Cruzain is a key cysteine protease that is essential for the survival and replication of Trypanosoma cruzi (T. cruzi), a protozoan parasite that is the causative agent of Chagas disease. Inhibition of cruzain has been validated as a viable strategy for the development of small molecule therapeutics for Chagas disease. To date, reports of small molecule cruzain inhibitors have been primarily of those which contain an electrophilic warhead and act by irreversible, covalent modification of the enzyme. As such, we sought to discover novel reversible, non-covalent inhibitors of cruzain using a combination of docking, co-crystallization and high-throughput screening. In this report, we describe the discovery of ML217, which exhibits trypanocidal activity against the T. Cruzi parasite, while having minimal toxicity to the host cell. Moreover, ML217 represents the first reversible non-covalent inhibitor of cruzain to demonstrate efficacy in a Chagas disease mouse model.
Assigned Assay Grant #: DA024891
Screening Center Name & PI: NIH Chemical Genomics Center, Christopher P. Austin
Chemistry Center Name & PI: NIH Chemical Genomics Center, Christopher P. Austin
Assay Submitter & Institution: Brian Shoichet, University of California, San Francisco
PubChem Summary Bioassay Identifier (AID): 2413
Probe Structure & Characteristics
Recommendations for Scientific Use of the Probe
Researchers in the Chagas disease field will find utility in ML217 due to its efficacy and its non-covalent, reversible mechanism of action. In addition to its T. Cruzi growth inhibition, researchers in related parasitic-based neglected tropical diseases may find ML217 to have a common mechanism and potential anti-parasitic activity. Lastly, due to its potential for medicinal chemistry optimization, ML217 is a suitable starting point for small molecule-based therapeutic development for Chagas disease.
At the outset of this project we had three goals in mind. The first goal was to discover and develop novel cruzain inhibitors with different mechanisms of action to ultimately aid in the development of antitrypanosomal agents. Second, using the cruzain enzymatic assay as a model screening system of an enzyme carrying a reactive cysteine catalytic residue, we aimed to profile the MLSMR collection with respect to several sources of false-positive or promiscuous types of inhibition: compound autofluorescence1, colloidal aggregation, and reactive compounds. The third goal was to improve the chances for discovering novel competitive cruzain inhibitors by using two techniques, virtual screen and HTS, by performing a virtual screen of the collection and characterizing the top docking hits using the same cruzain enzymatic assay. The first goal of finding novel cruzain inhibitors has clearly been met with the declaration of two probes, ML091 and ML0922, which both have different modes of inhibition, non-covalent and covalent inhibition, respectively. Moreover, we describe herein the development of a third structurally distinct chemotype which offers advantages over our two previously described probes (vide infra).
The second goal for the project has also been addressed through the comprehensive analyses of aggregation, autofluorescence, and reactivity artifacts for 197,861 compounds that are a part of the MLSMR.3 In this study, several key observations were noted, the first being that false positives resulting from aggregation far-outnumbered those resulting from either autofluorescent and/or reactive mechanisms. Second, despite this assay being potentially sensitive to reactive functionality as a result of a nucleophilic active-site cysteine, very few compounds possessing suspected reactive functionalities showed up as hits. Finally, we found that aggregation appears to be context dependent. When studying aggregators across assay platforms, enzymes, and detection formats, we found little correlation between detergent-sensitive inhibition, suggesting that one cannot simply remove these compounds from the library based on a singular result. This also supports the notion that reducing the artifacts resulting from aggregation can be achieved by the inclusion of detergent in the assay medium whenever possible.
Our final outlined goal was to prospectively and retrospectively compare the outcome of virtual and high-throughput screens of the same compound collection.4 This approach allowed us to prioritize molecules that were both predicted by docking and that were active in the HTS, and ultimately led to the selection of the novel cruzain inhibitor described in this report. Through this comparative analysis, several non-covalent competitive inhibitors were discovered, and while co-crystallization efforts results in crystals for several of these inhibitors, only compound 1 had desirable electron density (vide infra). Interestingly, this case is a good example of both methods having complementary strengths and weaknesses, as this compound had a docking rank of 7,560, and likely would not have been pursued in the absence of HTS data. Having confirmed the lack of covalent attachment through the X-ray structure, we were eager to pursue this compound in additional studies.
Cruzain inhibitors 25,36 and our recently reported ML probe (ML092), all utilize an electrophilic warhead moiety which is attacked by the active-site cysteine moiety to form a colavent interaction. Both the vinyl sulfone (compound 2) and the tetrafluorophenoxymethyl ketone (compound 3) act by irreversible inhibition, while the purine nitrile (ML092) can be classified as a reversible, covalent inhibitor of cruzain. In contrast, our other ML probe (ML091), was initially found to be a reversible, non-covalent inhibitor.7 All of the covalent inhibitors of cruzain (shown above) exhibit potent in vitro inhibition of cruzain (2: IC50 = 1.5 nM, 3: Ki = 460 nM, and ML092: IC50 = 0.2 nM, whereas the noncovalent inhibitor, ML091, has an IC50 value of 1.2 μM. Despite the potent activity of compound 3, the T. Cruzi IC50 was 5.1 μM and the compound had a host cell IC50 value of >10 μM, indicating limited selectivity. Our previously reported probe compound, ML092, was found to have potent efficacy against T. Cruzi but was also fairly toxic to the host cell and non-covalent inhibitor, ML091, was found to have minimal effect on the T. Cruzi parasite. In contrast, the probe described in this report exhibits T. Cruzi activity comparable to irreversible covalent inhibitors (2 and 3), is non-toxic to the host cell, and displays in vivo activity in a Chagas disease mouse model at 10 mg/kg SID. While both 2 and 3 also show favorable activity in mouse models, the frequency of treatment and dosing levels are higher, at 100 mg/kg and 20 mg/kg BID respectively.
2. Materials and Methods
General Methods for Chemistry. Unless otherwise stated, all reactions were carried out under an atmosphere of dry argon or nitrogen in dried glassware. Indicated reaction temperatures refer to those of the reaction bath, while room temperature (rt) is noted as 25 °C. All solvents were of anhydrous quality, purchased from Aldrich Chemical Co. and were used as received. Commercially available starting materials and reagents were purchased from Aldrich and were also used as received. Analytical thin layer chromatography (TLC) was performed with Sigma Aldrich TLC plates (5 × 20 cm, 60 Å, 250 μm). Visualization was accomplished by irradiation under a 254 nm UV lamp. Chromatography on silica gel was performed using forced flow (liquid) of the indicated solvent system on Biotage KP-Sil pre-packed cartridges and the Biotage SP-1 automated chromatography system. 1H- and 13C NMR spectra were recorded on a Varian Inova 400 MHz spectrometer. Chemical shifts are reported in ppm with the solvent resonance as the internal standard (CDCl3 7.26 ppm, 77.00 ppm, DMSO-d6 2.49 ppm, 39.51 ppm for 1H, 13C respectively). Data are reported as follows: chemical shift, multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, br = broad, m = multiplet), coupling constants, and number of protons. Low resolution mass spectra (electrospray ionization) were acquired on an Agilent Technologies 6130 quadrupole spectrometer coupled to the HPLC system. High resolution mass spectral data was collected in-house using an Agilent 6210 time-of-flight mass spectrometer, also coupled to an Agilent Technologies 1200 series HPLC system. If needed, products were purified via a Waters semi-preparative HPLC equipped with a Phenomenex Luna® C18 reverse phase (5 micron, 30 × 75 mm) column having a flow rate of 45 ml/min. The mobile phase was a mixture of acetonitrile (0.025% TFA) and H2O (0.05% TFA), and the temperature was maintained at 50 °C.
Samples were analyzed for purity on an Agilent 1200 series LC/MS equipped with a Luna® C18 reverse phase (3 micron, 3 × 75 mm) column having a flow rate of 0.8–1.0 mL/min over a 7-minute gradient and a 8.5 minute run time. Purity of final compounds was determined to be >95%, using a 3 μL injection with quantitation by AUC at 220 and 254 nm (Agilent Diode Array Detector).
Kinetic Fluorogenic qHTS Assay for Inhibitors of Cruzain: Three μL of reagents were dispensed into 1536-well Greiner black solid-bottom assay plate. Compounds and controls (23 nL) were transferred via Kalypsys PinTool equipped with 1536-pin array (10 nL slotted pins, V&P Scientific, Palo Alto, CA). The plate was incubated for 15 min at room temperature, and then a 1 μL aliquot of 8 μM substrate solution was added to start the reaction. The plate was transferred to ViewLux high-throughput CCD imager (Perkin-Elmer, Waltham, MA), where kinetic measurements (4 reads, one read every 30 seconds) of the AMC fluorescence were acquired using standard 340 nm excitation and 450 nm emission filter sets. During dispense, reagent bottles were kept submerged into a 4 °C recirculating chiller bath and all liquid lines were covered with aluminum foil to minimize fluorophore degradation. All screening operations were performed on a fully integrated robotic system (Kalypsys Inc, San Diego, CA) as described elsewhere. Plates containing DMSO only (instead of compound solutions) were included approximately every 50 plates throughout the screen to monitor any systematic trend in the assay signal associated with reagent dispenser variation, or any decrease in enzyme specific activity. Pubchem AID 1478.
Kinetic Fluorogenic qHTS Assay for Inhibitors of Cruzain (Detergent-Free Screen): Three μL of reagents were dispensed into 1536-well Greiner black solid-bottom assay plate. Compounds and controls (23 nL) were transferred via Kalypsys PinTool equipped with 1536-pin array (10 nL slotted pins, V&P Scientific, Palo Alto, CA). The plate was incubated for 15 min at room temperature, and then a 1 μL aliquot of 8 μM substrate solution was added to start the reaction. The plate was transferred to ViewLux high-throughput CCD imager (Perkin-Elmer, Waltham, MA), where kinetic measurements (4 reads, one read every 30 seconds) of the AMC fluorescence were acquired using standard 340 nm excitation and 450 nm emission filter sets. During dispense, reagent bottles were kept submerged into a 4 °C recirculating chiller bath, and all liquid lines were covered with aluminum foil to minimize fluorophore degradation. All screening operations were performed on a fully integrated robotic system (Kalypsys Inc, San Diego, CA) as described elsewhere. Plates containing DMSO only (instead of compound solutions) were included approximately every 50 plates throughout the screen to monitor any systematic trend in the assay signal associated with reagent dispenser variation or decrease in enzyme specific activity. Pubchem AID 1476.
Confirmatory Assay: Top inhibitors, including some potentially covalent modifiers such as triazine nitriles, were cherry-picked for follow-up confirmation using the primary screening protocol. Several commercially available compounds, including new benzimidazoles, were also purchased for follow-up based on the inhibitor clusters. A total of 599 compounds were tested in the confirmatory assay. All original qHTS actives confirmed while several of the newly purchased analogs were inactive. Pubchem AID 2158.
qHTS Counterscreen Assay for Papain Inhibition: Although high selectivity was not a stated prerequisite for probe nomination, we wished to evaluate the selectivity of our top small molecules against related proteases. Papain, another cysteine protease known to utilize the same Z-FR-AMC fluorogenic substrate, was selected as a convenient profiling target. The fluorogenic assay used for cruzain was used, with the exception of the reducing agent, which was 5 mM cystein instead of DTT. A total of 414 compounds were tested in this counter screen assay. The benzimidazole series was inactive against papain, though other series such as the triazine nitriles, inhibited papain. Pubchem AID 2161.
T. Cruzi Growth Inhibition Assay:8 NIH/3T3 cells and parasites were harvested, washed once and resuspended in DMEM supplemented with 2% FBS and Pen-Strep-Glut. DMEM did not contain phenol red to avoid interference with the assay absorbance readings at 590 nM. Different numbers of NIH/3T3 cells were seeded in 96-well plates. After 3 hours, compounds were added at the indicated concentrations and mixed by pipetting. BZN tablets (Rochagan, Roche) dissolved in DMSO and 4 μM Amphotericin B solution (Sigma-Aldrich) were used as positive controls. Different numbers of T. cruzi parasites were added in a final volume of 200 μl/well. After 4 days, 50 μL of PBS containing 0.5% of the detergent NP40 and 100 μM Chlorophenol Red-β-D-galactoside (CPRG) (Fluka) were added. Plates were incubated at 37 °C for 4 hours and absorbance was read at 590 nm using a Tecan Spectra Mini plate reader. To calculate the Z′ factor, we used the formula described by Zhang et al.:9 Z′ = 1−[(3σc++3σc−)/|μc+−μc−|] where σc+ = standard deviation (SD) of positive control, σc− = SD of negative control, μc+ = mean of positive control, μc− = mean of negative control. Subsequently, the best ratio was used for all growth inhibition assays (50,000 cells and parasites, multiplicity of infection (MOI) 1:1). To determine IC50 values, β-gal activity (Abs590) was plotted against compound concentration for each compound. The IC50 was determined as the concentration at which the activity (absorbance) was half that in the absence of compound. Mean IC50 values are the average of independent experiments performed in triplicate on three different days.
Cytotoxicity Assay: Cells (NIH/3T3) were washed, plated using 200 μL at a density of 50,000 cells/well in 96-well plates, and were allowed to adhere for 3 hours. Twenty-four hour assays were done in DMEM without phenol red supplemented with 10% FBS and Pen-Strep-Glut, while 4-day assays were done in the same medium containing 2% FBS. Drugs were added and mixed. After 1 or 4 days, 20 μL of Alamar Blue (Biosource, Invitrogen) was added. Plates were incubated for 4 – 6 hours (NIH/3T3) at 37 °C and fluorescence was read using a Labsystems Fluoroskan II plate reader (excitation: 544 nm, emission: 590 nm). To determine TC50 values, fluorescence was plotted against inhibitor concentration. TC50 was determined as the concentration at which cytotoxicity (fluorescence) was half that in the absence of inhibitor.
2.2. Probe Chemical Characterization
*Purity >95% as judged by LC/MS and 1H NMR
N-(2-(1-(2-(p-tolyloxy)ethyl)-1H-benzo[d]imidazol-2-yl)ethyl)cyclohexanecarboxamide.1H NMR (CDCl3) δ 1.09–1.28 (m, 3 H), 1.29–1.51 (m, 2 H), 1.53–1.88 (m, 5 H), 1.96–2.11 (m, 2 H), 2.24 (s, 3 H), 3.17 (t, J = 6.0 Hz, 2 H), 3.86 (q, J = 6.0 Hz, 2 H), 4.22 (t, J = 5.0 Hz, 2 H), 4.49 (t, J = 4.6 Hz, 2 H), 6.66 (d, J = 8.6 Hz, 2 H), 6.94 (brs, 1 H), 7.01 (d, J = 8.6 Hz, 2 H), 7.20–7.32 (m, 1 H), 7.36–7.46 (m, 1 H) and 7.66–7.81 (m, 1 H). 13C (CDCl3) δ 20.40, 20.44, 25.58, 25.70, 25.74, 27.31, 29.25, 29.53, 36.09, 43.23, 45.44, 66.01, 109.37, 114.10, 119.15, 119.18, 122.18, 122.46, 129.95, 130.73, 134.93, 142.39, 153.70, 155.76 and 176.31. HRMS (ESI) m/z (M+H)+ calcd. for C25H32N3O2N, 406.2489; found 406.2499.
- LC/MS (Agilent system) Retention time t1 (short) = 3.17 min and t2 (long) = 4.82
- Column: 3× 75 mm Luna C18, 3 micron
- Run time: 4.5 min (short); 8.5 min (long)
- Gradient: 4% to 100%
- Mobile phase: Acetonitrile (0.025 % TFA), water (0.05 % TFA).
- Flow rate: 0.8 to 1.0 mL
- Temperature: 50 °C
- UV wavelength: 220 nm, 254 nm
MLS Numbers for probe and analogs
Probe in vitro ADME properties
Figure 1Buffer Stability (48 hours at 25 °C) of ML217
2.3. Probe Preparation
Preparation of tert-butyl 2-(1H-benzo[d]imidazol-2-yl)ethylcarbamate (Step 1). BOC-anhydride (0.99 mL, 4.27 mmol), 1 N NaOH (15 mL, 14.95 mmol) was added to a suspension of commercially available 2-(1H-benzo[d]imidazol-2-yl)ethanamine dihydrochloride (Sigma Alrich) (1.00 g, 4.27 mmol) in CH2Cl2. The biphasic reaction was stirred at room temperature for 5 hrs, then the solids were removed by fitration, washed with CH2Cl2, concentration under reduced pressure then under high vacuum to give a greater than 98% pure product in a 70% yield. LC-MS (min) = 2.69 min; 1H NMR (DMSO-d6) δ 1.35 (s, 9 H), 2.91 (t, J = 7.3 Hz, 2 H), 3.36 (m, 2 H), 6.95 (brs, 1 h), 7.08 (dd, J = 6.0 Hz and 3.2 Hz, 2 H) and 7.44 (dd, J = 6.0 Hz and 3.2 Hz, 2 H).
Preparation of tert-butyl-2-(1-(2-(p-tolyloxy)ethyl)-1H-benzo[d]imidazol-2-yl)ethylcarbamate (step 2). General procedure for alkylation: A suspension of tert-butyl 2-(1H-benzo[d]imidazol-2-yl)ethylcarbamate (0.52 mg, 1.99 mmol), 1-(2-bromoethoxy)-4-methylbenzene (0.43 g, 1.99 mmol), and cesium carbonate (1.95 g, 5.97 mmol) in acetonitrile/DMF (10 ml/5 ml) was heated to 8 5°C in for 4 hrs. The reaction mixture was cooled to room temperature, filtered and the solids were washed with acetone, and concentrated leaving only DMF remaining. The resulting residue was diluted with ethyl acetate and washed three times with water, dried with Na2SO4, filtered, and concentrated to yield clear glass-like oil in 89% yield. LC-MS (min) 3.14; 1H NMR (CDCl3) δ 1.14 (s, 9 H), 2.23 (s, 3H), 3.15 (t, J = 6.0 Hz, 2 H), 3.75 (d, J = 6.0 Hz, 2 H), 4.21 (t, J = 5.4 Hz, 2 H), 4.49 (t, J = 5.3 Hz, 2 H), 6.67 (d, J = 6.0 Hz, 2 H), 7.00 (d, J = 6.0 Hz, 2 H), 7.18–7.27 (m, 2H), 7.43 (d, J = 4.3 Hz, 1H) and 7.70 (d, J = 4.2 Hz, 1 H).
Preparation of 2-(1-(2-(p-tolyloxy)ethyl)-1H-benzo[d]imidazol-2-yl)ethanamine dihydrochloride (step 3). Tert-butyl-2-(1-(2-(p-tolyloxy)ethyl)-1H-benzo[d]imidazol-2-yl)ethylcarbamate was treated with 4M HCl/Dioxane and stirred for 2 hrs. The reaction mixture was then filtered and the solids were washed with CH2Cl2. The resulting colorless solid was dried under high vacuum and carried used in the next reaction without further purification. LC-MS (min) = 2.73.
Preparation ofN-(2-(1-(2-(p-tolyloxy)ethyl)-1H-benzo[d]imidazol-2-yl)ethyl) cyclohexanecarboxamide [ML217)] (Step 4). Diisopropylethylamine (71.0 μL, 0.41 mmol) was added to a solution containing 2-(1-(2-(p-tolyloxy)ethyl)-1H-benzo[d]imidazol-2-yl)ethanamine dihydrochloride (0.050 g, 0.14 mmol) in CH2Cl2 (1.5 mL). The resulting suspension was stirred until clear, then cyclohexanecarbonyl chloride (74 μL, 0.14 mmol) was added. The reaction mixture for 30 min. at room temperature was then diluted with CH2Cl2 and washed with 1N NaOH, brine and dried over Na2SO4. The organic layer was filtered then concentrated under reduced pressure and purified by reversed-phase HPLC (see methods section for details).
General Synthesis for Various Phenoxy-Derivatives (Mitsunobu Reaction). A solution containing DIAD (0.13 mL, 0.67 mmol), and Ph3P (0.18g, 0.67 mmol) in toluene/THF was stirred at room temperature for 15 min. The requisite phenol (0.62 mmol) was added and the reaction mixture was stirred at room temperature for 15 min before cooling in an ice bath. N-(2-(1-(2-hydroxyethyl)-1H-benzo[d]imidazol-2-yl)ethyl)cyclohexanecarboxamide (0.14 g, 0.44 mmol) was then added to the cooled solution and allowed to stir at this temperature for 30 min, at which time was warmed to room temperature and stirred for an additional 3 hrs. The resulting orange solution was poured into ethyl acetate and washed with water, 1 N NaOH, and brine, then dried over Na2SO4, filtered and concentrated under reduced pressure to yield an orange oil, which was purified by reversed-phase HPLC.
3.1. Summary of Screening Results
To measure the enzymatic activity of cruzain, we utilized its model fluorogenic substrate Z-Phe-Arg-AMC, which is converted to a highly fluorescent 7-amino-4-methylcoumarin reporter upon cruzain-catalyzed hydrolysis. Assay optimization was performed directly in 1,536-well format at a final reaction volume of 4 μL. For the detergent-present screen, Triton X-100 was used at 0.01%, and the cruzain concentration selected for screening was 1.5 nM. During the optimization of the detergent-free assay, low stability and high variability in the specific activity of cruzain was noted, likely due to protein denaturation promoted by the large surface-to-volume ratio in the 1,536-well polystyrene assay plates. In order to stabilize the enzyme’s performance, its final concentration was raised to 3 nM and a trace of Triton X-100 (final concentration of 0.00005%) was included in the detergent-free assay; we note that a similar adjustment step was needed for the AmpC beta-lactamase aggregation screen described earlier. At the conditions selected – 1.5 nM or 3 nM final cruzain concentration and 2 μM final substrate concentration (the latter was chosen to match previously reported conditions and being close to the Km value for this substrate) – the signal evolution was robust, and low substrate conversion was monitored over the course of 1 minute, making the assay highly sensitive to cruzain inhibitors. All assay components were tested and found stable for at least 24 hours when formulated as stock solutions at their working concentrations, in both the detergent-present and the detergent-free buffers (data not shown). Such demonstrated stability permitted the implementation of an unattended overnight screening operation.
The screens of the 197,864-compound collection were completed within a period of two separate workweeks. During the first week, approximately 60% of the 1,107 1,536-well plates library were screened in the detergent-free assay first; this was immediately followed by a screen of the same set of library plates in the detergent-present assay. The remaining 40% of the library was screened in the same manner during the second week. The screening of each compound against the two assays in close succession minimized the possibility for sample-age related differences in results. Overall, two sets of 1,107 1,536-well assay plates were run under the detergent-free and detergent-present conditions, respectively, leading to the generation of 197,864 concentration responses consisting of at least seven points per compound per assay, and corresponding to a total of approximately 1.5 million samples tested per detergent condition.
The Z′ screening factors associated with each plate and each screening condition remained high and stable throughout the two screens (Figure 2): the average Z′ for the detergent-free screen was 0.78, while the corresponding average for the detergent-present screen was 0.93. Of the two screens comprising a total of 2,214 plates, only a group of six plates failed and was re-screened immediately using the same batches of enzyme and substrate. As a further quality control measure, we included a concentration response of the known vinyl sulfone cruzain inhibitor K1177721 (2), added as a 16-point dilution series in duplicate between 5.7 μM and 0.175 nM into the second column of every assay plate. The shape and quality of the concentration response remained consistent throughout both screens (Figure 2b, green data points) with the associated minimum significant ratios10 of 2.5 and 1.4 for the detergent-free and detergent-present conditions, respectively, further indicating stable runs.
The cumulative effect of all library compounds on the cruzain activity at each screening condition is shown in Figure 3. On the plots, concentration responses were color-coded and positionally sorted based on activity with inactive samples (flat concentration responses) represented by the black dots, activators in red, and inhibitors in blue; green points in the very front of the 3D plots represented the 1,107 duplicate responses of the intra-plate control titration of K11777 (2). Similar to our AmpC beta-lactamase profiling, the outcomes from the detergent-free versus detergent-present screens were strikingly different: the detergent-free screen yield over 15 times more hits than its detergent-present counterpart. Also, as noted previously, we observed a large number (~12,000) of apparent activators in the detergent-free screen. Of these samples, almost all turned completely inactive upon inclusion of detergent. Over 85% of the activators were associated with partial or single-point top concentration responses, indicating that the condition-dependent activation was being observed only at the highest compound concentrations where complicating phenomena such as transient precipitation, light scatter, and compound aggregate-assisted enzyme stabilization have been known to lead to false positive effects. We did not consider those compounds further.
The two primary screens led to a large number of potential actives that needed substantial filtering. A large percentage of the actives (12,746) were detergent-sensitive ‘activators’ as mentioned above. These were identified and filtered out using our curve classification method.11 The screen with detergent helped eliminate another 10,399 detergent-sensitive weak curves. An additional 3,844 detergent-sensitive inhibitors gave significant response but were inactive in the presence of detergent. These were considered likely aggregators and were filtered out. Next, the kinetic reads captured in the primary screen provided background auto-fluorescence data, which were used to reject 507 false positive concentration response curves. Substructure patterns of 243 reactive and problematic functional groups were used to filter out another 428 compounds. Finally, 550 weak inhibitors were filtered out due to low potency. The remaining 493 compounds were clustered, and among the chemotypes identified was a benzimidazole-based series.
3.2. Dose Response Curves for Probe
T. Cruzi growth inhibition activity vs. cytotoxicity is shown in Figure 4.
3.3. Scaffold/Moiety Chemical Liabilities
While many of the series found in the qHTS contained reactive functional groups, the series chosen for probe development did not contain any serious liabilities.
3.5. Cellular Activity
3.6. Profiling Assays
See in vitro ADME profiling properties in Section 2.2.
In our previous effort which focused on reversible, covalent inhibitors of cruzain (probe: ML92), we found that the potency could be improved greatly through subtle modification to the aniline moiety which occupied the S2 region of the enzyme (see Figure 6). Moreover, we found the purine nitrogen (N-9) reaches toward the surface of the enzyme (S1 region) and makes no direct interactions, thus substitution of this position was used to modulate ADME properties (e.g. solubility). As shown in the co-crystal structure of N-(2-(1H-benzo[d]imidazol-2-yl)ethyl)-2-(2-bromophenoxy)acetamide (1), the molecule adopts a similar orientation as the purine nitrile probe, where the 2-bromo-phenol occupies the S2 pocket of the enzyme, a predominately hydrophobic pocket, and the benzimidazole core is exposed to solvent (Figure 7). A key difference is the seemingly critical interactions involving the amide moiety (NH-Asp161 and C=O to Gly66). The importance of this interaction is something we planned to investigate early on in the SAR studies. However, before doing so, we did some exploratory SAR to study variations of the amide moiety and substitution of the benzimidazole nitrogen. Interestingly, replacing the 2-bromophenoxy-acetamide moiety with a simple cyclohexane ring (5) retained activity (6.7 μM for 1 and 6.4 μM for 5). Given that this compound should be more stable and is easier to synthesize, we decided to use this group for initial investigations of substitution on benzimidazole nitrogen. These studies revealed that a two carbon linker with a terminal phenoxy group improved potency (e.g. analog 18, IC50 = 1.8 μM); this provided the opportunity for us to further explore the SAR of this position in addition to preparing other amide analogs. Before utilizing parallel library synthesis to explore the SAR of the amide and phenoxy moiety, we aimed to investigate the effect of varying the length of the C-2 side chain and modifications to the amide (Table 1). Changing the C-2 side chain length from n = 2 to n = 3 (analog 4) resulted in complete loss of activity as did shortening the side chain (n = 1, data not shown). Methylation of the amide nitrogen (compound 6) and removal of the amide carbonyl (compound 7) both resulted in loss of activity, which supports the binding mode depicted in the co-crystal structure. Interestingly, the sulfonamide analog 8 maintained some activity (16 μM), whereas the urea derivative 9 lost all activity.
Having established some key SAR points from these initial efforts, we next aimed to explore the SAR around the amide moiety (e.g. cyclohexyl group) and additional modifications to the C-1 side chain. Given our previously reported ability to modulate potency via optimization of the group occupying the S2 region of the enzyme, we were hopeful that analogs of the amide group would provide the desired improvements in potency. Analogs 12–15 looked at differing ring sizes (cyclopentane 12, cycloheptane 13, cyclopropane 14 and cyclohexane 15) and while 12, 13 and 15 all had comparable potencies; the smaller cyclopropane ring analog 14 was essentially inactive. This result can be rationalized by the need for larger hydrophobic groups to fully occupy this portion of the binding site. Next we investigated modifications to the terminal phenoxy group. As anticipated, changes to this portion of the molecule were well tolerated (compounds 16–21, 33–35). Analogs 19 and 20 (R2 = 3-NH2 and 2-NH2; IC50 = 3.2 and 4.6 μM respectively) had comparable potency to the lead compound and should have improved solubility. We then turned our attention to further modification of the amide moiety, holding the p-tolyl-oxyethyl group constant for direct comparison to previous analogs. Replacement of the cyclohexane ring with a phenyl group resulted in a ~6-fold loss in potency. Incorporation of halogens, particularly fluorine, into this portion of the molecule on our purine scaffold provided great improvements in potency. However, in this case, though potency was improved the increased affinity was much less pronounced. Potency tended to decrease with decreasing size of the substituent (e.g. analog 23 (R1 = 2-fluorophenyl, IC50 = 6.5 μM), 27 (R1 = 2-bromophenyl, IC50 = 12.9 μM), 38 (R1 = 2-CF3-phenyl, IC50 = 16 μM), 24 (R1 = 3-fluorophenyl, IC50 = 6.5 μM) and 28 (R1 = 3-chlorophenyl, IC50 = 9.1 μM). Moreover, electron-withdrawing groups appear to be favored over electron-donating groups, exemplified by analogs 31 and 32 (2-methylphenyl and 4-methylphenyl) which are both less potent than the corresponding analogs containing electron-withdrawing substituents in these positions. Finally, we looked to explore changes to C-2 side chain, including varying the length (compound 43), replacing the oxygen with carbon (analog 11), and adding unsaturation (compound 44). These investigations revealed that the oxygen was not required for activity, varying the length of this side-chain had minimal effect, and more rigid, unsaturated side chains were also tolerated.
Generally, the SAR around this scaffold appeared relatively flat, with the essential functional group being the amide moiety, as any changes to this resulted in a drastic loss of activity. In contrast to our reversible, covalent inhibitors that had IC50 values in the low nM to pM range, these compounds hovered around 1 μM. These efforts, combined with those appearing in the literature, seem to indicate that for low to mid-nM inhibition of cruzain, covalent modification of the enzyme is strongly preferred.
T. Cruzi Activity
Despite the moderate in vitro potency of these compounds against cruzain, we were quite encouraged by experiments performed in the lab of Professor Ana Rodriquez, where they investigated activity against the T. Cruzi parasite and activity against the host cells (cytotoxicity) as shown in Figure 4. At the time of these studies, only a few reasonably potent benzimidazole analogs had been synthesized (e.g. compounds 15 and 16), so these were tested along with our previously reported purine and triazine nitrile covalent inhibitors. Despite being much less potent in the in vitro assay, these non-covalent inhibitors displayed encouraging IC50 values against the T. Cruzi parasite and were not very toxic against the host cell as shown in Figure 4 (at levels over 10-fold the EC50 value against the parasite). Having an acceptable EC50/TC50 ratio prompted us to investigate these compounds in the T. Cruzi in vivo mouse model. This in vivo model utilizes bioluminescent T. Cruzi, in which the firefly luciferase enzyme is stably expressed within the parasite.12 This method allows for rapid analysis of potential drugs for the treatment of Chagas disease (Figure 5A). Our compounds were dosed at 10 mg/kg, once/day (SID) and monitored compared to a control. Treatment began 7 days post infection and the parasite burden was analyzed at day 12 (data not shown) and day 19 (Figure 5B). We found a significant reduction of parasite burden at day 19, compared to the control and after 5 days treatment. These results are significant as other reported in vivo T. Cruzi parasite killing assays were conducted with higher doses (20–100 mg/kg) and more frequent treatment regiments (BID). These results suggest that over a longer treatment time, and using comparable dosing, these compounds would have an even more pronounced effect.
Our current probe molecule shows activity against cruzain, T. Cruzi parasite killing, is non-toxic to the host cell (NIH/3T3 cells), and shows significant reduction of parasite burden in Chagas disease mouse models. However, this compound has only limited aqueous solubility (5 μM), and thus could have poor absorption and oral bioavailability. As such, we are encouraged by the improved solubility of compounds 19 and 20, which have much improved solubility in PBS buffer (pH 7.4) of >150 μM while maintaining comparable potency of the probe compound (15). These compounds have not yet been tested in the T. Cruzi assay or the Chagas disease mouse model, but the preliminary results suggest that the solubility liability has already been addressed.
4.1. Comparison to Existing Art and How the New Probe is an Improvement
While there are numerous small molecules reported to have efficacy against the T. Cruzi parasite, treatment is still limited to nitroaromatic compounds, benznidazole and nifurtimox. Both of these compounds are not only toxic, but are also ineffective against the frequently occurring chronic stage of Chagas disease. The only known cruzain inhibitors that have progressed into animal models for Chagas infection include the vinyl sulfone 2 (and related analogs) and a tetrafluorophenoxymethyl ketone containing inhibitor 3, both of which are irreversible covalent modifiers of cruzain. Compound 2 has shown to be effective in curing T. Cruzi infection in both cell and animal models, but is plagued by low oral bioavailabilty and a short half-life. Both of these compounds have thus far proven save in animal models, but the potential for toxicity associated with irreversible modification exists. As such, this report represents to the best of our knowledge the first example of a reversible, non-covalent inhibitor of cruzain showing substantial efficacy in a Chagas disease mouse model. Considering our compounds were dosed at 10 mg/kg SID compared to 20 mg/kg BID (both via IP) for compound 2, the potential exists for even greater reduction in parasite burden for our compounds if we use a comparable dosing regimen. Additionally, these compounds had over 10-fold selectivity.
4.2. Mechanism of Action Studies
See earlier discussion section on co-crystallization studies of compound related to the probe molecule.
4.3. Planned Future Studies
Currently, Ana Rodriguez is testing approximately 80 new benzimidazole analogs in the high-throughput T. Cruzi assay and host cell toxicity studies, most of which are highlighted in this report. The top candidates which emerged as having optimal IC50/TC50 ratios and ADME properties will be progressed to her in vivo model for Chagas disease (as described above). We will also investigate whether these compounds are trypanostatic (inhibit replication within the host cell) or trypanocidal (induce lysis of intracellular amastigotes). Trypanostatic compounds may be more effective at treating Chagas during the acute phase of the disease, whereas typanocidal compounds may be required for activity against chronic forms of the disease. To investigate whether the parasites have been completed eliminated, the mice will be immuno-suppressed, and we will wait to see if the parasites come back. One of the final studies would be to see if these compounds act against the Chronic stage of the disease. This is important because the majority of cases are found in this stage; however, this requires long protocols (3 month wait after infection) followed by treatment. As such, this study will be limited only to compound(s), which have been analyzed in the other studies. Importantly, none of the compounds used today are effective towards this stage of the disease.
Concurrently, we will be investigating the in vitro ADME properties of several of the top actives to determine which compounds are best suited for in vivo PK. We have already discovered compounds that improved solubility compared to our probe molecule, while maintaining potency as shown above. Given the apparent metabolic liability of this probe compound, we plan to investigate METID studies to help establish the metabolic soft-spots. Preliminary analysis seems to indicate that the cyclohexane is the primary spot for metabolism. A compound with improved ADME properties would presumably translate to improved in vivo efficacy. We plan to submit an extended characterization proposal to provide funds for the studies mentioned above.
- Simeonov A, Jadhav A, Thomas CJ, Wang Y, Huang R, Southall NT, Shinn P, Smith J, Austin CP, Auld DS, Inglese J. Fluorescence spectroscopic profiling of compound libraries. J. Med. Chem. 2008;51:2363–2371. [PubMed: 18363325]
- Mott BT, Ferreira RS, Simeonov A, Jadhav A, Ang KK, Leister W, Shen M, Silveira JT, Doyle PS, Arkin MR, McKerrow JH, Inglese J, Austin CP, Thomas CJ, Shoichet BK, Maloney DJ. Identification and optimization of inhibitors of Trypanosomal cystein proteases: cruzain, Rhodesian, and TbCatB. J. Med. Chem. 2010;53:52–60. [PMC free article: PMC2804034] [PubMed: 19908842]
- Jadhav A, Ferreira RS, Klumpp CK, Mott BT, Austin CP, Inglese J, Thomas CJ, Maloney DJ, Shoichet BK, Simeonov A. Quantitative analyses of aggregation, autofluorescence, and reactivity artifacts in a screen for inhibitors of a thiol protease. J. Med. Chem. 2010;53:37–51. [PMC free article: PMC2992957] [PubMed: 19908840]
- Ferreira RS, Simeonov A, Jadhav A, Eidam O, Mott BT, Keiser MJ, McKerrow JH, Maloney DJ, Irwin JJ, Shoichet BK. Complementarity between a docking and a high-throughput screen in discovering new cruzain inhibitors. J. Med. Chem. 2010;53:4891–4905. [PMC free article: PMC2895358] [PubMed: 20540517]
- Roush WR, Gwaltney SL II, Cheng J, Scheidt KA, McKerrow JH, Hansell E. Vinyl sulfonate esters and vinyl sulfonamides: potent, irreversible inhibitors of cysteine proteases. J. Am. Chem. Soc. 1998;120:10994–10995.
- Engel JC, Doyle PS, Hsieh I, McKerrow JH. Cysteine protease inhibitors cure an experimental Trypanosoma cruzi infection. J. Exp. Med. 1998;188:725–734. [PMC free article: PMC2213346] [PubMed: 9705954]
- Barr SC, Warner KL, Kornreic BG, Piscitelli J, Wolfe A, Benet L, McKerrow JH. A cysteine protease inhibitor protects dogs from cardiac damage during infection by Typanosoma cruzi. Antimicrob. Agents Chemother. 2005;49:5160–5161. [PMC free article: PMC1315979] [PubMed: 16304193]
- Doyle PS, Zhou YM, Engel JC, McKerrow JH. A cysteine protease inhibitor cures Chagas’ diease in an immunodeficient mouse model of infection. Antimicrob. Agents Chemother. 2007;51:3932–3939. [PMC free article: PMC2151429] [PubMed: 17698625]
- Brak K, Kerr ID, Barrett KT, Fuchi N, Debnath M, Ang K, Engel JC, McKerrow JH, Doyle PS, Brinen LS, Ellman JA. Nonpeptidic tetrafluorophenoxymethyl ketone cruzain inhibitors as promising new leads for Chagas disease chemotherapy. J. Med. Chem. 2010;53:1763–1773. [PMC free article: PMC2838180] [PubMed: 20088534]
- Recent evidence from the Shoichet lab suggests that this compound may in fact be a covalent modifier of the enzyme.
- Bettiol E, Samanovic M, Murkin AS, Raper J, Buckner F, Rodriguez A. Identification of three classes of heteroaromatic compounds with activity against intracellular Trypanosoma cruzi by chemical library screening. PLoS Hegl Trop Dis. 2009;3:e384. [PMC free article: PMC2639639] [PubMed: 19238193]
- Zhang JH, Chung TD, Oldenburg KR. A simple statistical parameter for use in evaluation and validation of high throughput screening assays. J. Biomol. Screen. 1999;4:67–73. [PubMed: 10838414]
- Eastwood BJ, Farmen MW, Iversen PW, Craft TJ, Smallwood JK, Garbison KE, Delapp NW, Smith GF. The Minimum Significant Ratio: A Statistical Parameter to Characterize the Reproducibility of Potency Estimates from Concentration-Response Assays and Estimation by Replicate-Experiment Studies. J. Biomol. Screen. 2006;11:253–261. [PubMed: 16490778]
- Inglese J, Auld DS, Jadhav A, Johnson RL, Simeonov A, Yasgar A, Zheng W, Austin CP. Quantitative high-throughput screening: a titration-based approach that efficiently identifies biological activities in large chemical libraries. Proc. Nat. Acad. Sci. USA. 2006;103:11473–11478. [PMC free article: PMC1518803] [PubMed: 16864780]
Diane Luci,a Wendy Lea,a Rafaela Ferreira,b Brian Shoichet,a Anton Simeonov,a Ana Rodriguez,c Ajit Jadhav,a and David J. Maloneya,*.
Received: April 15, 2011; Last Update: February 28, 2013.
National Center for Biotechnology Information (US), Bethesda (MD)
Luci D, Lea W, Ferreira R, et al. Reversible and non-covalent benzimidazole-based in vivo lead for Chagas disease. 2011 Apr 15 [Updated 2013 Feb 28]. In: Probe Reports from the NIH Molecular Libraries Program [Internet]. Bethesda (MD): National Center for Biotechnology Information (US); 2010-.