The Second-Generation PIM Kinase Inhibitor TP-3654 Resensitizes ABCG2-Overexpressing Multidrug-Resistant Cancer Cells to Cytotoxic Anticancer Drugs

Human ATP-binding cassette (ABC) subfamily G member 2 (ABCG2) mediates the transport of a wide variety of conventional cytotoxic anticancer drugs and molecular targeted agents. Consequently, the overexpression of ABCG2 in cancer cells is linked to the development of the multidrug resistance (MDR) phenotype. TP-3654 is an experimental second-generation inhibitor of PIM kinase that is currently under investigation in clinical trials to treat advanced solid tumors and myelofibrosis. In this study, we discovered that by attenuating the drug transport function of ABCG2, TP-3654 resensitizes ABCG2-overexpressing multidrug-resistant cancer cells to cytotoxic ABCG2 substrate drugs topotecan, SN-38 and mitoxantrone. Moreover, our results indicate that ABCG2 does not mediate resistance to TP-3654 and may not play a major role in the induction of resistance to TP-3654 in cancer patients. Taken together, our findings reveal that TP-3654 is a selective, potent modulator of ABCG2 drug efflux function that may offer an additional combination therapy option for the treatment of multidrug-resistant cancers.

Given the lack of U.S. Food and Drug Administration (FDA)-approved therapeutic agents for the treatment of multidrug-resistant cancer patients, we and others have been exploring the possibility of repositioning FDA-approved drugs or therapeutic agents that have undergone clinical testing for the reversal of MDR mediated by ABCB1 and/or  50 values are mean ± SD calculated from dose-response curves obtained from at least three independent experiments using cytotoxicity assay as described in Section 4.

TP-3654 Selectively Reverses Multidrug Resistance in ABCG2-Overexpressing Cancer Cells
To determine the selectivity and potency of TP-3654 on reversing transporter-mediated MDR, we investigated the chemosensitizing effect of TP-3654 on MDR mediated by ABCB1 or ABCG2 in ABCB1-and ABCG2-overexpressing multidrug-resistant cells. We first examined the effect of TP-3654 on ABCB1-mediated resistance to doxorubicin, paclitaxel and colchicine, three well-known substrate drugs of ABCB1 [42], in ABCB1-overexpressing NCI-ADR-RES, KB-V-1 cancer cells, and ABCB1-transfected MDR19-HEK293 cells. As summarized in Table 2, we found that at submicromolar concentrations (100-500 nM), TP-3654 had only a marginal effect on ABCB1-mediated MDR in all three ABCB1-overexpressing cell lines. In contrast, we discovered that TP-3654 significantly resensitized ABCG2overexpressing S1-M1-80 ( Figure 1A-C), H460-MX20 ( Figure 1D-F) cancer cells and ABCG2transfected R482-HEK293 cells ( Figure 1G-I) to ABCG2 substrate drugs mitoxantrone, SN-38 and topotecan [43,44], in a concentration-dependent manner. The IC 50 values and extent of reversal by TP-3654, represented by the fold-reversal (FR) values [16,45], are summarized in Tables 2 and 3. Of note, since a moderate basal level of ABCG2 is expressed in H460 cells [46], we also detected some MDR reversal effects by TP-3654 in H460 cells. Our results revealed that at submicromolar concentrations, TP-3654 selectively reversed ABCG2-mediated MDR in ABCG2-overexpressing multidrug-resistant cancer cells in a concentration-dependent manner.

TP-3654 Selectively Inhibits the Drug Efflux Function of ABCG2
The effect of TP-3654 on the drug efflux function of ABCB1 and ABCG2 was determined by treating ABCB1-overexpressing cells and ABCG2-overexpressing cells, respectively, with calcein-AM or pheophorbide A (PhA), in the absence or presence of TP-3654 as described in Section 4. As shown in Figure

TP-3654 Enhances Drug-Induced Apoptosis in ABCG2-Overexpressing Cancer Cells
Next, knowing that TP-3654 selectively reverses ABCG2-mediated MDR (Figure 1) by inhibiting the drug transport function of ABCG2 (Figure 2), we examined the effect of TP-3654 on drug-induced apoptosis in ABCG2-overexpressing cancer cells. Drug-sensitive parental S1 and ABCG2-overexpressing multidrug-resistant S1-M1-80 cancer cells were treated with DMSO (control), 0.5 µM of TP-3654, 5 µM of topotecan, or a combination of 5 µM of topotecan with 0.5 µM of TP-3654 for 48 h before processed as detailed in Section 4. As shown in Figure 3, TP-3654 had no significant apoptotic effect on S1 or S1-M1-80 cancer cells, whereas a substantial increase of apoptosis, from approximately 2% basal to 35% total apoptosis, was observed in S1 cancer cells. Notably, we discovered that TP-3654 enhanced topotecan-induced apoptosis, from approximately 4% to 23% total apoptosis, in S1-M1-80 cancer cells. Our results here confirmed that TP-3654 resensitized ABCG2-overexpressing cancer cells to ABCG2 substrate drug by enhancing drug-induced apoptosis, and not by initiating growth retardation.

TP-3654 Treatment Does Not Affect the Protein Expression of ABCG2 in ABCG2-Overexpressing Cancer Cells
Natarajan et al. reported previously that the PIM kinase inhibitor SG-1776 reversed ABCG2-mediated MDR and increased apoptosis of cells overexpressing ABCG2 by decreasing the protein expression of ABCG2 [41]. To this end, we examined the effect of TP-3654 on the protein expression of ABCG2 in ABCG2-overexpressing cancer cells (Figures 4 and S1). ABCG2-overexpressing S1-M1-80 and H460-MX20 cancer cells were treated with DMSO (control) or increasing concentrations (0.1-0.5 µM) of TP-3654 for 72 h followed by Western blot analysis as described in Section 4. We discovered that TP-3654 did not significantly affect the protein expression of ABCG2 in S1-M1-80 ( Figure 4A) or H460-MX20 ( Figure 4B) cancer cells. Our data indicated that TP-3654 reverses ABCG2-mediated MDR and increased apoptosis of ABCG2-overexpressing cancer cells by inhibiting the drug transport function of ABCG2.

Docking of TP-3654 in the Drug-Binding Pocket of ABCG2
To further understand the binding interaction of TP-3654 with ABCG2, a binding study was performed. TP-3654 was docked into the substrate-binding cavity between the transmembrane helices of the human ABCG2 structure (PDB:6VXH), and the best binding conformation was selected with the binding energy calculated to be −58.23 kcal/mol. Hydrophobic interactions were observed between TP-3654 with both monomers A and B. Val 546 , Met 549 on monomer A and Phe 432 on monomer B were predicted to interact with the cyclohexane moiety on TP-3654. Met 549 and Val 546 on monomer B were found to interact with the imidazo [1,2-b]pyridazine ring. More interactions were also predicted between Val 546 on monomer B with the phenyl ring and Phe 439 with CF 3 . One hydrogen bond was predicted between Thr 435 on monomer B and the hydroxyl group on TP-3654 ( Figure 5). S1 or S1-M1-80 cancer cells, whereas a substantial increase of apoptosis, from approximately 2% basal to 35% total apoptosis, was observed in S1 cancer cells. Notably, we discovered that TP-3654 enhanced topotecan-induced apoptosis, from approximately 4% to 23% total apoptosis, in S1-M1-80 cancer cells. Our results here confirmed that TP-3654 resensitized ABCG2-overexpressing cancer cells to ABCG2 substrate drug by enhancing drug-induced apoptosis, and not by initiating growth retardation. . TP-3654 improves drug-induced apoptosis in ABCG2-overexpressing cancer cells. The extent of apoptosis was determined by treating drug-sensitive S1 and ABCG2-overexpressing multidrug-resistant S1-M1-80 cancer cells with DMSO (control), 500 nM of TP-3654 (+ TP-3654), 5 μM of topotecan (+ topotecan), or a combination of 5 μM of topotecan with 500 nM of TP-3654 (+ topotecan + TP-3654) and analyzed by flow cytometry as described in Section 4. The quantification Figure 3. TP-3654 improves drug-induced apoptosis in ABCG2-overexpressing cancer cells. The extent of apoptosis was determined by treating drug-sensitive S1 and ABCG2-overexpressing multidrugresistant S1-M1-80 cancer cells with DMSO (control), 500 nM of TP-3654 (+TP-3654), 5 µM of topotecan (+topotecan), or a combination of 5 µM of topotecan with 500 nM of TP-3654 (+topotecan +TP-3654) and analyzed by flow cytometry as described in Section 4. The quantification results are presented as mean ± S.D. calculated from at least three independent experiments. ** p < 0.01, versus the same treatment in the absence of TP-3654.

TP-3654 Treatment Does Not Affect the Protein Expression of ABCG2 in ABCG2-Overexpressing Cancer Cells
Natarajan et al. reported previously that the PIM kinase inhibitor SG-1776 reversed ABCG2-mediated MDR and increased apoptosis of cells overexpressing ABCG2 by decreasing the protein expression of ABCG2 [41]. To this end, we examined the effect of TP-3654 on the protein expression of ABCG2 in ABCG2-overexpressing cancer cells (Figures 4 and S1). ABCG2-overexpressing S1-M1-80 and H460-MX20 cancer cells were treated with DMSO (control) or increasing concentrations (0.1-0.5 μM) of TP-3654 for 72 h followed by Western blot analysis as described in Section 4. We discovered that TP-3654 did not significantly affect the protein expression of ABCG2 in S1-M1-80 ( Figure 4A) or H460-MX20 ( Figure 4B) cancer cells. Our data indicated that TP-3654 reverses ABCG2mediated MDR and increased apoptosis of ABCG2-overexpressing cancer cells by inhibiting the drug transport function of ABCG2.

Docking of TP-3654 in the Drug-Binding Pocket of ABCG2
To further understand the binding interaction of TP-3654 with ABCG2, a binding study was performed. TP-3654 was docked into the substrate-binding cavity between the transmembrane helices of the human ABCG2 structure (PDB:6VXH), and the best binding

Discussion
Preclinical development of synthetic inhibitors of ABCB1 and/or ABCG2 has no successful due to the lack of selectivity and unexpected adverse drug-drug intera [3,[47][48][49][50]. In recent years, many protein kinase inhibitors were found to interact ABCB1 and/or ABCG2. Some kinase inhibitors such as almonertinib [24], sitravatinib erdafitinib [23], avapritinib [45], and midostaurin [52], inhibit drug efflux mediat ABCB1 and/or ABCG2, whereas some kinase inhibitors such as osimertinib encorafenib [54], ibrutinib [55], and vemurafenib [56], are substrates of ABCB1 a ABCG2. More importantly, the results of combination therapy trials of erlotinib gemcitabine for advanced pancreatic cancer patients [57,58], as well as lapatinib capecitabine for human epidermal growth factor receptor 2 (HER2)-positive adv breast cancer patients [59,60] demonstrated the advantages of combination thera kinase inhibitors with conventional chemotherapeutic drugs over monothe Furthermore, findings from a more recent trial of doxorubicin in combination wi ABCB1-modulating nilotinib showed the benefits of including a kinase inhibit combination therapy against multidrug-resistant cancers [61]. These findings prom us to investigate the interactions between TP-3654 and ABCB1 and ABCG2.
In this study, we discovered that the PIM kinase inhibitor TP-3654 could i ABCG2-mediated drug transport in a concentration-dependent manner. Consequ the extent of drug-induced apoptosis and MDR mediated by ABCG2 were signific reversed by TP-3654 in ABCG2-overexpressing multidrug-resistant cells. We fu demonstrated that TP-3654 has no significant effect on ABCG2 protein expressi ABCG2-overexpressing cancer cells. In contrast, TP-3654 had a minimal effect o transport function of ABCB1, and it did not resensitize ABCB1-overexpressing ce ABCB1 substrate drugs. Our results indicate that TP-3654 is selective to ABCG2 re to ABCB1. The in silico molecular docking analysis of TP-3654 in the inward Figure 5. The binding mode of TP-3654 with ABCG2 protein structure (PDB:6VXH) was predicted by Accelrys Discovery Studio 4.0 software as described in Section 4. TP-3654 is shown as a molecular model with highlighted yellow color and the atoms for interacting amino acid residues were colored as carbon (gray), oxygen (red), hydrogen (light gray), and nitrogen (blue). Dotted lines indicate proposed interactions.

Discussion
Preclinical development of synthetic inhibitors of ABCB1 and/or ABCG2 has not been successful due to the lack of selectivity and unexpected adverse drug-drug interactions [3,[47][48][49][50]. In recent years, many protein kinase inhibitors were found to interact with ABCB1 and/or ABCG2. Some kinase inhibitors such as almonertinib [24], sitravatinib [51], erdafitinib [23], avapritinib [45], and midostaurin [52], inhibit drug efflux mediated by ABCB1 and/or ABCG2, whereas some kinase inhibitors such as osimertinib [53], encorafenib [54], ibrutinib [55], and vemurafenib [56], are substrates of ABCB1 and/or ABCG2. More importantly, the results of combination therapy trials of erlotinib and gemcitabine for advanced pancreatic cancer patients [57,58], as well as lapatinib and capecitabine for human epidermal growth factor receptor 2 (HER2)-positive advanced breast cancer patients [59,60] demonstrated the advantages of combination therapy of kinase inhibitors with conventional chemotherapeutic drugs over monotherapy. Furthermore, findings from a more recent trial of doxorubicin in combination with the ABCB1-modulating nilotinib showed the benefits of including a kinase inhibitor in combination therapy against multidrug-resistant cancers [61]. These findings prompted us to investigate the interactions between TP-3654 and ABCB1 and ABCG2.
In this study, we discovered that the PIM kinase inhibitor TP-3654 could inhibit ABCG2-mediated drug transport in a concentration-dependent manner. Consequently, the extent of drug-induced apoptosis and MDR mediated by ABCG2 were significantly reversed by TP-3654 in ABCG2-overexpressing multidrug-resistant cells. We further demonstrated that TP-3654 has no significant effect on ABCG2 protein expression in ABCG2overexpressing cancer cells. In contrast, TP-3654 had a minimal effect on the transport function of ABCB1, and it did not resensitize ABCB1-overexpressing cells to ABCB1 substrate drugs. Our results indicate that TP-3654 is selective to ABCG2 relative to ABCB1. The in silico molecular docking analysis of TP-3654 in the inward-open conformation of human ABCG2 shows the predicted interactions between TP-3654 and several residues within the substrate-binding pocket of ABCG2. Moreover, despite the interaction with ABCG2, we found that TP-3654 is equally cytotoxic to ABCG2-overexpressing cell lines as to their respective drug-sensitive parental cell lines. Our data suggest that ABCG2 does not confer significant resistance to TP-3654 and may not play a major role in the induction of resistance to TP-3654 in cancer patients. Notably, Natarajan et al. reported that the first clinically tested PIM inhibitor SGI-1776 [62] increased substrate drug-induced apoptosis in ABCB1-and ABCG2-overexpressing multidrug-resistant cancer cells. Moreover, at a non-cytotoxic concentration of 1 µM, SGI-1776 resensitized ABCB1-overexpressing cancer cells to ABCB1 substrate drug daunorubicin with FR values of 2.9 and 4.0, whereas it resensitized ABCG2-overexpressing cancer cells to ABCG2 substrate drug mitoxantrone with FR values of 2.7 and 2.4 [41]. In contrast, at a non-cytotoxic concentration of 500 nM, TP-3654 did not resensitize ABCB1-overexpressing cancer cells to ABCB1 substrate drugs, but significantly resensitized ABCG2-overexpressing cancer cells to ABCG2 substrate drugs topotecan, SN-38, and mitoxantrone, with FR values ranging from approximately 4 to 32 (Table 3). Our results suggest that TP-3654 is more potent and selective than SGI-1776 against ABCG2-mediated MDR.

Cytotoxicity Assay
Cells were seeded in 96-well flat-bottom plates and allowed to attach overnight at 37 • C in 5% CO 2 humidified air. Varying concentrations of TP-3654 alone or in combination with chemotherapeutic agents were added to each plate and incubated for an additional 72 h before processed as previously described [71]. Viable cells were quantified based on the cytotoxic MTT assay reported by Ishiyama et al. [72]. IC 50 values were calculated using the fitted concentration-response curve of each drug regimen from at least three independent experiments. The extent of reversal was presented by a fold-reversal (FR) value, determined by adding TP-3654 or tariquidar or Ko143 to the cytotoxicity assays as described previously [16].

Apoptosis Assay
The concurrent staining of annexin V-FITC and propidium iodide (PI) method was used according to the manufacturer's instructions (BD Pharmingen) to determine the extent of apoptosis induced by a cytotoxic drug and as previously described [51]. Briefly, S1 and S1-M1-80 cancer cells were treated with DMSO, 500 nM of TP-3654 alone, 5 µM of topotecan alone, or the combination of 5 µM of topotecan and 500 nM of TP-3654 as indicated for 48 h before stained with annexin V-FITC (1.25 µg/mL) and PI (0.1 mg/mL) for 15 min at room temperature. Samples were analyzed by FACScan equipped with the CellQuest software (Becton-Dickinson Biosciences, San Jose, CA, USA) as previously described [19].

Flow Cytometry
Flow cytometry assays with the ABCB1 substrate calcein-AM and the ABCG2 substrate PhA were performed as described previously [51]. Briefly, trypsinized cells were incubated in phenol red-free Iscove's modified Dulbecco's medium (IMDM) supplemented with 10% FCS and 100 units of penicillin/streptomycin/mL with calcein-AM or PhA in the presence of DMSO (control) or 3 µM of tariquidar or 1 µM of Ko143 or increasing concentrations of TP-3654. The relative fluorescence intensity was detected using a FACSort flow cytometer (Becton-Dickinson) and analyzed using FlowJo software (Tree Star, Inc., Ashland, OR, USA), as described previously [38,73].

Docking Analysis of TP-3654 with ABCG2
The structures of ABCG2 protein (PDB:6VXH) [74] and TP-3654 were first prepared with CHARMM force field at pH 7.4 using Accelrys Discovery Studio 4.0. Docking of TP-3654 in ABCG2 was performed using CDOCKER module of the same software. The docked poses with the lowest CDOCKER interaction energy were selected and the respective interaction energy was calculated as described previously [9].

Quantification and Statistical Analysis
Experimental results were obtained from at least three independent experiments. The differences were analyzed by a two-tailed Student's t-test and labeled with asterisks as "statistically significant" if the probability, p, was less than 0.05 compared with control. GraphPad Prism software (GraphPad Software, La Jolla, CA, USA) was used for curve plotting, and KaleidaGraph software (Synergy Software, Reading, PA, USA) was used for statistical analysis.

Conclusions
In summary, we revealed that the second-generation PIM kinase inhibitor TP-3654 resensitizes ABCG2-overexpressing multidrug-resistant cancer cells to cytotoxic anticancer drugs by attenuating the drug efflux function of ABCG2 ( Figure 6). Although it is possible that unforeseen drug-drug interactions and adverse drug reactions may occur, our results