Discovery of Novel Symmetrical 1,4-Dihydropyridines as Inhibitors of Multidrug-Resistant Protein (MRP4) Efflux Pump for Anticancer Therapy

Despite the development of targeted therapies in cancer, the problem of multidrug resistance (MDR) is still unsolved. Most patients with metastatic cancer die from MDR. Transmembrane efflux pumps as the main cause of MDR have been addressed by developed inhibitors, but early inhibitors of the most prominent and longest known efflux pump P-glycoprotein (P-gp) were disappointing. Those inhibitors have been used without knowledge about the expression of P-gp by the treated tumor. Therefore the use of inhibitors of transmembrane efflux pumps in clinical settings is reconsidered as a promising strategy in the case of the respective efflux pump expression. We discovered novel symmetric inhibitors of the symmetric efflux pump MRP4 encoded by the ABCC4 gene. MRP4 is involved in many kinds of cancer with resistance to anticancer drugs. All compounds showed better activities than the best known MRP4 inhibitor MK571 in an MRP4-overexpressing cell line assay, and the activities could be related to the various substitution patterns of aromatic residues within the symmetric molecular framework. One of the best compounds was demonstrated to overcome the MRP4-mediated resistance in the cell line model to restore the anticancer drug sensitivity as a proof of concept.


Introduction
Cancer is an ongoing burden for humanity with increasing costs for therapy of patients [1][2][3]. Older chemotherapeutic agents affected all dividing human cells and had strong side effects [4,5]. However, besides tumor surgery as the best alternative, they were the only weapon to combat tumor growth. Meanwhile, specific targeting drugs have been developed which address tumor-specific structures [4]. They cause fewer side effects but have a limited use only for special kinds of cancer in single cases [4]. Most prominent drugs have been monoclonal antibodies, but they are expensive, and their benefit is limited due to a partial short elongation of life for the patients [6,7]. Protein kinase inhibitors are established alternatives with lower costs but increasing resistances as a result of single mutations that reduce the drug-binding affinities [8]. Inhibitors emerged with more than one protein kinase as a target structure [9]. However, these inhibitors are also affected by resistance mechanisms [10].
Besides the various resistance mechanisms that occur toward single drugs, multidrug resistance (MDR) is the main obstacle in cancer therapy [11,12]. Transmembrane efflux pumps are known as causative agents in most cases of MDR [11,13]. They pump anticancer drugs that were taken up by the cancer cells out of the cells so that the cell becomes resistant toward the treatment with the used drug [13]. Moreover, the used drug itself induces the 1, methyl propiolate 2, and a benzylamine compound 3 as shown in Scheme 1. Thus the synthesis is different from the known Hantzsch-synthesis of 1,4-dihydropyridines in that a dicarbonyl compound is used instead of methyl propiolate [44]. Our target compounds 4-21 own no substituents in the 2-and 6-positions where those formed with the dicarbonyl compound own alkyl substituents at the respective 1,4-dihydropyridine positions.
After finishing the reaction that was followed by thin-layer chromatography (TLC), the solvent was removed. The compound purification and final isolation led to solid products that were spectroscopically characterized to confirm the given structures with proton signals for the hydrogen atoms at the 2-and 6-positions at about 7.45 ppm and for the 4hydrogen atom at about 4.75 ppm in the 1

MRP4 Efflux Pump Inhibition with the 1,4-Dihydropyridines
However, due to the recent promotion of MRP4 to an anticancer target, there is a certain lack of MRP4 inhibitors. From those few inhibitors with low potency, the most prominent one is the cysteinyl leukotriene receptor 1 antagonist MK571 that shows additional pharmacological activities as a phosphodiesterase inhibitor [26,45]. Thus, novel inhibitors are necessary.
MRP4 has a symmetric molecular framework built of two transmembrane domains (TMDs) that each consist of six α-helical subunits surrounding the inner channel for outward transport of the MRP4 substrates [46,47]. The substrate-binding cavity for outward transport consists of amino acid residues with arranged subunits similar to the symmetric P-gp [47]. Therefore, we wondered whether our symmetric 1,4-dihydropyridines may serve as promising inhibitors of the MRP4 activity. In our assay system, we determined the uptake of the fluorescent MRP4 substrate calcein in the pancreatic colo357 and the colo357 MRP4-overexpressing cell lines both with and without our 1,4-dihydropyridines as potential inhibitors. In the case of inhibition of MRP4, an increase of the uptake of the fluorescent MRP4 substrate calcein was observed in the MRP4-overexpressing cell line. The fluorescence amount of the MRP4 substrate was measured for the inhibited cell lines where it increased only in the MRP4-overexpressing After finishing the reaction that was followed by thin-layer chromatography (TLC), the solvent was removed. The compound purification and final isolation led to solid products that were spectroscopically characterized to confirm the given structures with proton signals for the hydrogen atoms at the 2-and 6-positions at about 7.45 ppm and for the 4-hydrogen atom at about 4.75 ppm in the 1 H NMR spectra.

MRP4 Efflux Pump Inhibition with the 1,4-Dihydropyridines
However, due to the recent promotion of MRP4 to an anticancer target, there is a certain lack of MRP4 inhibitors. From those few inhibitors with low potency, the most prominent one is the cysteinyl leukotriene receptor 1 antagonist MK571 that shows additional pharmacological activities as a phosphodiesterase inhibitor [26,45]. Thus, novel inhibitors are necessary.
MRP4 has a symmetric molecular framework built of two transmembrane domains (TMDs) that each consist of six α-helical subunits surrounding the inner channel for outward transport of the MRP4 substrates [46,47]. The substrate-binding cavity for outward transport consists of amino acid residues with arranged subunits similar to the symmetric P-gp [47]. Therefore, we wondered whether our symmetric 1,4-dihydropyridines may serve as promising inhibitors of the MRP4 activity. In our assay system, we determined the uptake of the fluorescent MRP4 substrate calcein in the pancreatic colo357 and the colo357 MRP4-overexpressing cell lines both with and without our 1,4-dihydropyridines as potential inhibitors. In the case of inhibition of MRP4, an increase of the uptake of the fluorescent MRP4 substrate calcein was observed in the MRP4-overexpressing cell line. The fluorescence amount of the MRP4 substrate was measured for the inhibited cell lines where it increased only in the MRP4-overexpressing cell line due to the inhibition while in the non-inhibited cells fluorescence remained unchanged. A difference in fluorescence was calculated from these values for each cell line. The given values were divided to result in a fluorescence activity ratio (FAR). The FAR values are shown in Table 1. The higher the inhibition effect was, the higher were the resulting FAR values. We started with a meta trifluoromethyl substitution of the 4-phenyl residue and combined it with N-benzyl ring substitutions in derivatives 4-7. Both meta trifluoromethyl 4-phenyl and N-benzyl substituted compound 4 with a FAR value of 1.28 was much more active than the used MRP4 inhibitor MK571 for that a value of 0.82 was determined. When the meta trifluoromethyl substituent in the N-benzyl residue was replaced with a methoxy substituent, the activity of compound 5 was reduced. If that methoxy substituent moved to the para position in compound 6, a further reduction of activity was found. A combination of both a meta and para methoxy N-benzyl substitution in derivative 7 led to an almost unchanged activity.
If the trifluoromethyl substitution in the 4-phenyl residue moved to the para position, the activity of the resulting compound 8 was found to increase compared to derivative 5 with both the trifluoromethyl and methoxy substituent in the meta position of the aromatic residues. When the meta methoxy substitution in the N-benzyl substituent moved to the para position, the activity of compound 9 decreased. A combination of both meta and para methoxy N-benzyl substitutions in compound 10 resulted in a slightly decreased activity compared to the only meta substituted compound 8. It can be concluded that the meta substitution of the N-benzyl ring is more favorable than the para substitution with all compounds showing a better activity than MK571.
Next, we distributed the fluoro functions from the trifluoro substituent to the meta and para functions of the 4-phenyl residue in compounds 11-13. Combined with the meta methoxy function at the N-benzyl phenyl ring in compound 11, we reached an increase in activity compared to both the trifluoromethyl substituted derivatives 5 and 8. If the methoxy function moved to the para position of the N-benzyl residue, a slight decrease resulted for compound 12. A combination of a para und meta methoxy substitutions in the N-benzyl residue of derivative 13 also resulted in a favorable activity compared to the two trifluoromethyl compounds 7 and 10. The only meta fluoro 4-phenyl substituted derivative 14 and the just para fluoro substituted compound 15 were less active than the difluoro 4-phenyl compound 13. Therefore it can be concluded that the disubstitution of the 4-phenyl ring resulted in the best activities so far.
Next, we combined a para methoxy and a meta trifluoromethyl residue as disubstitution at the 4-phenyl substituent with the respective N-benzyl ring substituents of the first compounds 4-7. The meta trifluoromethyl N-benzyl substitution in derivative 16 resulted in the almost similar activity of both trifluoromethyl substitutions in compound 4. If the meta trifluoro substitution in the N-benzyl ring was replaced with a methoxy substituent, we found the main increases in activity for derivative 17 with a FAR value of 1.55. If that meta methoxy function moved to the para position in compound 18, we found a decreased activity, and both meta and para methoxy substituted compound 19 resulted in an activity lower than that of derivative 17 and better than that of compound 18. Again, the disubstitution of the 4-phenyl ring resulted in improved activities with the meta substitution in the N-benzyl ring being more favorable than the para position.
Finally, we investigated a dimethoxy function in the 4-phenyl ring and combined it with the meta trifluoro substituent of the N-benzyl phenyl ring in compound 20. Compared to derivative 16, the activity was found to decrease. When the dimethoxy substitution was replaced with a dibenzyloxy function, the activity of compound 21 was found to increase.
Generally, the disubstituted 4-phenyl compounds with a meta substitution in the N-benzyl phenyl ring resulted in the best MRP4 inhibiting activities. In order to gain a first insight into an MRP4-specific inhibition effect, we determined the corresponding FAR values for the best MRP4-inhibiting compounds 11, 12, 17, and 21 in an MRP1-overexpressing ovarian carcinoma cell line [48]. Compared to the used MRP1 standard probenecid for which a FAR value of 1.23 was determined at a used concentration of 10 µM, we found partly mainly reduced values of 0.92 for compound 11, 0.97 for compound 12, 1.01 for compound 17, and finally 0.82 for compound 20. If compared to the compounds MRP4 activities in relation to the MK571 standard that means just a low activity toward MRP1.

In Vitro MRP4 Resistance Studies of Drug Reversal
MRP4 has been associated with various kinds of cancer due to an observed overexpression that was mostly based on determined mRNA analysis. Described compound effects on MRP4 efflux inhibition are rare and mostly limited to drugs that were discovered from compound libraries with pharmacological properties different from the MRP4 inhibition, therefore a perspective use for therapy was excluded [23,37]. As recently reported for efflux pump inhibitors for MDR cancer therapy, their use will be effective if the respective efflux pump is expressed by the tumor tissue to be really blocked by the inhibitor so that the anticancer drug resistance mediated by that efflux pump can be overcome.
In order to perspectively profile our novel compound class of effective MRP4 inhibitors, we further evaluated the inhibitor ability to resensitize the colo357 MRP4-overexpressing cell line toward a cytostatic drug as MRP4 substrate. We used 6-mercaptopurine as antimetabolite in cancer therapy and MRP4 substrate [25] and determined the toxicity of 6-mercaptopurine in the colo357 and the MRP4-overexpressing cell line in an MTT assay to detect the cellular toxicity as a result of damage of the mitochondrial activity of MTT reduction.
In the anticancer drug-sensitive cell line colo357, we determined an IC 50 value of 24.88 µM for 6-mercaptopurine. In the MRP4-overexpressing cell line, the IC 50 value was determined with 224.0 µM which means a loss of sensitivity for 6-mercaptopurine. Then we determined the toxicity in the MRP4-overexpressing cell line with the use of 10 µM of the MRP4 inhibitor MK571. The IC 50 value of 6-mercaptopurine was determined with 313.6 µM which practically means no effect of an efflux pump inhibition that would have reduced the IC 50 values of mercaptopurine. Then we used our best MRP4 inhibitor compound 17 from our uptake assay studies. We determined an IC 50 value for mercaptopurine of 33.73 µM. That IC 50 value almost reached the value determined for mercaptopurine in the non-MRP4-expressing colo357 cell line. Cellular toxic effects of compound 17 itself and the other derivatives could be excluded by previous studies of FACS analyses with the used compound preincubated cell lines that would have indicated such cellular effects in the cell sorting process.
Therefore we found that the anticancer drug sensitivity was restored by the use of our best inhibitor as a proof of principle.

Chemical Reagents and Instruments
Commercial reagents were used without further purification. The 1 H-NMR spectra (500 MHz) were measured using tetramethylsilane as an internal standard. Thin-layer chromatography (TLC) was performed on E. Merck 5554 silica gel plates. The highresolution mass spectra were recorded on a Finnigan LCQ Classic mass spectrometer (Mudelein, IL, USA).

General Procedure for the Synthesis of Compounds 4-21
One equivalent of the aromatic aldehyde 1, two equivalents of the methyl propiolate 2, and one equivalent of the benzylamine 3 were dissolved in 2 mL of freshly distilled acetic acid and heated under reflux under stirring at 80 • C. The reaction procedure was followed by TLC and mass spectrometry. After finishing, a brine solution was added, and the mixture was extracted with chloroform several times. The unified organic layer was reduced in volume, and residual acetic acid was removed with toluene. Then methanol or a mixture of diethyl ether and ethanol was added under cooling to give the final product under crystallization. In the case of non-crystallizing, the reaction mixture had to be purified by column chromatography using silica gel and eluent mixtures of cyclohexane and ethyl acetate. The unified compound-containing fractions were evaporated, and again methanol or a mixture of diethyl ether and methanol was added and the purified compound crystallized.

MRP Inhibition Assay
The human pancreatic carcinoma cell lines colo357 and colo357 MRP4 that resulted from transfection of the colo357 cells with an MRP4 vector construct were used [49], and the MRP4 overexpression was confirmed by Western blot as shown in the Supplementary Materials. Both cell lines and the ovarian carcinoma cell line A2780 were cultured in RPMI-1640 medium that was supplemented with fetal calf serum (10%), minimal essential amino acids (1%), and penicillin/streptomycin (1%) at 37 • C and under carbon dioxide atmosphere (5%).
In the assay, every 200.000 cells were given in an Eppendorf tube. The cells were centrifuged at 2000 RPM at 4 • C. The supernatant was removed, and the samples were stored on ice. Then they were resuspended in RPMI-1640 medium, and test compounds and MK571 or probenecid control were added from stock solutions of 4000 µM in DMSO to give a final concentration of 10 µM. DMSO was used as control with no effect on the fluorescence. The samples were cultured at 37 • C for 20 min and 1200 RPM in a thermomixer. Then the fluorescent calcein was added from a PBS solution to give a final concentration of 0.005 µM in the colo cells, and the fluorescent CFDA was added from a PBS solution with the same final concentration to the ovarian cells. The samples were centrifuged again, and the supernatant was removed. Then PBS was added, and the samples were centrifuged again. The washing procedure was repeated. Finally, the fluorescence of the resuspended cells was measured by flow cytometry using 10.000 cells and a MACSQuant Analyzer (Bergisch Gladbach, Germany). The measurement was conducted three times each for inhibitortreated and untreated cells of both cell lines. The MRP4 FAR values were calculated as the ratio of the fluorescence of the treated colo357 MRP4 to the treated colo357 cells with each value corrected by the fluorescence of the untreated colo cells. The MRP1 FAR values were calculated as the ratio of the fluorescence of the treated ovarian cells related to the untreated control cells.

MRP4 Reversal Assay
10.000 cells of each cell line were cultured in wells of a 96-well plate at 37 • C under a carbon dioxide atmosphere (5%) for 24 h. Increasing concentrations of the MRP4 substrate 6-mercaptopurine from 0.0001 µM to finally 100 µM were added after the test compound and MK571, each at a concentration of 10 µM, had been added. The plate was incubated for 48 h under the original culture conditions. Then the MTT reagent was added to each well (10 µL of a stock solution of 5 mg/mL in PBS) and incubation continued for 4 h. Then 100 µL DMSO was added to each well to solve the formazan reduction product. The plate was shaken for 30 min on a plate shaker and, finally, the formazan absorption was measured. The described method was repeated three times. The IC 50 values were determined from the resulting sigmoid curves.

Conclusions
MDR is the main problem in anticancer treatment, especially in the case of metastatic cancer that generally cannot be treated as it would afford the effectiveness of various anticancer drugs. As novel anticancer drugs are specific to single cancer drug targets, their use is limited to certain kinds of cancer but not for the treatment of MDR. Moreover, these novel drugs also become affected by the MDR phenomenon. The aim to block the MDR by the use of effective drugs is still a challenge. If a respective tumor expresses a transmembrane efflux pump as the main cause of the MDR phenomenon, the use of an inhibitor will treat the MDR and resensitize the tumor toward the anticancer drug treatment.
We developed a novel class of MRP4 inhibitors that was evaluated to block the MRP4 activity toward the fluorescent MRP4 substrate calcein. Certain substituent-dependent effects were observed with best activities for a disubstituted 4-phenyl residue and a meta substituted N-benzyl ring residue at the symmetric molecular scaffold. All compounds showed a mainly better activity than the used MRP4 inhibitor MK571. In order to further profile our compound class, the best inhibitor was investigated to restore the MRP4mediated anticancer drug toxicity in the used MRP4-overexpressing cell line model. The use of that compound almost reached the anticancer drug toxicity determined in the non-MRP4-expressing cell line. Thus our novel compound class is promising for further studies to perspectively combat MRP4-mediated anticancer drug resistance.