Identification of Novel Bisbenzimidazole Derivatives as Anticancer Vacuolar (H+)-ATPase Inhibitors †

The vacuolar (H+)-ATPases (V-ATPases) are a family of ATP-driven proton pumps and they have been associated with cancer invasion, metastasis, and drug resistance. Despite the clear involvement of V-ATPases in cancer, the therapeutic use of V-ATPase-targeting small molecules has not reached human clinical trials to date. Thus, V-ATPases are emerging as important targets for the identification of potential novel therapeutic agents. We identified a bisbenzimidazole derivative (V) as an initial hit from a similarity search using four known V-ATPase inhibitors (I–IV). Based on the initial hit (V), we designed and synthesized a focused set of novel bisbenzimidazole analogs (2a–e). All newly prepared compounds have been screened for selected human breast cancer (MDA-MB-468, MDA-MB-231, and MCF7) and ovarian cancer (A2780, Cis-A2780, and PA-1) cell lines, along with the normal breast epithelial cell line, MCF10A. The bisbenzimidazole derivative (2e) is active against all cell lines tested. Remarkably, it demonstrated high cytotoxicity against the triple-negative breast cancer (TNBC) cell line, MDA-MB-468 (IC50 = 0.04 ± 0.02 μM). Additionally, it has been shown to inhibit the V-ATPase pump that is mainly responsible for acidification. To the best of our knowledge the bisbenzimidazole pharmacophore has been identified as the first V-ATPase inhibitor in its class. These results strongly suggest that the compound 2e could be further developed as a potential anticancer V-ATPase inhibitor for breast cancer treatment.


Introduction
The vacuolar (H + )-ATPases (V-ATPases) are a family of ATP-driven proton pumps that couple ATP hydrolysis with the translocation of protons across membranes, and they are present in both the intracellular and cell surface membranes of eukaryotes. The V-ATPase proton pump is a macromolecular complex composed of at least 14 subunits organized into two functional domains, proton channel [1][2][3][4][5]. The V-ATPase plays a major role in the regulation of cellular pH conditions, and it has been associated with cancer invasion, metastasis, and drug resistance [6][7][8][9]. Despite knowing the clear involvement of V-ATPases in cancer, the therapeutic use of V-ATPase-targeting small molecules has not reached the clinic currently. Thus, V-ATPases are emerging as important targets for the identification of potential novel chemo therapeutic agents [10]. Macrolide antibiotics such as bafilomycin and concanamycin potently inhibit V-ATPases [11][12][13][14][15], but their use is complicated by adverse side effects on other targets. Few unnatural small molecule drugs have been explored as V-ATPase inhibitors, but the need for compounds that specifically target the mechanism(s) responsible for the low pH of tumors is growing. The development of useful V-ATPase inhibitors has been limited, even though huge efforts by both pharmaceutical industry and academic medicinal chemists are made, because of the complicated chemical structures of existing natural inhibitors.
Novel small molecules with defined mechanisms of inhibition against V-ATPase are needed to evaluate their therapeutic potential in cancer. Among many mechanisms that regulate the tumor microenvironment, V-ATPases are especially significant because they can be inhibited by proton pump inhibitors. The easiest method to design and develop new V-ATPase inhibitors is the screening processes. Drug screening can be performed using various in vitro assays. In vitro drug screening provides a quick, cost-effective method for the selection of potent V-ATPase inhibitors. Planning the course of effective screening mainly involves the selection of a proper set of compounds, built inhouse or obtained from commercially available libraries. In an effort to identify the new small molecule V-ATPase inhibitors, we used four known small molecule V-ATPase inhibitors to perform similarity searches against the University of Cincinnati, Drug Discovery Center (UC DDC) Library, consisting of 362,000 compounds, and the National Cancer Institute (NCI) Small Molecule Repository (SMR), with 263,365 compounds. Similarity searches were performed in Accelrys' Pipeline Pilot using 13 fingerprint types in parallel [16][17][18]. Compounds I and II were developed from the most promising member of the indole class of V-ATPase inhibitors, (2Z,4E)-5-(5,6-dichloro-2-indolyl)-2-methoxy-N-(1,2,2,6,6-pentamethylpiperidin-4-yl)-2,4-pentadienamide (SB 242784) [19,20]. Compound III, Enoxacin, is a fluoroquinolone antibiotic that was identified by structure-based virtual screening [21]. The natural product Diphyllin (IV) functions as a potent inhibitor of osteoclastic V-ATPase [22] ( Figure 1).

Similarity Search
We performed the similarity search process based on the four compounds listed above (I-IV; Figure 1). Initially, we selected approximately 429 compounds for the study. Subsets of 28 compounds were manually selected for anticancer screening using Lipinski's rules and a visual perception of relatedness to the compounds of interest. All 28 compounds were screened for a single human breast cancer cell line, MDA-MB-231, and identified the bisbenzimidazole derivative (compound V) as an initial hit ( Figure 2). The compound V served as a starting point in our program for finding potent new anticancer V-ATPase agents by designing and synthesizing a small set of bisbenzimidazole analogs. These new analogs were screened for antiproliferative activity using a panel of breast and ovarian cancer cell lines. The selected potent anticancer compounds were measured for proton pump activity. We performed the similarity search process based on the four compounds listed above (I-IV; Figure 1). Initially, we selected approximately 429 compounds for the study. Subsets of 28 compounds were manually selected for anticancer screening using Lipinski's rules and a visual perception of relatedness to the compounds of interest. All 28 compounds were screened for a single human breast cancer cell line, MDA-MB-231, and identified the bisbenzimidazole derivative (compound V) as an initial hit ( Figure 2). The compound V served as a starting point in our program for finding potent new anticancer V-ATPase agents by designing and synthesizing a small set of bisbenzimidazole analogs. These new analogs were screened for antiproliferative activity using a panel of breast and ovarian cancer cell lines. The selected potent anticancer compounds were measured for proton pump activity.

Chemical Synthesis
We quickly developed a fast and efficient synthetic procedure to prepare novel bisbenzimidazole analogs using the one pot procedure.

Antiproliferative Activity
Our research group is extensively involved in studying the role of vacuolar (H + )-ATPase in breast as well as ovarian cancer cell lines [23][24][25][26]. We examined the antiproliferative activity of the novel bisbenzimidazole derivatives  [27,28]. We employed the CellTiter 96 AQueous One solution cell proliferation assay (MTS; Promega) [29,30] to measure the in vitro cell viability of the cells.
Our initial hit (compound V, Figure 3) demonstrated a single digit micoromolar activity (IC50 range 0.72 ± 0.08-9.33 ± 0.69 μM) towards all of the cell lines tested. The newly prepared compounds (2a-e) have shown encouraging antiproliferative activity  in comparison with the positive control, Baf A1 ( Figure 9) (Table 1).
Our initial hit (compound V, Figure 3) demonstrated a single digit micoromolar activity (IC 50 range 0.72 ± 0.08-9.33 ± 0.69 µM) towards all of the cell lines tested. The newly prepared compounds (2a-e) have shown encouraging antiproliferative activity  in comparison with the positive control, Baf A1 ( Figure 9) (Table 1).  , respectively. Unfortunately, these compounds lack selectivity towards the normal breast epithelial cell line, MCF10A. We planned to introduce the hydrophobic end group with the two carbon linker (compound 2e), since the two carbon linker compounds were better than the three carbon linker compounds in the case of the hydrophilic end group compounds. To our surprise, compound 2e demonstrated a selective cytotoxicity towards only the MDA-MB-468 cell line (IC50 = 0.04 ± 0.02 μM), but it is not selective for other breast cancer cell lines. It is nearly 40 times less toxic than the normal breast epithelial cell line, MCF10A (IC50 = 1.62 ± 0.14 μM). The compound 2e showed good antiproliferative activity towards the cisplatin-resistant ovarian cancer cell line (Cis-A2780: IC50 = 1.34 ± 0.14 μM), whereas it showed moderate activity against cisplatin-sensitive ovarian cancer cell lines (A2780: IC50 = 2.77 ± 0.05 μM; PA-1: IC50 = 2.87 ± 0.15 μM). This indicates that it could be further developed for resistant cell lines to overcome drug resistance. With our quick SAR studies, we identified the highly potent anticancer agent, compound 2e.

Proton (H + ) Pump Activity
Since our hit compound (V) has been identified by the literature known V-ATPases, we hypothesized that these newly synthesized bisbenzimidazole derivatives (2a-e) will work as proton pump inhibitors (PPIs). To neutralize the acidic microenvironment, PPIs have been used to target the V-ATPase pumps present on the cell membrane. PPIs have been shown to be highly effective at inhibiting V-ATPases in vitro and displayed efficacy in animal models [31][32][33][34]. Therefore, we selected efficacious anticancer agents for the proton pump activity. We performed the standard proton pump activity using the acridine orange (AO) fluorescence quenching method [35]. We selected two highly efficacious compounds (2d and 2e) and our initial hit (V) for proton pump activity. Baf A1 is a known V-ATPase inhibitor and is used as a positive control. The proton pump activity of cells was measured as a decrease in the fluorescence intensity per min. The activity of cells in the presence of Baf A1 and the potent compounds (V, 2d, and 2e) was calculated as a percentage of the untreated control. The values represented are mean values of duplicate measurements ( Figure 10). Baf A1 (1 μM) showed a 25% inhibition of the H + -ATPase activity, whereas our hit compound (V) exhibited 30% at 10 μM. Our most efficacious anticancer compounds, (2d) and (2e), have shown 14% and 42% inhibition, respectively, at 10 μM.

Proton (H + ) Pump Activity
Since our hit compound (V) has been identified by the literature known V-ATPases, we hypothesized that these newly synthesized bisbenzimidazole derivatives (2a-e) will work as proton pump inhibitors (PPIs). To neutralize the acidic microenvironment, PPIs have been used to target the V-ATPase pumps present on the cell membrane. PPIs have been shown to be highly effective at inhibiting V-ATPases in vitro and displayed efficacy in animal models [31][32][33][34]. Therefore, we selected efficacious anticancer agents for the proton pump activity. We performed the standard proton pump activity using the acridine orange (AO) fluorescence quenching method [35]. We selected two highly efficacious compounds (2d and 2e) and our initial hit (V) for proton pump activity. Baf A1 is a known V-ATPase inhibitor and is used as a positive control. The proton pump activity of cells was measured as a decrease in the fluorescence intensity per min. The activity of cells in the presence of Baf A1 and the potent compounds (V, 2d, and 2e) was calculated as a percentage of the untreated control. The values represented are mean values of duplicate measurements ( Figure 10). Baf A1 (1 μM) showed a 25% inhibition of the H + -ATPase activity, whereas our hit compound (V) exhibited 30% at 10 μM. Our most efficacious anticancer compounds, (2d) and (2e), have shown 14% and 42% inhibition, respectively, at 10 μM.

Proton (H + ) Pump Activity
Since our hit compound (V) has been identified by the literature known V-ATPases, we hypothesized that these newly synthesized bisbenzimidazole derivatives (2a-e) will work as proton pump inhibitors (PPIs). To neutralize the acidic microenvironment, PPIs have been used to target the V-ATPase pumps present on the cell membrane. PPIs have been shown to be highly effective at inhibiting V-ATPases in vitro and displayed efficacy in animal models [31][32][33][34]. Therefore, we selected efficacious anticancer agents for the proton pump activity. We performed the standard proton pump activity using the acridine orange (AO) fluorescence quenching method [35]. We selected two highly efficacious compounds (2d and 2e) and our initial hit (V) for proton pump activity. Baf A1 is a known V-ATPase inhibitor and is used as a positive control. The proton pump activity of cells was measured as a decrease in the fluorescence intensity per min. The activity of cells in the presence of Baf A1 and the potent compounds (V, 2d, and 2e) was calculated as a percentage of the untreated control. The values represented are mean values of duplicate measurements ( Figure 10). Baf A1 (1 µM) showed a 25% inhibition of the H + -ATPase activity, whereas our hit compound (V) exhibited 30% at 10 µM. Our most efficacious anticancer compounds, (2d) and (2e), have shown 14% and 42% inhibition, respectively, at 10 µM.

Chemical General Information
All reagents were purchased from Sigma-Aldrich Chemical Co. The reactions were carried out in an argon atmosphere. Routine thin-layer chromatography (TLC) was performed on aluminum-backed Uniplates (Analtech, Newark, DE, USA). Melting points were determined on a Stuart™ melting point apparatus SMP10 (Sigma-Aldrich) and are uncorrected. 1 H and 13 C nuclear magnetic resonance (NMR) spectra were determined in DMSO-d6/or MeOD at 400 MHz and 100 MHz, respectively, using Agilent 400 MHz Spectrometer (Agilent Technologies, Santa Clara, CA, USA). Chemical shifts are reported as shifts (δ) in parts per million (ppm) relative to TMS. Mass spectra were collected on a Shimadzu LCMS-2020 electrospray/ion trap instrument (Shimadzu Scientific Instruments, Columbia, MD, USA) in positive modes. High Resolution Mass Spectrometry (HRMS) for 2a-e was carried out at the University of Illinois Mass Spectrometry Lab in the School of Chemical Sciences (Urbana, IL, USA). Yields refer to purified products.

Chemical General Information
All reagents were purchased from Sigma-Aldrich Chemical Co.

Cell Viability Assay
CellTiter 96 AQueous One Solution cell proliferation assay (MTS) [29,30] was used to measure the in vitro cell viability of cells, and the MTS purchased from Promega (Madison, WI, USA). Cells (5 × 10 3 ) were seeded into each well of 96-well plates and incubated overnight. Dimethyl Sulfoxide (DMSO) stock solutions of the compounds (V, 2a-e) were diluted in corresponding media and exposed to different concentrations for 48 h. All three independent experiments were performed in quadruplicates. Untreated cells were used as a negative control and data were presented as the mean ± standard deviation (S.D.). The absorbances at 492 nm were measured using a Microplate Reader (Synergy HT, Biotek Instruments, Winooski, VT, USA). The compound concentration that inhibited cell growth by 50% of the untreated control (IC 50 ) was calculated from the dose response curves constructed by plotting percentages of cell viability versus drug concentrations using nonlinear regression analysis by GraphPad Prism Software 7 (GraphPad Software, San Diego, CA, USA).

Measurement of the Proton (H + ) Pump Activity
The proton pump activity was assessed by using the AO fluorescence quenching method [35]. Bafilomycin A1 (Baf A1), AO, and nigericin were purchased from Sigma-Aldrich. Hank's Balanced Salt Solution (HBSS) was purchased from Gibco-Life Technologies (Carlsbad, CA, USA). Human breast cancer MDA-MB-231 (ATCC, HTB-26) cells were washed twice with Hank's Balanced Salt Solution (HBSS). The cells were re-suspended at a density of 5 × 10 6 cells per mL in the same buffer and placed on ice. H + uptake was initiated by the addition of MDA-MB-231 cells (10 µL; 0.1 × 10 6 ) to pre-warmed HBSS (2.4 mL) containing AO (6 µL; 30 µM) and placed in a cuvette. AO fluorescence was measured at an excitation wavelength of 495 nm and an emission wavelength of 540 nm using the LS-50B Luminescence Spectrometer (Perkin Elmer, Software: FL WinLab, Version 4.00.03). Nigericin (6 µL; 6 µM) was added to collapse the pH gradient and unquench the AO fluorescence (See Supplementary Figure S1: 1a and 1b). MDA-MB-231 cells (10 µL; 0.1 × 10 6 ) were treated with the standard V-ATPase inhibitor Baf A1 (1 µM), along with compounds V, 2d, and 2e (10 µM), for 20 min followed by addition of AO (6 µL; 30 µM) for the proton pump activity (See Supplementary Figure S2: 2A, 2B, 2C, and 2D). A decrease in the fluorescence intensities per min of Baf A1 and the potent compounds (V, 2d, and 2e) was calculated and is expressed as a percentage of the control (mean ± SD of two independent experiments) ( Figure 10).

Conclusions
In conclusion, starting from the virtual screening practice using known small molecule V-ATPase inhibitors, we identified a bisbenzimidazole analog (V) as an initial hit compound. Synthesis and anticancer screening of a focused set of novel bisbenzimidazole derivatives indicated that they were potent against selected breast and ovarian cancer cell lines in vitro. The selected bisbenzimidazole analogs (2d and 2e) have demonstrated selective potency towards the TNBC cell line, MDA-MB-468. The compound 2e potently inhibited the proton pump activity, which is largely responsible for the acidification, and it showed selective anticancer activity. To the best of our knowledge, the bisbenzimidazole pharmacophore has been identified as the first V-ATPase inhibitor in its class. Together, our findings demonstrated that the bisbenzimidazole analog (2e) could be further developed as an anticancer vacuolar (H + )-ATPase inhibitor for the treatment of breast cancer.