Chiral Thioxanthones as Modulators of P-glycoprotein: Synthesis and Enantioselectivity Studies

Recently, thioxanthone derivatives were found to protect cells against toxic P-glycoprotein (P-gp) substrates, acting as potent inducers/activators of this efflux pump. The study of new P-gp chiral modulators produced from thioxanthone derivatives could clarify the enantioselectivity of this ABC transporter towards this new class of modulators. The aim of this study was to evaluate the P-gp modulatory ability of four enantiomeric pairs of new synthesized chiral aminated thioxanthones (ATxs) 1–8, studying the influence of the stereochemistry on P-gp induction/ activation in cultured Caco-2 cells. The data displayed that all the tested compounds (at 20 μM) significantly decreased the intracellular accumulation of a P-gp fluorescent substrate (rhodamine 123) when incubated simultaneously for 60 min, demonstrating an increased activity of the efflux, when compared to control cells. Additionally, all of them except ATx 3 (+), caused similar results when the accumulation of the P-gp fluorescent substrate was evaluated after pre-incubating cells with the test compounds for 24 h, significantly reducing the rhodamine 123 intracellular accumulation as a result of a significant increase in P-gp activity. However, ATx 2 (−) was the only derivative that, after 24 h of incubation, significantly increased P-gp expression. These results demonstrated a significantly increased P-gp activity, even without an increase in P-gp expression. Therefore, ATxs 1–8 were shown to behave as P-gp activators. Furthermore, no significant differences were detected in the activity of the protein when comparing the enantiomeric pairs. Nevertheless, ATx 2 (−) modulates P-gp expression differently from its enantiomer, ATx 1 (+). These results disclosed new activators and inducers of P-gp and highlight the existence of enantioselectivity in the induction mechanism.


Introduction
P-glycoprotein (P-gp) is the best characterized efflux transporter belonging to the superfamily of ATP-binding cassette (ABC) transporters [1][2][3][4]. The structure of P-gp complies with the prototype characteristics described for the ABC transporters, being a polypeptide made up of two homologous halves that arise from a gene duplication event [5][6][7]. Each half contains a transmembrane domain for mefloquine, a drug used in the prevention and treatment of malaria: (11R,12S)-(+)-mefloquine presents a much higher human brain penetration than (11S,12R)-(−)-mefloquine [55].
Herein, we investigated the potential modulatory effect on P-gp activity and expression of four newly synthetized pairs of enantiomers of aminated thioxanthones (ATxs 1-8, Figure 1), to clarify the enantioselectivity of this efflux pump towards the ABC transporters class. Moreover, although some examples of biological enantioselectivity were reported in the literature for structurally related xanthones [56][57][58], for thioxanthones, and specifically for P-gp modulation, data is presented here for the first time.
Molecules 2018, 23, x 3 of 20 (11R,12S)-(+)-mefloquine presents a much higher human brain penetration than (11S,12R)-(−)-mefloquine [55]. Herein, we investigated the potential modulatory effect on P-gp activity and expression of four newly synthetized pairs of enantiomers of aminated thioxanthones (ATxs 1-8, Figure 1), to clarify the enantioselectivity of this efflux pump towards the ABC transporters class. Moreover, although some examples of biological enantioselectivity were reported in the literature for structurally related xanthones [56][57][58], for thioxanthones, and specifically for P-gp modulation, data is presented here for the first time.

Synthesis of Thioxanthones
The new chiral ATxs 1-8 were synthesized by copper-catalyzed Ullmann cross-coupling reactions between the thioxanthone derivative 1-chloro-4-propoxy-9H-thioxanthen-9-one (Tx) and eight enantiomerically pure amino alcohols 9-16 in alkaline medium (Scheme 1). The amino alcohols selected for this study fitted on the pharmacophore model previously built for P-gp activators and their design was based on a hit compound found as a P-gp activator/inducer, 1-[(3-hydroxypropyl)amino]-4-propoxy-9H-thioxanthen-9-one, a thioxanthone with an amino alcohol chain at position 1 [30]. Therefore, we decided to modify this amino alcohol chain introducing also a stereogenic center to explore the influence of chirality in this class of derivatives.

Synthesis of Thioxanthones
The new chiral ATxs 1-8 were synthesized by copper-catalyzed Ullmann cross-coupling reactions between the thioxanthone derivative 1-chloro-4-propoxy-9H-thioxanthen-9-one (Tx) and eight enantiomerically pure amino alcohols 9-16 in alkaline medium (Scheme 1). The amino alcohols selected for this study fitted on the pharmacophore model previously built for P-gp activators and their design was based on a hit compound found as a P-gp activator/inducer, 1-[(3-hydroxypropyl)amino]-4-propoxy-9H-thioxanthen-9-one, a thioxanthone with an amino alcohol chain at position 1 [30]. Therefore, we decided to modify this amino alcohol chain introducing also a stereogenic center to explore the influence of chirality in this class of derivatives. Initial investigations of the catalyst to be used in the Ullman asymmetric reactions revealed copper iodide as a more efficient catalyst when compared to copper oxide used previously in the synthesis of aminated thioxanthones [30].
The purity of each synthesized chiral thioxanthone was determined by high performance liquid chromatography with diode-array detection (HPLC-DAD) analysis. All tested compounds presented a purity of at least 95%. Their enantiomeric purity was also measured by chiral HPLC using an amylose-derived column under normal-phase or polar organic mode, achieving enantiomeric excess (e.e.) values higher than 99%, as exemplified in Figure 2 for the enantiomeric pair ATx 1 (+) and ATx 2 (−). Initial investigations of the catalyst to be used in the Ullman asymmetric reactions revealed copper iodide as a more efficient catalyst when compared to copper oxide used previously in the synthesis of aminated thioxanthones [30].
The structure elucidation of all new thioxanthonic derivatives (ATxs 1-8) was established by spectroscopic methods: infrared (IR), 1 H-and 13 C-nuclear magnetic resonance (NMR), and bidimensional heteronuclear single quantum correlation (HSQC) and heteronuclear multiple bond correlation (HMBC). Since these determinations were performed in achiral environments, no significant differences were noted for each pair of enantiomers obtained, as expected. The 13 C-NMR assignments were made by HSQC and HMBC experiments and by comparison with assignments of similar molecules [25,58]. The HMBC data were important to deduce the chemical shifts of quaternary carbons and confirm the C-N coupling product.

Flow Cytometry Analysis of P-gp Expression
The structure elucidation of all new thioxanthonic derivatives (ATxs 1-8) was established by spectroscopic methods: infrared (IR), 1 H-and 13 C-nuclear magnetic resonance (NMR), and bidimensional heteronuclear single quantum correlation (HSQC) and heteronuclear multiple bond correlation (HMBC). Since these determinations were performed in achiral environments, no significant differences were noted for each pair of enantiomers obtained, as expected. The 13 C-NMR assignments were made by HSQC and HMBC experiments and by comparison with assignments of similar molecules [25,58]. The HMBC data were important to deduce the chemical shifts of quaternary carbons and confirm the C-N coupling product.
The structure elucidation of all new thioxanthonic derivatives (ATxs 1-8) was established by spectroscopic methods: infrared (IR), 1 H-and 13 C-nuclear magnetic resonance (NMR), and bidimensional heteronuclear single quantum correlation (HSQC) and heteronuclear multiple bond correlation (HMBC). Since these determinations were performed in achiral environments, no significant differences were noted for each pair of enantiomers obtained, as expected. The 13 C-NMR assignments were made by HSQC and HMBC experiments and by comparison with assignments of similar molecules [25,58]. The HMBC data were important to deduce the chemical shifts of quaternary carbons and confirm the C-N coupling product.

Flow Cytometry Analysis of P-gp Expression
The ability of the tested chiral ATx's 1-8 to induce P-gp expression in Caco-2 cells was evaluated by flow cytometry, using a P-gp monoclonal antibody (UIC2) conjugated with phycoerythrin (PE), 24 h after exposure. As shown in Figure 4, ATx 2 (−) significantly increased P-gp expression by 36%, when compared to control cells. For the other tested chiral thioxanthones, no significant effect on P-gp expression was observed after 24 h of incubation. Furthermore, significant differences were detected in the P-gp induction mediated by the enantiomeric pair ATx 1 (+) : ATx 2 (−), demonstrating an enantioselectivity for P-gp induction. The ability of the tested chiral ATx's 1-8 to induce P-gp expression in Caco-2 cells was evaluated by flow cytometry, using a P-gp monoclonal antibody (UIC2) conjugated with phycoerythrin (PE), 24 h after exposure. As shown in Figure 4, ATx 2 (−) significantly increased P-gp expression by 36%, when compared to control cells. For the other tested chiral thioxanthones, no significant effect on P-gp expression was observed after 24 h of incubation. Furthermore, significant differences were detected in the P-gp induction mediated by the enantiomeric pair ATx 1 (+) : ATx 2 (−), demonstrating an enantioselectivity for P-gp induction. . Flow cytometry analysis of P-gp expression levels in Caco-2 cells exposed to ATxs 1-8 (20 μM) for 24 h. Results are presented as mean ± standard error mean (SEM) from 5 independent experiments (performed in duplicate). Statistical comparisons were made using One-way ANOVA, followed by Tukey's multiple comparisons test (*** p < 0.001 vs. control (0 μM); # p < 0.05 for ATx 1 (+) vs. ATx 2 (−)).

Evaluation of P-gp Transport Activity
The P-gp transport activity was evaluated by flow cytometry using rhodamine (RHO) 123, a P-gp fluorescent substrate [25,31]. Two different approaches were used. In the first approach, RHO 123 accumulation was evaluated in the presence of the tested chiral ATxs 1-8 (20 μM) during the RHO 123 accumulation phase that lasted 60 min, in order to evaluate the potential immediate effects of the tested compounds on P-gp activity as a result of a direct activation of the pump. P-gp activity was evaluated through the ratio between the mean intracellular RHO 123 fluorescence intensity (MFI) obtained under P-gp inhibition (IA) and the MFI observed in normal conditions (NA, in the absence of P-gp inhibition) (Equation (1)), and expressed as percentage of control cells (0 μM chiral ATxs).

Evaluation of P-gp Transport Activity
The P-gp transport activity was evaluated by flow cytometry using rhodamine (RHO) 123, a P-gp fluorescent substrate [25,31]. Two different approaches were used. In the first approach, RHO 123 accumulation was evaluated in the presence of the tested chiral ATxs 1-8 (20 µM) during the RHO 123 accumulation phase that lasted 60 min, in order to evaluate the potential immediate effects of the tested compounds on P-gp activity as a result of a direct activation of the pump. P-gp activity was evaluated through the ratio between the mean intracellular RHO 123 fluorescence intensity (MFI) obtained under P-gp inhibition (IA) and the MFI observed in normal conditions (NA, in the absence of P-gp inhibition) (Equation (1)), and expressed as percentage of control cells (0 µM chiral ATxs).

Determination of ATPase Activity
The effect of chiral ATxs 1-8 on the ATPase activity of P-gp was evaluated through the detection and quantification of the unmetabolized ATP, as a luciferase-generated luminescent signal, in human P-gp-enriched membranes. P-gp-dependent decreases in luminescence reflect ATP consumption by P-gp and, consequently, a decrease in the obtained signal reflects the presence of a compound that stimulates P-gp ATPase activity.
In the present study, Na 3 VO 4 (sodium orthovanadate) was used as a selective P-gp inhibitor and, thus, samples treated with Na 3 VO 4 have no P-gp ATPase activity. Consequently, the difference in luminescent signal between Na 3 VO 4 -treated samples and untreated samples (∆RLU basal ) reflects the basal P-gp ATPase activity, whereas the difference in luminescent signal between Na 3 VO 4 -treated samples and samples treated with the test compound (TC) (∆RLU TC ) represents the P-gp ATPase activity in the presence of the test compound. Therefore, the test compound can be a stimulator of P-gp ATPase activity if ∆RLU TC > ∆RLU basal ; can be an inhibitor of P-gp ATPase activity if ∆RLU TC < ∆RLU basal ; or can have no effect on P-gp ATPase activity if ∆RLU TC = ∆RLU basal . Verapamil (Ver) was used in the present assay as a positive control since it is a substrate for P-gp-mediated transport, thus stimulating P-gp ATPase activity. The results for the eight tested compounds and Ver are shown in Figure 7, as well as the basal P-gp ATPase activity. None of the tested compounds (ATx's 1-8) caused a significant change in the ∆RLU when comparing to the basal P-gp ATPase activity, demonstrating not being substrates for this efflux pump. The effect of chiral ATxs 1-8 on the ATPase activity of P-gp was evaluated through the detection and quantification of the unmetabolized ATP, as a luciferase-generated luminescent signal, in human P-gp-enriched membranes. P-gp-dependent decreases in luminescence reflect ATP consumption by P-gp and, consequently, a decrease in the obtained signal reflects the presence of a compound that stimulates P-gp ATPase activity.
In the present study, Na3VO4 (sodium orthovanadate) was used as a selective P-gp inhibitor and, thus, samples treated with Na3VO4 have no P-gp ATPase activity. Consequently, the difference in luminescent signal between Na3VO4-treated samples and untreated samples (ΔRLUbasal) reflects the basal P-gp ATPase activity, whereas the difference in luminescent signal between Na3VO4-treated samples and samples treated with the test compound (TC) (ΔRLUTC) represents the P-gp ATPase activity in the presence of the test compound. Therefore, the test compound can be a stimulator of P-gp ATPase activity if ΔRLUTC > ΔRLUbasal; can be an inhibitor of P-gp ATPase activity if ΔRLUTC < ΔRLUbasal; or can have no effect on P-gp ATPase activity if ΔRLUTC = ΔRLUbasal. Verapamil (Ver) was used in the present assay as a positive control since it is a substrate for P-gp-mediated transport, thus stimulating P-gp ATPase activity. The results for the eight tested compounds and Ver are shown in Figure 7, as well as the basal P-gp ATPase activity. None of the tested compounds (ATx's 1-8) caused a significant change in the ΔRLU when comparing to the basal P-gp ATPase activity, demonstrating not being substrates for this efflux pump. Results are expressed as mean ± SEM from three independent experiments (performed in triplicate). Statistical comparisons were estimated using One-way ANOVA, followed by Tukey's multiple comparisons test (** p < 0.01 vs. basal P-gp ATPase activity).

Discussion
All chiral aminated thioxanthones, ATxs 1-8, were synthesized in enantiomerically pure form in order to study the effect of the stereochemistry in P-gp modulation. The reaction conditions used in the Ullmann cross-coupling reaction between a suitable functionalized thioxanthone with enantiomerically pure building blocks allowed a successful synthesis of the desired chiral thioxanthones in moderate yields. The purification process applied allowed us to obtain the desired compounds (ATxs 1-8) with a purity of at least 95% and e.e. higher than 99%. Sufficient amounts were obtained to proceed with the evaluation of the biological activity. Structure elucidation of ATxs 1-8 was confirmed by spectroscopic methods. The e.e. values obtained by chiral HPLC evaluation as well as the absolute values of specific rotation confirmed that no racemization occurred under these reaction conditions.
Caco-2 cells express P-gp at levels similar to those found in normal human jejunum [62] and they were already validated as a suitable in vitro model for the evaluation of P-gp modulation [30][31][32]52]. Previous studies demonstrated that newly synthetized thioxanthone derivatives significantly Results are expressed as mean ± SEM from three independent experiments (performed in triplicate). Statistical comparisons were estimated using One-way ANOVA, followed by Tukey's multiple comparisons test (** p < 0.01 vs. basal P-gp ATPase activity).

Discussion
All chiral aminated thioxanthones, ATxs 1-8, were synthesized in enantiomerically pure form in order to study the effect of the stereochemistry in P-gp modulation. The reaction conditions used in the Ullmann cross-coupling reaction between a suitable functionalized thioxanthone with enantiomerically pure building blocks allowed a successful synthesis of the desired chiral thioxanthones in moderate yields. The purification process applied allowed us to obtain the desired compounds (ATxs 1-8) with a purity of at least 95% and e.e. higher than 99%. Sufficient amounts were obtained to proceed with the evaluation of the biological activity. Structure elucidation of ATxs 1-8 was confirmed by spectroscopic methods. The e.e. values obtained by chiral HPLC evaluation as well as the absolute values of specific rotation confirmed that no racemization occurred under these reaction conditions.
Caco-2 cells express P-gp at levels similar to those found in normal human jejunum [62] and they were already validated as a suitable in vitro model for the evaluation of P-gp modulation [30][31][32]52]. Previous studies demonstrated that newly synthetized thioxanthone derivatives significantly increased both P-gp expression and activity in Caco-2 cells, 24 h after exposure, being the first report on the ability of such compounds to act as P-gp inducers [30]. Moreover, although it is known that increases in protein expression may not necessarily result in proportional increases in the pump activity [31,60,61], for these compounds, the detected increases in the cell surface P-gp expression (since the UIC2 monoclonal antibody used in the experiments recognizes an external P-gp epitope) were accompanied by similar increases in its transport activity [30]. An important aspect to note among the obtained data was the ability of all the tested chiral thioxanthones to rapidly and significantly increase P-gp activity, without interfering with its expression, an effect compatible with a P-gp activation phenomenon [30].
In the present study, ATxs 1-8 also demonstrated to be P-gp activators since they were able to significantly increase P-gp transport function without interfering with the protein expression levels, as demonstrated by the data obtained in the RHO 123 accumulation assay performed in the presence of the chiral thioxanthones during a short 60 min accumulation phase. Concerning their structure, a chiral center further away of the thioxanthone moiety seems to prejudice the P-gp activation effect in this series of compounds. The RHO 123 accumulation evaluated using this protocol does not reflect a possible contribution of an increased P-gp expression in the increased activity due to the short duration of the contact between the chiral thioxanthones and the cells during the RHO 123 accumulation phase.
Furthermore, and contrarily to what was observed in the previously mentioned study [30], in the present work, the data obtained for the eight tested compounds demonstrated that, although only chiral ATx 2 (−) caused a significant increase in P-gp expression 24 h after exposure, all the tested compounds significantly increased P-gp activity when RHO 123 accumulation was evaluated 24 h after exposure, except for ATx 3 (+). In fact, and as previously mentioned, it is known that increases in protein expression may not necessarily result in proportional increases in pump activity, and that increases in activity can be observed independently of the level of expression. Indeed, although chiral ATxs 1 (+), 4 (−), 5 (+), 6 (−), 7 (+) and 8 (−) were not able to significantly increase P-gp expression 24 h after exposure, a significant increased P-gp activity was observed in Caco-2 cells pre-exposed to the tested compounds for this incubation period. Therefore, since these compounds demonstrated to be P-gp activators (as demonstrated by the significant differences observed in the RHO 123 accumulation assay performed with the compounds present only during the short RHO 123 accumulation phase), the observed increases in the pump activity must result from a direct pump activation, instead of an increased P-gp expression. The observed P-gp activation may be caused by the compound that remains intracellularly, once the cells are pre-exposed to the tested compounds for 24 h, and washed (to remove the tested compounds) prior to the evaluation of P-gp activity. In what concerns to P-gp ATPase activity, in contrast to previous findings [30], no significant differences were observed between the P-gp ATPase activity in the presence of the investigated thioxanthones and the basal P-gp ATPase activity. These ATxs 1-8 with a chiral center in the α-position of the aniline moiety seem not to affect the P-gp ATPase activity, and are unlikely to be substrates for P-gp-mediated transport, since P-gp substrates, like verapamil, typically stimulate its ATPase activity.
Although, some examples of enantioselective modulation of P-gp with other class of compounds have been described [54,55,63,64], no studies were reported focusing the enantioselectivity of chiral thioxanthones in modulating P-gp. Comparing the results obtained for the studied enantiomeric pairs (ATxs 1 (+) and 2 (−), ATxs 3 (+) and 4 (−), ATxs 5 (+) and 6 (−), and ATxs 7 (+) and 8 (−)), no significant differences were observed in P-gp activity. In contrast, for the P-gp induction effect for the chiral pair ATxs 1 (+) and 2 (−), significant differences were found. Noteworthy, a chiral center with R configuration closer to the thioxanthone moiety and with less hindered substituents seems to be critical for the induction effect in this series of compounds. This result highlights the existence of enantioselectivity in what concerns to induction mechanism, which is not a consequence of a direct interaction with P-gp. Previously, a similar behavior was described for the H1 anti-histaminic cetirizine, which was also ascribe to effects on P-gp expression [54]; the (R)-cetirizine upregulates P-gp expression, while (S)-cetirizine down-regulates it [54,65]; although, this library of compounds did not display any direct stereoselective modulation of P-gp, as previously observed for other substrates [54,55,63,64].

General Information
All reagents and solvents were purchased from Sigma Aldrich Co. (St. Louis, MO, USA), and no further purification was used. The amine building blocks 9-16 were (S)-(+)-amino-1-propanol (9) HPLC analysis were performed on a Finnigan Surveyor (Thermo Electron Corporation, Cleveland, OH, USA) equipped with an autosampler (AutoSampler Plus) and a TSP UV8000LP diode array detector (Thermo Electron Corporation). The treatment of the chromatographic data was performed using the Xcalibur ® 2.0 SUR1 software (Thermo Electron Corporation). For purity evaluation, the chromatographic separation was carried out on 250 × 4.6 mm i.d. FortisBIO C18 column (5 µm) with the mobile phase methanol:water:trimethylamine (80:20:0.5 v/v/v). The mobile phase was prepared in a volume/volume relation and degassed in an ultrasonic bath for 15 min before use. The flow rate was 1.0 mL/min. Aliquots of 20 µL of each synthesized chiral thioxanthone (ATxs 1-8) at the concentration of 20 µg/mL were injected. For e.e. evaluation, the chromatographic column used was a 250 × 4.6 mm i.d. Lux ® Amylose-1 (5 µm) from Phenomenex (Torrance, CA, USA). Analyses were performed in isocratic mode, at 22 ± 2 • C, in normal phase, using a mixture of n-hexane:ethanol (70:30 v/v) as mobile phase, or polar organic elution mode, using methanol or acetonitrile as mobile phase. The flow rate used was 0.5 mL/min. The UV detection was set at a wavelength of 254 nm and 20 µL of sample were injected in triplicate. The e.e. was determined by the relative percentages of the peak areas  (ATxs 1-8) 1-Chloro-4-propoxy-9H-thioxanthen-9-one (Tx, 450 mg, 1.48 mmol) and a suitable chiral amino alcohol (9-16, 1.72 mmol) were dissolved in methanol (30 mL) and CuI (0.15 mmol) and K 2 CO 3 (1.92 mmol) were added. The reaction mixture was heated at 100 • C in a muffle furnace for 48 h. After the completion of the reaction, the crude material was filtrated, washed with dichloromethane, and the organic solvents were evaporated under reduced pressure. Then, the obtained solid was dissolved in 50 mL of dichloromethane and extracted with HCl 1 M (3 × 50 mL). The aqueous layer was basified with NaOH 20% and extracted with dichloromethane (3 × 100 mL). The organic layers were gathered, washed with water (3 × 50 mL), dried over anhydrous sodium sulphate and the solvent was evaporated under reduced pressure. Then, a solid phase extraction using a cation exchange cartridge Discovery ® DSC-SCX was applied to further purify the extracted material. First, an activation of the cartridge with dichloromethane (50 mL) was performed followed by loading the cartridge with the sample (previously incorporated in silica). Then, elution was carried out with the following solvents: dichloromethane, a mixture of dichloromethane/methanol 5:5 (v/v), methanol 100% and NH 3 2% in methanol. The fractions containing the chiral ATxs were gathered and the solvent was evaporated under reduced pressure. A flash column chromatography with n-hexane/ethyl acetate in gradient and crystallization from chloroform and petroleum ether (4:1) were also performed to obtain pure compounds.   2875, 1606, 1566, 1556, 1501, 1451, 1434, 1328, 1286, 1270, 1249, 1227, 1158, 1133, 1102, 1060

Caco-2 Cell Culture
For the in vitro assays, the Caco-2 cell model was used, which derived from human colorectal adenocarcinoma. The cells were routinely cultured in 75 cm 2 flasks and maintained in a 5% CO 2 -95% air atmosphere, at 37 • C. Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 100 µM nonessential amino acids (NEAA), 100 U/mL penicillin, 100 µg/mL streptomycin and 6 µg/mL transferrin was used and changed every 2 to 3 days. Cultures were passaged weekly by trypsinization (0.25% trypsin/1 mM EDTA). For all the experiments, the cells were taken between the 27th and 36th passages. The cells were always seeded at a density of 60,000 cells/cm 2 and used 3 days after seeding, when confluence was reached. For the cytotoxicity assays the cells were seeded in 96-well plates, while for the evaluation of P-gp expression and P-gp transport activity, the cells were seeded in 12-well plates.

Compounds Cytotoxicity Assays
The cytotoxicity of the chiral ATxs 1-8 (0-50 µM) was accessed by the NR uptake assay, 24 h after exposure. The NR uptake assay provides a quantitative estimation of the number of viable cells in a culture based on the ability of viable cells to incorporate and bind the supravital dye NR in the lysosomes [30]. For that purpose, the cells were seeded onto 96-well plates and exposed to the eight tested ATxs 1-8 (0-50 µM) in fresh cell culture medium. After a 24 h incubation period, the cells were incubated with NR (40 µg/mL in cell culture medium) at 37 • C, in a humidified, 5% CO 2 -95% air atmosphere, for 60 min. The cell culture was then removed, the dye absorbed only by viable cells extracted (with absolute ethyl alcohol/distilled water (1:1 v/v) containing 5% acetic acid) and the absorbance measured at 540 nm in a multiwell plate reader (BioTek Synergy™ HT, BioTek, Winooski, VT, USA). At least five independent experiments were performed, in triplicate, and the percentage of NR uptake relative to that of the control cells was used as the cytotoxicity measure.

Flow Cytometry Analysis of P-gp Expression
P-gp expression was evaluated by flow cytometry using a P-gp monoclonal antibody [UIC2] conjugated with phycoerythrin (PE) [30]. For that purpose, Caco-2 cells were seeded onto 12-well plates, and exposed, three days after seeding, to the eight tested ATxs 1-8 (20 µM) in fresh cell culture medium. After 24 h of incubation, the cells were washed with HBSS buffer and harvested by trypsinization (0.25% trypsin/1 mM EDTA) to obtain a cellular suspension. After centrifugation (300× g, for 10 min, at 4 • C), the cells were suspended in HBSS buffer containing the P-gp antibody, being the antibody dilution selected accordingly to the manufacturer's instructions for flow cytometry. After a 30 min incubation, at 37 • C in the dark, with the UIC2 antibody, the cells were washed twice with HBSS containing 10% heat-inactivated FBS, centrifuged (300× g for 10 min), suspended in ice-cold HBSS buffer and kept on ice until analysis. The fluorescence measurements of isolated cells were performed with a flow cytometer (BD AccuriTM C6, BD Biosciences, San Jose, California, USA), being the fluorescence of the UIC2-PE antibody measured by a 585 ± 40 nm band-pass filter (FL2). The logarithmic fluorescence was recorded and displayed as a single parameter histogram, and based on the acquisition of data for 20,000 cells. The parameter used for comparison was the mean of fluorescence intensity (MFI) [calculated as percentage of control (0 µM)]. In order to detect a possible contribution from cells autofluorescence to the analyzed fluorescence signals, non-labelled cells (with or without chiral thioxanthones) were also analyzed in each experiment by a 585 ± 40 nm band-pass filter (FL2). Five independent experiments were performed in duplicate.

Evaluation of P-gp Transport Activity
The effect of the tested compounds on P-gp transport activity was accessed by flow cytometry using RHO 123 (2 µM) as a P-gp fluorescent substrate. Two different approaches were tested: a RHO 123 accumulation assay in the presence (IA) and absence (NA) of elacridar (10 µM), a known P-gp inhibitor, and with or without simultaneous exposure to ATxs 1-8 (20 µM) only during the RHO 123 accumulation phase; a RHO 123 accumulation assay, in the presence or absence of elacridar (10 µM), with or without pre-exposure of Caco-2 cells to ATxs 1-8 (20 µM) for 24 h.

RHO 123 Efflux Assay in the Presence of ATxs 1-8
For this purpose, the cells were seeded onto 75 cm 2 flasks and, after reaching confluence, harvested by trypsinization to obtain a cellular suspension, which was then divided into several aliquots of 300,000 cells. After centrifugation (300× g for 10 min), the cells were suspended in HBSS containing 10% heat-inactivated FBS and 2 µM RHO 123, with and without 20 µM ATxs 1-8, and incubated at 37 • C, for 60 min, in the presence and absence of elacridar (10 µM). The cells were then washed twice with ice-cold HBSS with 10% heat-inactivated FBS, centrifuged (300× g for 10 min at 4 • C) and kept on ice until flow cytometry analysis. The fluorescence measurements of isolated cells were performed as described in the section "Evaluation of P-gp expression", being the RHO 123 fluorescence measured by a 530 ± 15 nm band-pass filter (FL1). The activity of the efflux pump was accessed by the RHO 123 accumulation ratio (Equation (1) P-gp activity was assessed by ratio between the amount of RHO 123 accumulated under P-gp inhibition (with 10 µM elacridar) and the amount of RHO 123 accumulated in the absence of the P-gp inhibitor.
An increase in P-gp activity results in an increased RHO 123 efflux from the cells, which results in a decrease in the intracellular fluorescence intensity (decreased intracellular RHO 123 content). Therefore, a higher RHO 123 accumulation ratio (Equation (1)) is a consequence of a smaller MFI NA, which results from a higher P-gp activity since the dye is being effluxed out of the cells during the accumulation phase. Five independent experiments were performed in triplicate.

RHO 123 Efflux Assay in Caco-2 Cells Pre-Expose to ATxs 1-8 for 24 h
Caco-2 cells, seeded onto 12-well plates, were exposed to chiral thioxanthones (20 µM), in fresh cell culture medium, for 24 h, prior to the evaluation of P-gp activity. After exposure, the cells were harvested by trypsinization to obtain a cell suspension, being the cells of each well divided into two aliquots. One aliquot was submitted to a RHO 123 accumulation phase under inhibited conditions (RHO 123 accumulation performed in IA conditions) and the second aliquot was submitted to a RHO 123 accumulation phase performed under normal conditions (RHO 123 accumulation performed in NA conditions). For IA accumulation, the cells were centrifuged (300× g for 10 min) and suspended in HBSS buffer (pH 7.4) containing 10% heat-inactivated FBS, 2 µM RHO 123 and the P-gp inhibitor elacridar (10.0 µM), while for the NA accumulation a similar incubation was performed but in the absence of the P-gp inhibitor. In both cases, the cells were incubated with the P-gp substrate at 37 • C for 60 min to allow RHO 123 accumulation. After this accumulation period, the cells were washed twice with ice-cold HBSS buffer containing 10% heat-inactivated FBS and suspended in ice-cold HBSS buffer immediately before analysis. The evaluation of RHO 123 intracellular content was performed as described in the "Flow cytometry analysis of P-gp expression" section, being the RHO 123 fluorescence measured by a 530 ± 15 nm band-pass filter (FL1). The results were calculated according to the ratio defined in Equation (1) and expressed as percentage of control cells. Four independent experiments were performed in duplicate.

Evaluation of P-gp ATPase Activity
The effects of the tested ATxs 1-8 on P-gp ATPase activity were evaluated in recombinant human P-gp-enriched membranes, using the luminescent Pgp-Glo™ Assay (Promega Corporation, Madison, WI, USA), according to the manufacturer's instructions.
Briefly, recombinant human P-gp-enriched membranes (25 µg/well) were incubated with 20 µM chiral thioxanthones or sodium vanadate at 100 µM (positive control for P-gp ATPase activity inhibition) or verapamil at 200 µM (positive control for P-gp ATPase activity stimulation) in assay buffer, and 5 mM MgATP, for exactly 60 min, at 37 • C. After incubation, the reaction was stopped and the remaining ATP detected in a multi-well plate reader (BioTek Synergy™ HT), after a 20 min signal-developing period at room temperature, as a luciferase-generated luminescent signal.
The decreased luminescence of the untreated (NT) incubations relative to Na 3 VO 4 -treated incubations reflects the basal P-gp ATPase activity. The decreased luminescence of incubations performed with the positive control drug verapamil relative to Na 3 VO 4 -treated reactions reflects verapamil-stimulated P-gp ATPase activity. The luminescence of the chiral thioxanthones incubations relative to that of the Na 3 VO 4 -treated reactions reflects the effect, if any, of that compounds on P-gp ATPase activity (i.e., a decrease in luminescence reflects a stimulated P-gp ATPase activity). Three independent experiments were performed in triplicate for this assay and the results expressed as mean ± SEM.

Statistical Analysis
All statistical calculations were performed with the GraphPad Prism version 6.00 for Windows (GraphPad Software, San Diego, CA, USA). Normality of the data distribution was assessed by three different tests (KS normality test, D'Agostino & Pearson omnibus normality test and Shapiro-Wilk normality test). Data obtained from the thioxanthones cytotoxicity assays are expressed as mean ± SEM from at least five independent experiments (performed in triplicate) and the statistical comparisons were estimated using the nonparametric method of Kruskal-Wallis, followed by the Dunn's post hoc test. Data from the P-gp expression assays are presented as mean ± SEM from five independent experiments (performed in duplicate) and the statistical comparisons were made using One-way ANOVA, followed by the Tukey's multiple comparisons test. Results from the P-gp transport activity assays performed in the presence of chiral thioxanthones during the RHO 123 accumulation phase are presented as mean ± SEM from five independent experiments (performed in triplicate) and the statistical comparisons were estimated using One-way ANOVA, followed by the Tukey's multiple comparisons test. Results from the P-gp transport activity assays performed in cells pre-exposed to chiral thioxanthones for 24 h are presented as mean ± SEM from four independent experiments (performed in duplicate) and the statistical comparisons were estimated using One-way ANOVA, followed by the Tukey's multiple comparisons test. Results from the P-gp ATPase activity is presented as mean ± SEM from three independent experiments (performed in triplicate) and the statistical comparisons were estimated using One-way ANOVA, followed by the Tukey's multiple comparisons test. In all cases, p values lower than 0.05 were considered significant.

Conclusions
Thioxanthone derivatives and, particularly, aminated thioxanthones, have been characterized as P-gp modulators. Since the pharmacokinetics, efficacy and safety of drugs that are P-gp substrates depend on the level of expression and functionality of P-gp, and can be affected by the chirality of compounds, it is important to better understand the effect that this important property may have in P-gp modulation. Therefore, in order to accomplish the main aim of this study, four enantiomeric pairs of thioxanthones were synthesized through a copper catalyzed Ullman asymmetric coupling reaction that provided the desired chiral thioxanthones with an e.e. > 99%. The assays carried out for the evaluation of the biological activity of the tested chiral ATxs 1-8 demonstrated that these compounds have the ability to immediately increase P-gp activity without interfering with its levels of expression and, therefore, can be characterized as P-gp activators. Chiral thioxanthones 2 (−) and 7 (+) demonstrated to be the most efficient P-gp activators, among all the thioxanthones tested.
Comparing the results obtained with both enantiomers of each enantiomeric pair, no significant enantioselectivity was observed for the direct modulation of P-gp. Nevertheless, the introduction of a chiral center close to the thioxanthonic scaffold favored their activation effect. Differences were noted in what concerns the induction effect, which may be related to their mechanism of induction, which deserves to be further explored. However, it is well known that P-gp expression is regulated by multiple signaling pathways, each involving different molecular targets and transcription factors, making the elucidation of the induction mechanism underlying ATX2-mediated P-gp induction a complex task. One of the possible pathways involved in the transcriptional activation of the MDR1 gene expression involves the activation of the pregnane X receptor (PXR) [4,66], and several anticancer compounds, plant extracts, cholesterol-lowering statins, rifampicin and HIV protease inhibitors were already reported as PXR ligands [67][68][69][70][71][72]. Therefore, the potential enantioselectivity in what concerns to PXR activation deserves to be explored in future studies.