (1.59%) of the compounds were found to have inhibitory activity, of which only nine compounds gave reproducible phage lysis inhibition activity in shaken tube format. These nine compounds were taken on to a secondary screening process from which two capsule biogenesis inhibitors emerged. Malachite green oxalate (NCS5550), a compound not known to inhibit capsule biogenesis, was found to produce metabolites with previously reported toxicities to mammalian systems and was thus discarded. The second inhibitor, 2-(4-phenylphenyl)-benzo[g]quinoline-4-carboxylic acid (NSC136469), or C7, was employed as a prototype small molecule inhibitor of capsule biogenesis since it inhibited K1F phage lysis of UPEC K1 strain UTI80 reproducibly in the tests following HTS. Furthermore, this inhibition was found to behave in a concentration-dependent manner with the inhibitory effect reaching saturation at approximately 25 μM C7, producing approximately 50% inhibition of K1F phage lysis of UPEC at 12.5–25 μM (33).

This team previously reported a probe ML317 that arose from the same high-throughput screen that led to the probe ML333 reported herein. In short, the probe ML317 was found to have an IC₅₀ of 1.89 μM in a bacterial viability assay in the presence of K1 phage, a TC₅₀ (toxicity) of 51.6 μM, a selectivity of 27, a T7 phage bacterial viability assay IC₅₀ of <0.39 μM, and to reduce the bacterial capsule-dependent orcinol stain to 21% of the control sample. Additional biological characterization for probe ML317 is provided in Section 4.1 of this report.

Intellectual Property Landscape for the Probe Chemotype

On September 30^th, 2012, a search was performed using SciFinder to explore the intellectual property landscape around the probe compound ML333 and analogues from the probe chemotype. A substructure search using the structure shown revealed one patent and zero publications related to the preparation and use of N-(pyridin-4-yl)benzo[d]thiazole-6-carboxamide derivatives for use as tryptase inhibitors. No publications or patents reporting the use of ML333 derivatives as antibacterial agents were found.

ML333

1. “Preparation of (hetero)arylmethlamines as tryptase inhibitors” Lively, Sarah Elizabeth; Waszkowycz, Bohdan; Harrison, Martin James; Clase, Juha Andrew; Naylor, Neil Jason PCT Int. Appl. (2001), WO 2001027096 A1 20010419.

An exact structure search on the probe ML333 using SciFinder revealed that the compound was commercially available, and no references to publications or patents were found. An exact structure search for some of the most active compounds from this chemotype (i.e., CID 53484233, CID 53484226, CID 53484228, and CID 53484225) revealed no publications or patents. Overall, these search results suggest that compounds derived from the ML333 chemotype could be claimed for use as anti-bacterial agents, specifically for the inhibition of bacterial capsule biogenesis for the treatment of bacterial infection.

Overall Assay Strategy

The ability of the T1 bacteriophage to bind to bacterial capsule, infect, and lyse E. coli UT189 was used as the primary assay to identify the effects of screened compounds on bacterial capsule formation. The phenotypic end-point assay measured the fluorescence generated by cellular processing of alamarBlue as a direct indicator of cell viability. A total of 335,740 compounds were screened using the primary assay. Following this, a concentration-dependent confirmatory assay (in a compound concentration range of 50 – 0.19 μM) was used in parallel with a eukaryotic cytotoxicity counterscreen (same concentration range) to determine compound hit IC₅₀s and CC₅₀s (and selective indices). Hits were further evaluated using three secondary assays for alternative measurement of compound-induced reduction in bacterial capsule formation (secondary assays 1 and 2) and specificity (secondary assays 3 and 4). Confirmed actives with a selectivity index (SI) >5 were further investigated and subjected to chemical optimization, followed by secondary assay evaluation. Secondary assays more closely characterized the ability of the compounds to reduce bacterial capsule formation and can be used to examine the mechanism of action of the compounds. The primary assay (to measure capsule reduction), counter assay (for general eukaryotic cell toxicity) and secondary assay (to measure reduction in capsule formation and specificity) were combined to allow a determination of probe efficacy, selectivity, and specificity.

A chemical probe for this project was defined as a small molecule that:

• Has an IC₅₀ of < 10 μM in the primary and confirmatory alamarBlue Screen of E. coli strain UT189 lysis (primary assay)
• Has a therapeutic index of > 5 relative to the cytotoxicity in the bladder cell line Hu5637 (counterscreen)
• Has an IC₅₀ of < 10 μM in the confirmatory bacterial growth screen of E. coli strain UT189 lysis (secondary assay 1)
• Has an IC₅₀ of < 10 μM in the T7 lysis inhibition assay, indicating the desired target specificity (secondary assay 2)
• Will yield low orcinol levels that are 50% of the levels for capsule export control strain (secondary assay 3)
• Will yield K5 phage sensitivity in E. coli strain DS17 (secondary assay 4).

2.1. Assays

2.2. Probe Chemical Characterization

Probe Chemical Structure and Properties

Figure 1Structure Verification and Purity

1H NMR, 13C NMR, RP HPLC/UV/HRMS Data

Proton and carbon NMR data for ML333/MLS004555969/SID 103147597/CID 18109210: Detailed analytical methods used and the associated instrumentation are described in section 2.3, entitled “Probe Preparation”, under general experimental and analytical details. The numerical experimental proton and carbon NMR data are presented below. The associated spectra are included in Appendix B.

Proton NMR Data for ML333/SID 103147597/CID 18109210:¹H NMR (400 MHz, DMSO) δ 10.76 (s, 1H), 9.60 (s, 1H), 8.12 (d, J = 1.2 Hz, 1H), 8.51 (dd, J = 1.2, 5.2 Hz, 2H), 8.25 (d, J = 8.8 Hz, 1H), 8.12 (dd, J = 1.6, 6.8 Hz, 1H), 7.81 (dd, J = 1.2, 3.6 Hz, 2H) ppm.

Carbon NMR Data for ML333/SID 103147597/CID 18109210:¹³C NMR (100 MHz, DMSO) δ 165.9, 159.6, 155.1, 150.3, 145.8, 133.7, 131.3, 125.9, 122.9, 113.9 ppm.

RP HPLC/UV/HRMS Data for ML333/SID 103147597/CID 18109210: Detailed analytical methods used and the associated instrumentation are described in section 2.3, entitled “Probe Preparation”, under general experimental and analytical details. Purity assessment by RP HPLC/UV/HRMS at 214 nm for SID 103147597 (CID 18109210) revealed purity of 100.0 % (retention time = 2.287 minutes). The experimental RP HPLC/UV/HRMS spectra are included in Appendix B. HRMS (m/z) calcd. for C₁₃H₉N₃OS [M+H⁺] 256.0539, found 256.0543.

Aqueous Stability

Aqueous stability for the probe was assessed using two solvent systems (100% aqueous PBS, and 50:50 aqueous PBS:acetonitrile). The probe stability was measured in aqueous PBS (no antioxidants or other protectants, DMSO concentration below 0.1%, room temperature) and the results are reported as circles in the graph in Figure 2. The probe stability was also measured in 50:50 aqueous PBS and acetonitrile and the results are reported as squares in the graph in Figure 2. Stability data in each case is depicted as the loss of compound with time over 48 hours with a minimum of six time points and the percent compound remaining after 48 hours. ML333 was found to be quite stable (>75% remaining) in both solvent systems under these experimental conditions over the 48 hour study (Figure 2).

Figure 2

Graph depicting the stability for ML333/KUC107756 after 48 hours in two separate solvent systems.

Thiol Stability

Compounds were dissolved at 10 μM in PBS at pH 7.4 (1% DMSO) and incubated at room temperature with either no thiol source (as a negative control), 50 μM glutathione (GSH), or 50 μM dithiothreitol (DTT). The mixtures were sampled every hour for eight hours and analyzed by RP HPLC/UV/HRMS. The analytical RP HPLCUV/HRMS system utilized for the analysis was a Waters Acquity system with UV-detection and mass-detection (Waters LCT Premier). The analytical method conditions included a Waters Acquity HSS T3 C18 column (2.1 × 50mm, 1.8um) and elution with a linear gradient of 1% water to 100% CH₃CN at 0.6 mL/min flow rate. Peaks on the 214 nm chromatographs were integrated using the Waters OpenLynx software. Absolute areas under the curve were compared at each time point to determine relative percent compound remaining. The masses of potential adducts were searched for in the final samples to determine if any detectable adduct formed. All samples were prepared in duplicate (Figure 3). Ethacrynic acid, a known Michael acceptor, was used as a positive control. If solubility of the compound was an issue at 10 μM, PBS was substituted with PBS with 50% acetonitrile (35).

Figure 3

Graph depicting stability of the probe to thiol.

ML333 was found to be completely stable to the presence of five times its concentration of both glutathione and DTT across all time points in our analysis (Figure 3). Statistically, there is no significant drop in the % remaining concentration of ML333 during the course of these experiments. This data is easily contrasted to the data for the ethacrynic acid positive control compound, which is shown to react rapidly with five times its concentration of glutathione and even more rapidly with DTT.

Synthesis Route

The probe compound ML333 and numerous analogues were synthesized using the reaction sequence shown in Figure 4.

Figure 4

Synthetic route for probe and analogue generation.

Submission of Probe and Five Analogues to the MLSMR

Samples of the probe and five analogues were prepared, analytically characterized, and shipped to the MLSMR (Table 1). The structures for the five supporting analogues are shown in Figure 5.

Table 1

Five probe analogues submitted to the NIH MLSMR, and their associated screening data.

Figure 5

Five analogues chosen to support probe ML333.

2.3. Probe Preparation

3.1. Summary of Screening Results

Primary assay: An end-point assay to measure the amount of K1 phage-induced bacterial cell lysis was employed to determine test compound effect on capsule biogenesis. K1 bacteriophage specifically binds the bacterial capsule during the initial stages of infection. Bacteria without capsule cannot be infected and lysed by the bacteriophage. The ability of the test compounds to inhibit the K1 capsule formation was measured by an increase in the fluorescence of alamarBlue, which correlated with the amount of intact bacterial cells compared to control reaction wells. Inhibition of phage-induced lysis indicated an active compound in the primary screen. A total of 338,740 compounds were screened at 100 μM in the primary screen. Inhibition of phage lysis was calculated relative to the mean of the bacterial (positive) control on each microtiter plate. Primary screen average Z values = 0.75; average signal to background (S/B) = 11; and average coefficient of variance = 4.6%. For calculation of S/B/Z value, and CV, the following formulae (in which SD stands for standard deviation) were used: S/B = mean signal/mean background; Z = 1 − 3SD of sample + 3SD of control mean of sample - mean of control; % CV = (SD mean signal/mean signal) × 100.

Confirmatory efficacy: The background cutoff (15% inhibition) for the primary screen was calculated using the mean of all compound results plus 3× SD. Compounds that inhibited more than 30% (1,767) were considered for evaluation by confirmatory dose response and cytotoxicity counter screening assays. Only 1,219 compounds were available for confirmatory assays. The confirmatory efficacy assay was performed as described for the primary screen except that each compound was tested at 10 concentration points starting from 300 μM and continuing to lower concentrations by serial 2-fold dilutions to 0.69 μM. Twenty-six compounds were confirmed as effective at inhibiting the formation of capsule at compound concentrations below 50 μM. Confirmatory screen average Z values = 0.79; average signal to background (S/B) = 32; and average coefficient of variance = 6.4%. IC₅₀ values were calculated using a 4 parameter Levenburg-Marquardt algorithm, with the maximum and minimum locked at 0 and 100 respectively.

Cytotoxicity assay: The cytotoxicity assay was performed as described in Appendix A. Hit cytotoxicity and the 50% toxic concentration (CC₅₀) were determined and compared to the IC₅₀ to calculate the selectivity index. Twenty-nine compounds were tested, and 11 were inactive (CC₅₀ > 50 μM).

Figure 6HTS to compound hit flowchart

3.2. Dose Response Curves for Probe

The primary assay (Summary AID: 488970) and counterscreen methods (AID: 493020); were used to measure both probe efficacy and cytotoxicity. The probe ML333 potency in the primary phase lysis inhibition assay was determined to be IC₅₀ = 1.04 ± 0.13 μM, and the CC₅₀ 239 ± 89 μM. The calculated selectivity was 230 (CC₅₀/IC₅₀). The ML333 concentration response curves for efficacy and cytotoxicity are graphed (Figure 7).

Figure 7

Efficacy and cytotoxicity analysis of ML333. The graph on the left (green line) shows the efficacy of the probe in the primary bacterial phage lysis inhibition assay (n = 8). The small gray inset magnifies the corresponding region in the larger graph. (more...)

3.3. Scaffold/Moiety Chemical Liabilities

The probe compound and analogues appears to be stable based on our observations from day-to-day handling related to their synthesis, analysis, dissolution-transfer, lyophilization, and storage. The probe compound and analogues do not contain moieties that are known or expected to be reactive. The probe was found to be stable in aqueous solution and in the presence of thiol (please see section 2.2).

3.4. SAR Tables

The high throughput screen of 338,740 compounds at 100 μM concentration in the primary screen (AID 488966) resulted in a number of validated compound hits. One of the compound hit chemotypes the team chose to explore further resided in a cluster of nine compounds, in which one, Table 2, entry 1, exhibited 53% inhibition at 100 μM concentration. Related compounds in this chemotype, Table 2, entries 2–9 were found to be inactive at 100 μM.

Table 2

HTS data for compounds in the benzothiazole chemotype.

Considering the limited SAR information available for this chemotype, for the initial SAR exploration, the team chose to focus on varying the benzothiazole, amide, and pyridyl sub-structural components of the chemotype, as shown schematically in Figure 8.

Figure 8

Regions of the compound hit identified for initial SAR exploration.

In particular, in a combinatorial fashion, the benzothiazole moiety was replaced using the quinolone, indole, and benzimidazole ring systems, while the 4-pyridyl moiety was replaced using p-methylphenyl, p-trifluoromethyl-phenyl, p-methoxyphenyl, 2-pyridyl, 3-pyridyl, o-fluorophenyl, m-fluorophenyl, p-fluorophenyl, p-chlorophenyl, p-bromophenyl, p-acetylphenyl, and furan-2-yl-methyl moieties. Each compound was screened to determine IC₅₀ values in the K1 phage, human bladder cell cytotoxicity, and T7 phage assays. To our surprise, only one of the 49 analogues synthesized (Table 3, entry 42) showed any hint of activity.

Table 3

First round of SAR study on synthesized analogues.

Since none of the more significant structural modifications studied in the first round of SAR exploration resulted in active compounds, the team decided that the second round of compounds for SAR study should consist of only slightly modified analogues (Table 4). Specifically, the pyridyl moiety was substituted to give 2-bromopyridyl, 2,6-dichloropyridyl, 2-methoxypyridyl, 2-chloropyridyl, 2-hydroxypyridyl, 2-fluoropyridyl, and 3-fluoropyridyl analogues. To explore the SAR around the amide bond, we included the N-methyl analogue (entry 1), the reverse amide (entry 7), as well as amide-cyclized oxazolo[4,5-b]pyridyl and oxazolo[5,4-c]pyridine analogues (entries 8 and 9). Each compound was screened to determine IC₅₀ values in the K1 phage, human bladder cell cytotoxicity, and T7 phage assays. Entries 2, 5, 6, and 12 of Table 4 all exhibited activity within the range of our desired probe criteria, showing that mono-halogenation and methoxy substitutions on the pyridyl moiety were tolerated. However, N-methylation of the amide bond, reversal of the amide bond, di-substitution on the pyridyl ring, hydroxyl substitution of the pyridyl ring or cyclization through the amide bond onto the pyridyl ring were not tolerated. Replacing the pyridyl group in the molecule with a phenyl group (entry 16) reduced activity significantly.

Table 4

Second round of SAR study on synthesized analogues.

The third round of SAR study focused on very specific modifications to the pyridyl portion of the hit compound (Table 5). Saturated, homologated, and dihomologated analogues were prepared. The parent carboxylic acid, benzothiazole-6-carboxylic acid, was also tested. Each compound was screened to determine IC₅₀ values in the K1 phage, human bladder cell cytotoxicity, and T7 phage assays. None of these analogues showed biological activity.

Table 5

Third round of SAR study on synthesized analogues.

For the final round of SAR study, the team explored combining the most promising substitutions discovered from our studies on the pyridyl ring with substitutions on the, as yet, unexplored 2-position of the benzothiazole ring (Table 6, entries 1–32). At this stage of the project, compounds were screened for activity in the T7 phage assay (and were not screened using the K1 phage assay) and for cytotoxicity against human bladder cell carcinoma 5637 cells. This final round of SAR resulted in some of the most potent and selective analogues prepared, to date, entries 13 and 17. In the case of entry 13, methyl-group substitution adjacent to the nitrogen atom of the pyridine ring and introduction of a Boc-protected-amine at the 2-position of the benzothiazole provided a compound with an IC₅₀ of 490 nM in the T7 phage assay and a selectivity index of nearly 200. In addition, methyl-group (entries 10 and 11) and amine-group (entry 23) substitution was shown to be tolerated at the 2-position of the benzothiazole.

Table 6

Fourth round of synthetic analogues.

3.5. Cellular Activity

ML333 was identified using a phenotypic bacterial cell-based assay that determined the amount of bacterial capsule formation in the presence of the probe compound. No biochemical assays were used for determination of compound biological activity. Additional cell-based assays were performed to determine probe specificity. Also, a counterassay was performed to determine eukaryotic cytoxicity using human bladder cell carcinoma 5637 cells, which are considered the physiologically-relevant target cell type. The compound was shown to have no effect on bacterial cell viability in the highest tested concentration (300 μM), and the eukaryotic 50% cytotoxic concentration (CC₅₀) was 239 ± 89 μM, which afforded a calculated selectivity index of 230. Although the primary and secondary assays were performed using bacterial cultures, it is recognized that probe will be used in in vitro eukaryotic and in vivo assay systems. Because of this, each synthesized probe and analog were routinely tested using the cytotoxicity counterassay.

3.6. Profiling Assays

In vitro pharmacokinetics profiling

The in vitro pharmacokinetic (PK) properties of the probe (ML333) were profiled using a standard panel of assays (Table 7) across which the probe displayed encouraging results.

Table 7

Summary of in vitro ADME properties of ML333.

4.1. Comparison to existing art and how the new probe is an improvement

The probe described herein is entirely novel with the only precedent being the structurally dissimilar molecule 2-(4-phenylphenyl)benzo[g]quinoline-4-carboxylic acid that we previously described in proof-of-concept studies (33) and the previous MLPCN probe compound ML317. The probe described herein is uniquely poised for development as an infection-specific prevention and treatment therapeutic that enhances natural immune clearance of an infection, namely UTI. The highly soluble probe abrogates capsule development in several K type UPEC as demonstrated through capsule-specific phage assays and orcinol biochemical tests. However, the probe has no effect on in vitro growth and viability of UPEC in the absence of immune factors. Unlike with many traditional antibiotics, exposure of UPEC to a range of concentrations of the probe has not resulted in the emergence of spontaneous resistance in the laboratory, consistent with the concept that this probe does not induce stress and adaptive changes that confer resistance. Furthermore, this probe would be expected to act upon traditional antibiotic resistant strains of E. coli.

As previously outlined, 2-(4-phenylphenyl)-benzo[g]quinoline-4-carboxylic acid C7, was used as the prototype small-molecule inhibitor of capsule biogenesis, and, along with the MLPCN probe ML317, serves as the only prior art for this probe development project. Relative to C7, the probe discovered during this MLPCN project demonstrates improved potency in the K1 phage assay, a better selectivity index, similar performance in the orcinol screen, and greater than an order-of-magnitude lower IC₅₀ value in the T7 phage assay (Figure 9). Furthermore, we have demonstrated the synthetic tractability for the ML333 chemotype, and prepared an array of highly active analogues, whereas C7 does not share either of positive these attributes. To further demonstrate the usefulness of ML333 as a probe, we have shown ML333 to have acceptable in vitro pharmacokinetic properties. Aqueous stability in 1× PBS at pH 7.4, aqueous stability in LB, both human and mouse hepatic microsomal stability, PAMPA permeability and both human and mouse stability were all found to be acceptable. In contrast, while C7 has not been screened in such in vitro PK assays, one might predict poor results based on its chemical structure and calculated physiochemical properties. For example, one would not imagine a compound such as C7 to have good solubility characteristics. Lastly, the many aryl rings of the C7 structure pose liabilities to oxidative metabolism that might render this compound less attractive compared to ML333. Compared to the previous MLPCN probe ML317, the current probe compound ML333, has a better selectivity index (BC5637 bladder cells TC₅₀/K1 phage IC₅₀), comparable aqueous solubility, permeability, and plasma stability (human and mouse), whereas the hepatic microsome stability is observed to be slightly less promising.

Figure 9

Structures and select properties for C7, ML317, and ML333.

4.2. Mechanism of Action Studies

Moving forward, genetic and biochemical approaches are being employed to identify the mechanism of action for the probe. Currently, overexpression of a whole genome open reading frame library of E. coli is being used to identify gene products that enhance or reduce susceptibility to the probe. A number of plasmids expressing open reading frames have been selected in the screen and the identity and function of gene products are being analyzed. The target of the probe may be within a signal transduction pathway to capsule regulation but distinct from the actual transcription factor affecting capsule expression. Cellular localization of probe-interacting factors and biochemical studies of probe-target interactions will be explored.

Recent studies have also employed chemical mutagenesis to derive strains resistant to the action of the probe. Whole genome sequencing using Illumina Hiseq technology has been completed on five independent isolates. An analysis of polymorphisms shared among independent resistant bacterial clones will be used to localize putative factors that interact with the probes. These may be transporters or the actual target of the probe. The genes from the mutant and wild type strains will be cloned and expressed in the isogenic bacterial backgrounds to ascertain their roles in sensitivity to the probe.

4.3. Planned Future Studies

The probe will be optimized further for potency, selectivity, physiochemical properties, and in vitro pharmacokinetics (using the K1 phage, T7 phage, human bladder cell cytotoxicity, aqueous solubility, aqueous stability, plasma stability, plasma protein binding, logD, cell permeability, and microsomal stability assays). Based on the current SAR story, substitution at the positions R¹, R², and R³ will be explored, along with the introduction of nitrogen atoms at the positions X, as shown in Figure 10.

Figure 10

Future SAR studies.

Subsequently, in vivo pharmacokinetic measurements will be performed in a preclinical murine model with additional optimization of the probe for bioavailability and renal excretion. After completion of these studies and refinements to the probe, the refined probe will be employed in prophylaxis and treatment trials using preclinical murine infection models of E. coli urinary tract infection and bloodstream infection.

Screening for Inhibitors of Bacterial Capsule Biogenesis

Purpose: The primary screening was conducted to screen the MLSMR 300K compound library for selection of 1,767 hit compounds.

Summary AID: 488966

Assigned AID: 488970

Counterscreen: Cytotoxicity screening for potential Inhibitors of Bacterial Capsule Biogenesis

Purpose: This cell-based assay measures the cytotoxicity of compounds in bladder carcinoma 5637 cells using luminescent cell viability assay readout.

Summary AID: 488970

Assigned AID: 493020, 504769, 504831, 588399

Assay Description:Cytotoxicity: Dose response testing (dose range = 300–0.58 μM) established the hit cytotoxicity data. Compounds were plated in 384-well microplates in a stacked dose response format using the same doses used in the efficacy dose response. Bladder carcinoma 5637 cells were added to the compounds, and 72 hr later cell viability was measured using CellTiter Glo (38). Hit cytotoxicity and the 50% toxic concentration (TC₅₀) was determined and compared to the IC50 to calculate the therapeutic index. Test compounds are serially diluted in a plate to plate matrix or stacked plate matrix. All 320 compounds in a source plate are diluted together resulting in a 10 point dose response dilution series. It is visualized as a serial dilution series proceeding vertically through a stack of plates with the high dose plate on top and the low dose plate on the bottom.

Control Drug: Hyamine was used as a positive cytotoxic control. All wells contained 0.5% DMSO.

Preparation of Bladder carcinoma 5637 cells: Cells are harvested and resuspended to 80,000 cells per ml in Complete DMEM/F12^®.

Endpoint Read: Following the three day incubation period, the assay plates were equilibrated to room temperature for 30 min and an equal volume (30 μL) of Cell Titer-Glo^® reagent (Promega Inc.) is added to each well using a WellMate™ (Matrix, Hudson, NH) and plates are incubated for an additional 10 min at room temperature. At the end of the incubation, luminescence is measured using a Perkin Elmer Envision™ multi-label reader (PerkinElmer, Wellesley, MA) with an integration time of 0.1 s.

Secondary Assay: Screening for Inhibitors of Bacterial Capsule Biogenesis E.coli strain UT189

Purpose: This confirmatory cell-based assay provides an alternative measurement of inhibitory activity on phage-induced lysis. It measures reduction in bacterial capsule formation using an absorbance readout at A₆₀₀ instead of the alamar Blue reagent.

Summary AID: 488970

Assigned AID: 504358, 504543, 504675, 504768, 588321, 588386, 588395

Assay Set-up: This secondary assay was conducted in the 96 well plate format. Bacterial cultures of E. coli K1 strain UTI89 (cystitis isolate) and isogenic capsule mutant strains (as controls for phage infection) were grown and prepared at the screening center immediately prior to use. Overnight starting cultures of UTI89 were grown at 37 °C, and diluted 1:100 into LB. Compounds were added to plates in a concentration dependent manner in the range of 100–0.39 μM, followed by addition of 100 μL of bacterial culture. Each concentration was tested in quadruplicate. 1% DMSO (final well concentration) was included. The plates were tape sealed and shaken vigorously for 1.5 hr. An initial OD₆₀₀ reading at the time of infection was measured to identify compounds that cause growth retardation or bacterial killing in the absence of phage. Next, K1F phage (5 μL) was added to all of the test wells. The plates were resealed and shaken vigorously at 37 °C, and measurements of OD₆₀₀ for phage-mediated lysis were taken after 3 hr.

Endpoint Read: The plates were read at ambient temperature from the bottom for absorbance at A₆₀₀ in an Envision plate reader (Perkin Elmer) and the degree of phage-mediated lysis was determined based on the absorbance.

Secondary Assay: Screening for Inhibitors of Bacterial Capsule Biogenesis - T7 Lysis Inhibition

Purpose: This secondary assay measures the compound mechanistic specificity for inhibition of bacterial capsule formation using a different bacterial phage (T7). In this assay, an increase in phage-induced lysis correlates to a decrease in capsule formation.

Summary AID: 488970

Assigned AID: 504349, 504538, 504676, 504767, 588322

Assay Description: T7 phage-mediated lysis assay: This assay determined if the mechanism of action of the compound is inhibition of phage infectivity or replication. The T7 phage has a nearly identical genome to K1F, without encoding an endosialidase. Its cycle of replication is similar to that of K1F as well. However, T7 entry into E. coli is inhibited by K capsules. This secondary assay was conducted in the 96 well plate format. Bacterial cultures of E. coli K1 strain UTI89 (cystitis isolate) and isogenic capsule mutant strains (as controls for phage infection) were grown and prepared immediately prior to use. Overnight starting cultures of UTI89 were grown at 37 °C, and diluted 1:100 into LB. Compounds were added in quadruplicate to plates in a concentration dependent manner in the range of 100–0.39 μM, followed by addition of 100 μL of bacterial culture. 1% DMSO (final well concentration) was included. The plates were tape sealed and shaken vigorously for 1.5 hr. An initial OD₆₀₀ reading was measured to identify compounds that cause growth retardation or bacterial killing in the absence of phage. Next, T7 phage (5 μL) was added to all of the test wells. The plates were resealed and shaken vigorously at 37 °C, and measurements of OD₆₀₀ for phage-mediated lysis were taken after 3 hr. True inhibitors of capsule yielded bacteria that were susceptible to T7 phage and lysed within 2 hr of the addition of phage. However, compounds inhibiting phage replication did not promote bacterial lysis. The positive control drug C7 (100 μM final well concentration) was used in this screen.

Endpoint Read: The plates were read at ambient temperature from the bottom for absorbance at A₆₀₀ in a BioTek Quantplate reader and the degree of phage-mediated lysis was determined based on the absorbance.

Secondary Assay: Orcinol Secondary Screening for Inhibitors of Bacterial Capsule Biogenesis

Purpose: This assay was performed only on the probe candidate. This is an end-point assay to measure the amount of K1 bacterial cell capsule formation in the presence of a test compound concentration range. The ability of the test compounds to inhibit the K1 capsule formation was measured by a reduction of absorbance from the complex formed by orcinol and the capsule polysaccharide. The biochemical determination of cell-surface associated capsule was performed using UTI89 or the delta Region I and Region II capsule mutant bacterial strains.

Summary AID: 488970

Assigned AID: 504733, 624060

Assay Description: The biochemical measurement of cell-surface associated capsule was performed. UTI89 or capsule mutants were grown in culture tubes with and without the test compound (50 μM). The cultures were centrifuged, and the cell pellets were washed in PBS and resuspended in Tris buffer, pH5. We have found that low pH releases surface polysaccharide without lysing the bacteria. Released polysaccharide was harvested by separation from whole bacteria by centrifugation followed by deproteination with phenol/chloroform and precipitation with ethanol. The precipitated material was subjected to acid hydrolysis (pH 2 at 80°C for 1 hr) and incubated with orcinol, which reacts with periodic intermediates to produce a violet color quantified at OD 570 (3). Inhibitors reducing or abrogating surface encapsulation yielded low orcinol levels similar to the capsule export and synthesis mutants (Region I and II). The positive control drug C7 (100 μM final well concentration) was used in this screen.

Endpoint Read: The samples were read at ambient temperature from the bottom for absorbance at A570 in a BioTek Quantplate reader and the degree of orcinol-reactive material was determined based on the absorbance compared to a wild-type encapsulated strain (UTI89) and a standard curve using purified sialic acid.

Secondary Assay: K5 Secondary Screening for Inhibitors of Bacterial Capsule Biogenesis

Purpose: This assay was performed only on the probe, and determined if compounds considered active in the T7 and orcinol secondary assays were able to inhibit also inhibit K5 capsule biogenesis.

Summary AID: 488970

Assigned AID: 624061

Assay Description: This assay was performed only on the probe candidate, and determined if compounds considered active in the T7 and orcinol secondary assays were able to also inhibit K5 capsule biogenesis. This assay was performed in a method identical to the T7 assay test, but a different bacterial test strain and bacteriophage are used. In this validation test, we used E. coli strain DS17, a pyelonephritis clinical isolate expressing a K5 capsule. DS17 is highly susceptible to K5 phage-mediated lysis. Thus, compounds that were active in the K1F phage assay but did not inhibit phage in the T7 phage assay were analyzed using this assay. This secondary assay was conducted in the 96 well plate format. Bacterial cultures of E. coli strain DS17 were grown and prepared immediately prior to use. Overnight starting cultures of DS17 were grown at 37 °C and diluted 1:100 in LB. Compounds were added to plates in quadruplicate at 50 and 100 μM, followed by addition of 100 μL of bacterial culture. 1% DMSO (final well concentration) was included. The plates were tape sealed and shaken vigorously for 1.5 hr. An initial OD₆₀₀ reading was measured to identify compounds that cause growth retardation or bacterial killing in the absence of phage. Next, K5 bacteriophage (5 μL) was added to all of the test wells. The plates were resealed and shaken vigorously at 37 °C, and measurements of OD₆₀₀ for phage-mediated lysis were taken after 3 hr. True inhibitors of capsule yielded bacteria that were not susceptible to K5 bacteriophage and did not show lysis within 2 hr of the addition of phage. The positive control drug C7 (100 μM final well concentration) was used in this screen.

Endpoint Read: The plates were read at ambient temperature from the bottom for absorbance at A₆₀₀ in a BioTek Quantplate reader and the degree of inhibition of phage-mediated lysis was determined based on the absorbance.

Study Objective

To evaluate, in radioligand binding assays, the activity of probe compound ML333 across a panel of 67 receptors.

Methods

Methods employed in this study have been adapted from the scientific literature to maximize reliability and reproducibility. Reference standards were run as an integral part of each assay to ensure the validity of the results obtained.

Where presented, IC₅₀ values were determined by a non-linear, least squares regression analysis using MathIQ™ (ID Business Solutions Ltd., UK). Where inhibition constant (K_i) are presented, the Ki values were calculated using the equation of Cheng and Prusoff (Cheng. Y., Prusoff, W.H., Biochem. Pharmacol. 22:3099–3108, 1973) using the observed IC₅₀ of the tested compound, the concentration of radioligand employed in the assay, and the historical values of the KD of the ligand (obtained experimentally at Ricerca Biosciences, LC). Where presented, the Hill coefficient (n_H), defining the slope of the competitive binding curve, was calculated using MathIQ™. Hill coefficients significantly different than 1.0, may suggest that the binding displacement does not follow the laws of mass action with a single binding site. Where IC₅₀, K_i, and/or nH data are presented without Standard Error of the Mean (SEM), data are insufficient to be quantitative, and the values presented (Ki, IC50, n_H) should be interpreted with caution.