The multi-targeted tyrosine kinase inhibitor SKLB610 resensitizes ABCG2-overexpressing multidrug-resistant cancer cells to chemotherapeutic drugs

Abstract

The overexpression of ATP-binding cassette (ABC) transporter ABCB1 (P-glycoprotein) or ABCG2 (BCRP/MXR/ABCP) in cancer cells is frequently associated with the development of multidrug resistance (MDR) in cancer patients, which remains a major obstacle to effective cancer treatment. By utilizing energy derived from ATP hydrolysis, both transporters have been shown to reduce the chemosensitivity of cancer cells by actively effluxing cytotoxic anticancer drugs out of cancer cells. Knowing that there are presently no approved drugs or other therapeutics for the treatment of multidrug-resistant cancers, in recent years, studies have investigated the repurposing of tyrosine kinase inhibitors (TKIs) to act as agents against MDR mediated by ABCB1 and/or ABCG2. SKLB610 is a multi-targeted TKI with potent activity against vascular endothelial growth factor receptor 2 (VEGFR2), platelet-derived growth factor receptor (PDGFR), and fibroblast growth factor receptor 2 (FGFR2). In this study, we investigate the interaction of SKLB610 with ABCB1 and ABCG2. We discovered that neither ABCB1 nor ABCG2 confers resistance to SKLB610, but SKLB610 selectively sensitizes ABCG2-overexpressing multidrug-resistant cancer cells to cytotoxic anticancer agents in a concentration-dependent manner. Our data indicate that SKLB610 reverses ABCG2-mediated MDR by attenuating the drug-efflux function of ABCG2 without affecting its total cell expression. These findings are further supported by results of SKLB610-stimulated ABCG2 ATPase activity and in silico docking of SKLB610 in the drug-binding pocket of ABCG2. In summary, we reveal the potential of SKLB610 to overcome resistance to cytotoxic anticancer drugs, which offers an additional treatment option for patients with multidrug-resistant cancers and warrants further investigation.

1. Introduction

ABCB1 (MDR1; P-glycoprotein) and ABCG2 (BCRP; MXR; ABCP) are two of the most well-characterized members of the human ATP-binding cassette (ABC) proteins known to utilize the chemical energy produced by ATP hydrolysis to translocate a wide range of structurally unrelated xenobiotics and chemotherapeutic drugs across biological membranes [1–3]. Many of the most commonly prescribed cytotoxic anticancer drugs, such as Vinca alkaloids, mitoxantrone, topotecan, taxanes and anthracyclines, as well as molecularly targeted agents, are known substrate drugs of ABCB1 and/or ABCG2 [3–10]. These transporters are recognized to play an important role in modulating the oral availability and tissue distribution of substrate drugs because of the protective barriers they form at the intestinal epithelium, blood–brain barrier (BBB), and the blood–placenta barrier (BPB) [3,11,12]. Moreover, the development of multidrug-resistance (MDR) associated with the overexpression of ABCB1 or ABCG2 in cancer cells has been linked to poor prognosis [13–18] and failure of cancer chemotherapy [1–3,19], especially in patients with solid tumors such as lung carcinoma [20], esophageal squamous cell carcinoma (ESCC)[21,22] and metastatic breast cancer (MBC) [18], and in individuals with blood cancers such as acute myelogenous leukemia (AML), acute lymphocytic leukemia (ALL), chronic lymphocytic leukemia (CLL) [13–16], and multiple myeloma (MM) [17,23,24]. Therefore, it is of great significance to discover selective and effective therapeutic agents against the activity of ABCB1 or ABCG2 for clinical application [3,4].

Despite the tremendous efforts that have been invested in the preclinical development of synthetic inhibitors of ABCB1 and ABCG2, the U.S. Food and Drug Administration (FDA) has yet to approve any inhibitor to overcome MDR in cancer patients; the progress has been slow due to unforeseen adverse drug effects [3,25–28]. For instance, the ABCB1 inhibitor tariquidar (XR9576) failed in two phase III clinical trials (ClinicalTrials.gov Identifier: NCT00042315 and NCT00042302) due to increased toxicity in patients. As a result, a drug repositioning (also known as drug repurposing or drug retasking) approach has been considered by many as an alternative approach to the traditional drug discovery process. SKLB610 is a multi-targeted inhibitor of angiogenesis that is highly effective against the activity of angiogenesis vascular endothelial growth factor receptor 2 (VEGFR2), platelet-derived growth factor receptor (PDGFR) and fibroblast growth factor receptor 2 (FGFR2) [29]. SKLB610 was found to inhibit the proliferation of a variety of human cancer cell lines with IC₅₀ values in the range of 5–26 μM [29]. Additional pharmacokinetic studies of SKLB610 were also performed [30,31]. In the present study, we explore the prospects of repurposing SKLB610 to sensitize multidrug-resistant cancer cells overexpressing ABCB1 or ABCG2 to conventional cytotoxic anticancer drugs. We found that neither ABCB1 nor ABCG2 mediates resistance to SKLB610 in cancer cells. More importantly, we discovered an additional pharmacological activity of SKLB610 as an inhibitor of ABCG2-mediated drug transport and its ability to sensitize ABCG2-overexpressing multidrug-resistant cancer cells to apoptosis and cytotoxicity induced by cytotoxic anticancer drugs. Collectively, in addition to its anti-proliferative activity, the additional pharmacological activity of SKLB610 against ABCG2-mediated MDR warrants further studies for its use in combination therapies in patients with multidrug-resistant cancers.

2. Materials and methods

2.1. Cell culture

The human epidermal cancer cell line KB-3–1 and its ABCB1-overexpressing variant KB-V1 [32], the human embryonic kidney 293 cell line (HEK 293) transfected with either empty pcDNA 3.1 vector, or human ABCB1 (MDR19-HEK293) [33] or human ABCG2 (R482-HEK293) [34] were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM) (Gibco, Invitrogen, Carlsbad, CA, USA). The human ovarian cancer cell line OVCAR-8 and its ABCB1-overexpressing variant NCI-ADR-RES [35], the human colon cancer cell line S1 and its ABCG2-overexpressing variant S1-MI-80 [36], the human non-small cell lung cancer (NSCLC) cell line H460 and its ABCG2-overexpressing variant H460-MX20 [37] were maintained in Roswell Park Memorial Institute (RPMI-1640) medium (Gibco, Invitrogen, Carlsbad, CA, USA). The HEK293 transfectants were maintained in the presence of 2 mg/mL of G418 [38]; KB-V1 cells were cultured in the presence of 1 μg/mL of vinblastine [39] and NCI-ADR-RES cells were cultured in the presence of 0.85 μM doxorubicin [35], whereas H460-MX20 cells were cultured in the presence of 20 nM of mitoxantrone [40] and S1-MI-80 cells were cultured in the presence of 80 μM mitoxantrone [36]. All cells were cultured at 37 °C in 5% CO₂ humidified air and grown in media supplemented with 10% FCS, L-glutamine and 100 units/mL of penicillin and streptomycin. Cell lines were generous gifts from Drs. Michael Gottesman and Susan Bates (NCI, NIH, Bethesda, MD, USA). Cells were screened periodically for mycoplasma contamination using a TOOLS Mycoplasma Detection Kit and maintained in drug-free medium for 7 days before assay.

2.2. Chemicals

The TOOLS Cell Counting (CCK-8) kit was purchased from Biotools Co., Ltd. (Taipei, Taiwan). The annexin V FITC-apoptosis detection kit was purchased from BD Pharmingen (San Diego, CA, USA). SKLB610 was obtained from Selleckchem (Houston, TX, USA). Ko143, tariquidar, and all other chemicals were purchased from Sigma (St. Louis, MO, USA), unless stated otherwise.

2.3. Cytotoxicity assays

To determine the sensitivity of cell lines to SKLB610 or a combination of SKLB610 with another cytotoxic drug, cytotoxic MTT and Cell Counting Kit-8 (CCK-8) assays were performed based on the method described by Ishiyama et al. [41] and as described previously [42]. Briefly, cells were seeded in 96-well flat-bottom plates and allowed to attach at 37 °C in 5% CO₂ humidified air. After 24 h, different concentrations of SKLB610 or drug combinations were added to each well with 0.5% (v/v) final concentration of DMSO in all wells and incubated for an additional 72 h before being processed with either MTT or CCK-8 reagent as described previously [38]. The IC₅₀ value of each drug regimen was calculated using a fitted concentration-response curve from at least three independent experiments. The extent of drug resistance was presented as a resistance-factor (R.F.) value as described previously [43]. For the drug resistance reversal assays, the extent of chemosensitization by SKLB610 was presented as a fold-reversal (F.R.) value, determined by adding SKLB610 or a reference inhibitor of ABCB1 or ABCG2, at nontoxic concentrations, to the cytotoxicity assays as described previously [44].

2.4. Flow cytometry

A BD FACScan flow cytometer (BD Biosciences) was used to determine the intracellular accumulation of pheophorbide A (PhA) (395 nm excitation and 670 nm emission), a known fluorescent substrate of ABCG2 [45], as described previously [46]. Briefly, cells were first trypsinized and resuspended in Iscove’s modified Dulbecco’s medium (IMDM) containing 5% FBS before PhA (1 μM) was added to 3 × 10⁵ cells in 4 mL of IMDM in the presence of DMSO (control), 10 μM SKLB610 or 5 μM Ko143, an ABCG2 reference inhibitor. The relative fluorescence intensity of PhA was analyzed using CellQuest software (Becton-Dickinson Biosciences, San Jose, CA) and FlowJo software (Tree Star, Inc., Ashland, OR, USA), as described previously [43,45].

2.5. Immunoblot analysis

Cancer cells were treated with either DMSO (control) or SKLB610 at 0.5–3.0 μM for 72 h before being harvested and subjected to SDS-polyacrylamide electrophoresis and Western blotting as described previously [38]. Primary antibodies BXP-21 (1:1000 dilution) and anti-alpha tubulin (1:100,000 dilution) (Sigma-Aldrich, St. Louis, MO, USA) were used to detect human ABCG2 and the positive loading control tubulin, respectively. Horseradish peroxidase-conjugated goat anti-mouse immunoglobulin G (IgG) (1:100,000 dilution) (Abcam, Cambridge, MA, USA) was used as a secondary antibody. Signals were detected using the enhanced chemiluminescence (ECL) kit (Merck Millipore, Billerica, MA, USA) as described previously [38].

2.6. Apoptosis assays

The concurrent annexin V–FITC and propidium iodide (PI) staining method was used to determine the pro-apoptotic effect of drug regimens in cancer cells as previously described [47]. Briefly, parental drug-sensitive and ABCG2-overexpressing multidrug-resistant cancer cells were treated with either DMSO, 10 μM of SKLB610 alone, or 5 μM of the known apoptotic inducer topotecan, or a combination of 5 μM topotecan and 10 μM SKLB610 for 48 h. Cells were subsequently processed using the FITC Annexin V Apoptosis Detection Kit (BD Pharmingen, San Diego, CA, USA) according to the manufacturer’s protocol and analyzed (10,000 cells per sample) using a FACSCalibur flow cytometer equipped with CellQuest software (Becton-Dickinson Biosciences, San Jose, CA) as described previously [42].

2.7. ATPase assay

The vanadate (Vi)-sensitive ATPase activity of ABCG2 was measured using membrane vesicles prepared from ABCG2 baculovirus-infected High-Five insect cells based on the endpoint P_i assay as described previously [48].

2.8. Docking analysis

The energy was minimized for both the structures of ABCG2 protein (PDB: 6VXH) [49] and SKLB610 with CHARMM force field at pH 7.4 using Accelrys Discovery Studio 4.0. Ligand preparation and docking of SKLB610 in ABCG2 was performed using the CDOCKER module of the same software. The respective interaction energy was calculated and the conformation with the lowest CDOCKER interaction energy was selected as described previously [50].

2.9. Quantification and statistical analysis

Experimental values are presented as mean ± standard deviation (S. D.) or as mean ± standard error of the mean (S.E.M) calculated from at least three independent experiments. GraphPad Prism software (GraphPad Software, La Jolla, CA, USA) and KaleidaGraph software (Synergy Software, Reading, PA, USA) were used for curve plotting and statistical analysis, respectively. Two-tailed Student’s t-tests were performed to analyze the difference between mean values of experimental and control or improvement in fit and labeled with asterisks as “statistically significant” if the probability, p, was less than 0.05.

3. Results

3.1. SKLB610 is equally cytotoxic to drug-sensitive and multidrug-resistant cells overexpressing ABCB1 or ABCG2

Previous studies have reported that ABCB1 and ABCG2 are capable of transporting and/or conferring resistance to some tyrosine kinase inhibitors (TKIs) [51–56]. Therefore, the intrinsic cytotoxicity of SKLB610, a multi-targeted inhibitor of tyrosine kinases, was examined in multiple pairs of drug-sensitive cell lines and the corresponding ABCB1- or ABCG2-overexpressing multidrug-resistant sublines. The extent of cellular resistance to SKLB610 mediated by ABCB1 or ABCG2 was represented as the resistance factor (R.F.), which was calculated by dividing the IC₅₀ value of SKLB610 in the multidrug-resistant subline by the IC₅₀ value of SKLB610 in the respective parental line. As summarized in Table 1, the ABCB1-overexpressing human epidermal cancer cell line KB-V1 and human ovarian cancer cell line NCI-ADR-RES, as well as the ABCG2-overexpressing human lung cancer cell line H460-MX20 and the human colon cancer cell line S1-M180, and their corresponding parental cell lines were equally sensitive to SKLB610. Similarly, SKLB610 was cytotoxic to the same level for HEK293 cells transfected with human ABCB1 (referred to as MDR19-HEK293) or human ABCG2 (referred to as R482-HEK293), and parental HEK293 transfected with empty pcDNA 3.1 vector (referred to as pcDNA-HEK293).

Table 1.

Cytotoxicity of SKLB610 in human cell lines overexpressing ABCB1 or ABCG2.

Cell line	Type	Transporter overexpressed	IC50 (μM)a	RFb
KB-3–1	Epidermal	–	15.75 ± 2.48	1.0
KB-V1	Epidermal	ABCB1	10.75 ± 2.32	0.7
OVCAR-8	Ovarian	–	20.92 ± 1.31	1.0
NCI-ADR-RES	Ovarian	ABCB1	25.07 ± 2.64	1.2
H460	Lung	–	31.07 ± 3.22	1.0
H460-MX20	Lung	ABCG2	25.10 ± 2.76	0.8
S1	Colon	–	16.06 ± 4.15	1.0
S1-MI-80	Colon	ABCG2	17.06 ± 2.26	1.1
pcDNA3.1-HEK293	–	–	54.72 ± 7.46	1.0
MDR19-HEK293	–	ABCB1	68.61 ± 12.58	1.3
R482-HEK293	–	ABCG2	77.76 ± 12.89	1.4

Abbreviations: RF, resistance factor.

^aIC₅₀ values were calculated from dose-response curves obtained from at least three independent experiments as described in Section 2.

^bRF values were obtained by dividing the IC₅₀ value of SKLB610 in the ABCB1- or ABCG2-overexpressing multidrug-resistant cell lines by the IC₅₀ value of SKLB610 in the corresponding drug-sensitive parental cell lines.

3.2. SKLB610 re-sensitizes ABCG2-overexpressing multidrug-resistant cancer cells to cytotoxic anticancer drugs

Given that some TKIs are known to reverse ABCB1- and ABCG2-mediated multidrug resistance in cancer cells [42,44,50,57–61], SKLB610 was assessed for its ability to re-sensitize multidrug-resistant human cancer cells overexpressing ABCB1 or ABCG2, as well as in HEK293 cells transfected with human ABCB1 or ABCG2 to cytotoxic drugs. As summarized in Table 2, SKLB610 at sub-toxic concentrations (0.5–3.0 μM) had no significant effect on ABCB1-mediated resistance to colchicine, vincristine or doxorubicin [62] in KB-V1, NCI-ADR-RES cancer cells or MDR19-HEK293 cells. In contrast, we found that SKLB610 could reverse ABCG2-mediated resistance to its substrates mitoxantrone (Fig. 1A–C), SN-38 (Fig. 1D–F) and topotecan (Fig. 1G–I) [63–65] in S1-MI-80 and H460-MX20 cancer cells, as well as in R482-HEK293 cells in a concentration-dependent manner. As shown in Table 3, the extent of re-sensitization by SKLB610 was represented as the fold-reversal (F.R.) value, which was calculated by dividing the IC₅₀ value of a particular cytotoxic drug in the absence of SKLB610 by the IC₅₀ value of the same cytotoxic drug in the presence of SKLB610 in the same cell line [44]. Tariquidar and Ko143 were used here as positive controls to demonstrate the reversal of drug resistance mediated by ABCB1 and ABCG2, respectively. Of note, other than the H460 cell line that expresses a noticeable amount of ABCG2 intrinsically, SKLB610 had no significant effect on the chemosensitivity of drug-sensitive parental cells. Our data revealed that SKLB610 selectively reversed ABCG2-mediated MDR in cancer cells.

Table 2.

The effect of SKLB610 on ABCB1-mediated multidrug resistance.

		Mean IC50a ± SD and (FRb)
Treatment	Concentration (μM)	OVCAR-8 (parental) [nM]	NCI-ADR-RES (resistant) [μM]
Colchicine	–	23.96 ± 6.25 (1.0)	2.75 ± 0.52 (1.0)
+ SKLB610	0.5	25.78 ± 7.41 (0.9)	3.07 ± 0.55 (0.9)
+ SKLB610	1.0	26.72 ± 8.02 (0.9)	3.58 ± 0.66 (0.8)
+ SKLB610	2.0	26.08 ± 7.30 (0.9)	2.74 ± 0.55 (1.0)
+ SKLB610	3.0	25.91 ± 7.31 (0.9)	2.70 ± 0.52 (1.0)
+ Tariquidar	1.0	24.44 ± 6.74 (1.0)	46.79 ± 15.01 [nM]*** (59)
		[nM]	[μM]
Vincristine	–	15.47 ± 1.80 (1.0)	6.09 ± 1.32 (1.0)
+ SKLB610	0.5	16.83 ± 2.42 (0.9)	6.06 ± 1.47 (1.0)
+ SKLB610	1.0	16.37 ± 2.45 (0.9)	5.33 ± 1.08 (1.1)
+ SKLB610	2.0	15.35 ± 1.82 (1.0)	4.12 ± 0.88 (1.5)
+ SKLB610	3.0	16.05 ± 2.66 (1.0)	3.13 ± 0.64** (1.9)
+ Tariquidar	1.0	13.34 ± 1.56 (1.2)	29.96 ± 5.25 [nM]** (203)
		[nM]	[μM]
Doxorubicin	–	296.98 ± 35.69 (1.0)	7.17 ± 0.77 (1.0)
+ SKLB610	0.5	276.25 ± 22.13 (1.1)	8.32 ± 0.80 (0.9)
+ SKLB610	1.0	267.15 ± 30.99 (1.1)	8.81 ± 0.93 (0.8)
+ SKLB610	2.0	253.97 ± 28.85 (1.2)	7.12 ± 0.62 (1.0)
+ SKLB610	3.0	227.27 ± 28.82 (1.3)	6.25 ± 0.62 (1.1)
+ Tariquidar	1.0	261.99 ± 33.88 (1.1)	0.33 ± 0.05*** (21.7)
Treatment	Concentration (μM)	KB-3–1 (parental) [nM]	KB-V1 (resistant) [μM]
Colchicine	–	11.35 ± 4.35 (1.0)	1.29 ± 0.16 (1.0)
+ SKLB610	0.5	11.72 ± 4.58 (1.0)	1.59 ± 0.18 (0.8)
+ SKLB610	1.0	11.78 ± 4.68 (1.0)	1.71 ± 0.23 (0.8)
+ SKLB610	2.0	12.00 ± 4.78 (0.9)	1.59 ± 0.25 (0.8)
+ SKLB610	3.0	11.82 ± 4.75 (1.0)	1.34 ± 0.23 (1.0)
+ Tariquidar	1.0	11.73 ± 4.62 (1.0)	14.18 ± 5.29 [nM]*** (91)
		[nM]	[nM]
Vincristine	–	1.46 ± 0.41 (1.0)	1609.50 ± 381.83 (1.0)
+ SKLB610	0.5	1.75 ± 0.54 (0.8)	1385.80 ± 308.97 (1.2)
+ SKLB610	1.0	1.90 ± 0.62 (0.8)	1289.69 ± 299.71 (1.2)
+ SKLB610	2.0	1.82 ± 0.58 (0.8)	1005.78 ± 230.61 (1.6)
+ SKLB610	3.0	1.96 ± 0.66 (0.7)	912.63 ± 214.78 (1.8)
+ Tariquidar	1.0	1.49 ± 0.41 (1.0)	2.79 ± 0.51** (577)
		[nM]	[μM]
Doxorubicin	–	47.03 ± 8.60 (1.0)	2.10 ± 0.22 (1.0)
+ SKLB610	0.5	42.56 ± 6.70 (1.1)	2.27 ± 0.22 (0.9)
+ SKLB610	1.0	40.90 ± 7.40 (1.1)	2.18 ± 0.22 (1.0)
+ SKLB610	2.0	41.54 ± 7.72 (1.1)	2.34 ± 0.28 (0.9)
+ SKLB610	3.0	42.95 ± 7.76 (1.1)	1.69 ± 0.21 (1.2)
+ Tariquidar	1.0	47.09 ± 9.26 (1.0)	0.15 ± 0.02*** (14.0)
Treatment	Concentration (μM)	pcDNA3.1-HEK293 (parental) [nM]	MDR19-HEK293 (resistant) [nM]
Colchicine	–	14.57 ± 4.88 (1.0)	147.25 ± 29.93 (1.0)
+ SKLB610	0.5	14.75 ± 5.02 (1.0)	181.71 ± 38.28 (0.8)
+ SKLB610	1.0	13.77 ± 4.57 (1.1)	169.78 ± 46.56 (0.9)
+ SKLB610	2.0	13.96 ± 4.54 (1.0)	198.09 ± 49.60 (0.7)
+ SKLB610	3.0	13.69 ± 4.56 (1.1)	175.21 ± 49.45 (0.8)
+ Tariquidar	1.0	15.00 ± 5.05 (1.0)	11.53 ± 4.10** (12.8)
		[nM]	[nM]
Vincristine	–	3.76 ± 0.60 (1.0)	547.08 ± 110.31 (1.0)
+ SKLB610	0.5	3.25 ± 0.44 (1.2)	655.39 ± 104.48 (0.8)
+ SKLB610	1.0	3.26 ± 0.46 (1.2)	572.03 ± 138.76 (1.0)
+ SKLB610	2.0	3.36 ± 0.56 (1.1)	523.91 ± 111.69 (1.0)
+ SKLB610	3.0	3.43 ± 0.62 (1.1)	418.40 ± 65.50 (1.3)
+ Tariquidar	1.0	2.10 ± 0.32* (1.8)	1.75 ± 0.37** (312.6)
		[nM]	[nM]
Doxorubicin	–	20.71 ± 3.84 (1.0)	413.62 ± 56.02 (1.0)
+ SKLB610	0.5	24.43 ± 5.45 (0.8)	462.24 ± 43.34 (0.9)
+ SKLB610	1.0	27.63 ± 6.51 (0.7)	496.15 ± 76.15 (0.8)
+ SKLB610	2.0	24.55 ± 4.28 (0.8)	501.70 ± 76.17 (0.8)
+ SKLB610	3.0	23.72 ± 4.25 (0.9)	461.56 ± 64.79 (0.9)
+ Tariquidar	1.0	21.05 ± 3.76 (1.0)	63.71 ± 6.71*** (6.5)

Abbreviations: FR, fold-reversal.

^**p < 0.01

^***p < 0.001.

^aIC₅₀ values were calculated as described in the legend to Table 1.

^bFR values were calculated by dividing the IC₅₀ value of a known ABCB1 substrate drug by the IC₅₀ value of the same substrate drug in the presence of SKLB610 or tariquidar.

Fig. 1.

SKLB610 resensitizes ABCG2-overexpressing multidrug-resistant cells to anticancer drugs in a concentration-dependent manner. The effect of SKLB610 on ABCG2-mediated resistance to mitoxantrone (A–C), SN-38 (D–F), and topotecan (G–I) was evaluated in the ABCG2-overexpressing human colon cancer cell line S1-MI-80 (A, D and E, left panels) and its drug-sensitive parental line S1 (A, D and E, right panels), ABCG2-overexpressing human NSCLC cell line H460-MX20 (B, E and H, left panels) and its drug-sensitive parental line H460 (B, E and H, right panels), ABCG2-transfected R482-HEK293 cell line (C, F and I, left panels) and the parental line HEK293 (C, F and I, right panels). Cells were treated with increasing concentrations of mitoxantrone, SN-38 or topotecan in the presence of DMSO (empty circles) or SKLB610 at 500 nM (empty squares), 1 μM (filled squares), 2 μM (empty triangles), or 3 μM (filled triangles) for 72 h before analysis as described in Section 2. Points, mean values from at least three independent experiments; bars; S.E.M.

Table 3.

The effect of SKLB610 on ABCG2-mediated multidrug resistance.

		Mean IC50a ± SD and (FRb)
Treatment	Concentration	S1 (parental) [nM]	S1-MI-80 (resistant)
	(μM)		[μM]
Mitoxantrone	–	8.20 ± 1.15 (1.0)	63.99 ± 4.49 (1.0)
+ SKLB610	0.5	6.41 ± 1.02 (1.3)	25.27 ± 2.83*** (2.5)
+ SKLB610	1.0	7.26 ± 1.31 (1.1)	14.05 ± 1.94*** (4.6)
+ SKLB610	2.0	7.22 ± 1.51 (1.1)	8.06 ± 1.06*** (7.9)
+ SKLB610	3.0	5.52 ± 1.91 (15)	4.22 ± 0.58*** (15.2)
+ Ko143	1.0	6.86 ± 1.04 (1.2)	0.73 ± 0.07*** (57.4)
		[nM]	[μM]
SN-38	–	5.69 ± 0.82 (1.0)	7.59 ± 2.16 (1.0)
+ SKLB610	0.5	6.61 ± 0.92 (0.9)	4.72 ± 1.69 (1.6)
+ SKLB610	1.0	6.62 ± 0.86 (0.9)	3.17 ± 1.23* (2.4)
+ SKLB610	2.0	6.52 ± 0.90 (0.9)	1.88 ± 0.49* (4.0)
+ SKLB610	3.0	6.63 ± 1.00 (0.9)	1.41 ± 0.39** (5.4)
+ Ko143	1.0	6.78 ± 0.96 (0.8)	110.32 ± 32.73 [nM]*** (69)
		[nM]	[μM]
Topotecan	–	47.97 ± 3.39 (1.0)	15.55 ± 2.67 (1.0)
+ SKLB610	0.5	49.86 ± 5.31 (1.0)	4.76 ± 1.01** (3.3)
+ SKLB610	1.0	48.74 ± 5.25 (1.0)	3.98 ± 0.86** (3.9)
+ SKLB610	2.0	44.91 ± 4.74 (1.1)	3.09 ± 0.72** (5.0)
+ SKLB610	3.0	43.41 ± 4.33 (1.1)	1.97 ± 0.43*** (7.9)
+ Ko143	1.0	44.36 ± 5.08 (1.1)	0.33 ± 0.07*** (47.1)
Treatment	Concentration	H460 (parental) [nM]	H460-MX20
	μM)		(resistant) [nM]
Mitoxantrone	–	35.04 ± 5.26 (1.0)	841.13 ± 99.22 (1.0)
+ SKLB610	0.5	24.57 ± 3.96 (1.4)	337.64 ± 65.21** (2.5)
+ SKLB610	1.0	20.13 ± 3.67* (1.7)	228.04 ± 51.99*** (3.7)
+ SKLB610	2.0	17.96 ± 3.15** (2.0)	186.38 ± 44.86*** (4.5)
+ SKLB610	3.0	17.87 ± 3.31** (2.0)	175.70 ± 48.63*** (4.8)
+ Ko143	1.0	17.83 ± 3.31** (2.0)	86.46 ± 25.36*** (9.7)
		[nM]	[μM]
SN-38	–	16.62 ± 1.98 (1.0)	417.04 ± 77.66 (1.0)
+ SKLB610	0.5	12.46 ± 1.76 (1.3)	214.29 ± 46.00* (1.9)
+ SKLB610	1.0	10.06 ± 1.63* (1.7)	139.10 ± 31.03** (3.0)
+ SKLB610	2.0	7.69 ± 1.24** (2.2)	103.37 ± 25.14** (4.0)
+ SKLB610	3.0	7.56 ± 1.30** (2.2)	86.54 ± 22.74** (4.8)
+ Ko143	1.0	4.90 ± 1.09*** (3.4)	9.56 ± 2.62*** (43.6)
		[nM]	[μM]
Topotecan	–	139.14 ± 18.39 (1.0)	1329.22 ± 308.64 (1.0)
+ SKLB610	0.5	93.70 ± 13.58* (1.5)	666.77 ± 149.84* (2.0)
+ SKLB610	1.0	80.22 ± 11.64** (1.7)	591.98 ± 126.24* (2.2)
+ SKLB610	2.0	66.32 ± 10.30** (2.1)	437.45 ± 96.69** (3.0)
+ SKLB610	3.0	56.40 ± 9.16** (2.5)	326.33 ± 75.55** (4.1)
+ Ko143	1.0	48.96 ± 8.50** (2.8)	85.75 ± 20.07** (15.5)
Treatment	Concentration	pcDNA3.1-HEK293	R482-HEK293
	(μM)	(parental) [nM]	(resistant) [nM]
Mitoxantrone	–	4.76 ± 1.15 (1.0)
+ SKLB610	0.5	4.51 ± 0.83 (1.1)	125.97 ± 17.05 (1.0) 63.54 ± 7.62** (2.0)
+ SKLB610	1.0	4.47 ± 0.92 (1.1)	55.34 ± 5.17** (2.3)
+ SKLB610	2.0	4.48 ± 0.99 (1.1)	38.78 ± 5.29** (3.2)
+ SKLB610	3.0	4.38 ± 0.95 (1.1)	31.27 ± 4.65*** (4.0)
+ Ko143	1.0	4.29 ± 0.84 (1.1)	12.20 ± 1.02*** (10.3)
SN-38	–	[nM] 4.26 ± 0.97 (1.0)	[nM] 335.81 ± 36.28 (1.0)
+ SKLB610	0.5	4.27 ± 1.00 (1.0)	131.70 ± 18.69*** (2.5)
+ SKLB610	1.0	4.18 ± 1.03 (1.0)	93.22 ± 17.39*** (3.6)
+ SKLB610	2.0	3.97 ± 0.90 (1.1)	49.99 ± 11.55*** (6.7)
+ SKLB610	3.0	3.61 ± 0.75 (1.2)	38.76 ± 7.92*** (8.7)
+ Ko143	1.0	3.75 ± 0.87 (1.1)	12.22 ± 2.43*** (27.4)
Topotecan	–	[nM] 37.24 ± 9.83 (1.0)	[nM] 731.78 ± 89.04 (1.0)
+ SKLB610	0.5	36.16 ± 9.12 (1.0)	423.60 ± 53.13** (1.7)
+ SKLB610	1.0	32.29 ± 7.10 (1.2)	301.48 ± 44.27** (2.4)
+ SKLB610	2.0	31.15 ± 7.84 (1.2)	203.06 ± 29.44*** (3.6)
+ SKLB610	3.0	28.80 ± 6.25 (1.3)	170.08 ± 30.54*** (4.3)
+ Ko143	1.0	34.71 ± 8.07 (1.1)	138.25 ± 25.38*** (5.3)

Abbreviations: FR, fold-reversal.

^*p < 0.05

^**p < 0.01

^***p < 0.001.

^aIC₅₀ values calculation (see legend to Table 1).

^bFR values were calculated by dividing the IC₅₀ value of a known ABCG2 substrate drug by the IC₅₀ value of the same substrate drug in the presence of SKLB610 or Ko143.

3.3. SKLB610 inhibits ABCG2-mediated drug efflux

Knowing that a common way to re-sensitize ABCG2-overexpressing multidrug-resistant cancer cells to cytotoxic drugs is by directly inhibiting the drug transport function of ABCG2 [27,58,59,66–68], we examined the ability of SKLB610 to inhibit efflux of Pheophorbide A (PhA), a known fluorescent substrate of ABCG2 [45], from ABCG2-overexpressing cells. Cells were incubated in IMDM containing PhA in the presence of DMSO (solid line), SKLB610 (filled solid line), or Ko143 (dotted line) for 45 min before the intracellular accumulation of PhA was determined as described in Materials and methods. We found that SKLB610 and Ko143 had no significant effect on the intracellular accumulation of PhA in drug-sensitive parental S1 (Fig. 2A, left panel), H460 (Fig. 2B, left panel) cancer cells, or pcDNA-HEK293 cells (Fig. 2C, left panel). In contrast, the intracellular fluorescence of PhA was significantly higher in ABCG2-overexpressing S1-MI-80 (Fig. 2A, right panel) and H460-MX20 (Fig. 2B, right panel) cancer cells, and ABCG2-transfected R482-HEK293 cells (Fig. 2C, right panel) in the presence of SKLB610 or Ko143 versus in the absence of any inhibitor.

Fig. 2.

SKLB610 attenuates the drug transport function of ABCG2. The intracellular accumulation of pheophorbide A (PhA), a known fluorescent substrate drug of ABCG2, was determined in (A) S1 and S1-MI-80, (B) H460 and H460-MX20, and (C) pcDNA-HEK293 and R482-HEK293 cells, in the presence of DMSO (control, solid lines), 10 μM SKLB610 (filled solid lines), or 5 μM Ko143, a benchmark inhibitor of ABCG2 (dotted lines). (D) The corresponding quantification of intracellular PhA accumulation in ABCG2-overexpressing S1-MI-80 (black bars), H460-MX20 (white bars) and R482-HEK293 (gray bars). Representative histograms and the quantification values presented as mean ± S.D. calculated from at least three independent experiments are shown. The fluorescence signal was analyzed by flow cytometry as described previously [38].

3.4. SKLB610 does not alter the protein expression of ABCG2 in multidrug-resistant cancer cells

In addition to inhibiting the drug transport function of ABCG2, studies have shown that drug-induced down-regulation of ABCG2 is also a common mechanism by which ABCG2-overexpressing multidrug-resistant cancer cells can be re-sensitized to cytotoxic drugs [69,70]. Therefore, S1-MI-80 and H460-MX20 cancer cell lines were treated with increasing concentrations (0.5–3 μM) of SKLB610 for 72 h and the protein expression of ABCG2 was examined as given in Materials and methods. As shown in Fig. 3, the expression of ABCG2 was not significantly affected by SKLB610 at the protein level in S1-MI-80 (Fig. 3A) or H460-MX20 (Fig. 3B) cancer cells over a period of 72 h. Our results suggest that SKLB610 re-sensitized ABCG2-overexpressing multidrug-resistant cancer cells to cytotoxic drugs by inhibiting the drug efflux function of ABCG2.

Fig. 3.

SKLB610 has no significant effect on ABCG2 protein expression. The ABCG2-overexpressing S1-MI-80 (A) and H460-MX20 (B) cancer cells were treated with DMSO (vehicle control) or SKLB610 at 0.5 μM, 1 μM, 2 μM, or 3 μM for 72 h and the cell lysates were processed for Western blotting with indicated antibodies as described in Section 2. Representative immunoblots (top) and the corresponding quantification (bottom) of human ABCG2 protein and the internal loading controlα-tubulin are shown. Values are presented as mean ± S.D. calculated from at least three independent experiments.

3.5. SKLB610 restores drug-induced apoptosis in ABCG2-overexpressing multidrug-resistant cancer cells

Knowing that drug-induced growth retardation could also appear to be chemosensitization of multidrug-resistant cancer cells, the effect of SKLB610 on topotecan-induced apoptosis was examined in ABCG2-overexpressing S1-MI-80 cancer cells. S1 and S1-MI-80 cancer cells were treated with DMSO (control), 10 μM SKLB610, 5 μM topotecan or a combination of 5 μM topotecan and 10 μM SKLB610 for 48 h and processed as described in Materials and methods. As shown in Fig. 4, topotecan substantially increased the apoptotic cell population from approximately 4% basal to 39% in the drug-sensitive S1 cancer cell line. In contrast, being a known substrate of ABCG2 [63], topotecan merely increased the apoptotic cell population from approximately 4% basal to 7% in the ABCG2-overexpressing S1-MI-80 subline. However, without affecting the extent of overall apoptosis in S1 cancer cells, SKLB610 significantly enhanced topotecan-induced apoptosis in S1-MI-80 cancer cells from 7% to approximately 19% total apoptosis. Our results indicate that by blocking the drug efflux function of ABCG2, SKLB610 restored drug-induced apoptosis-mediated cytotoxicity in ABCG2-overexpressing multidrug-resistant cancer cells.

Fig. 4.

SKLB610 enhances drug-induced apoptosis in ABCG2-overexpressing cancer cells. The drug-sensitive human colon cancer cell line S1 and its ABCG2-overexpressing multidrug-resistant subline S1-MI-80 were treated with DMSO (control), 10 μM SKLB610 alone (+ SKLB610), 5 μM topotecan alone (+ topotecan), or a combination of topotecan and SKLB610 (+ topotecan + SKLB610) for 48 h, processed and analyzed by flow cytometry as described in Materials and methods. Representative flow cytometric dot plots are shown (top), and the corresponding quantifications (bottom) are presented as mean ± S.D. calculated from at least three independent experiments. **p < 0.01, versus the same treatment in the absence of SKLB610.

3.6. Effect of SKLB610 on ABCG2-mediated ATP hydrolysis

The effect of SKLB610 on the ATPase activity of ABCG2 was examined to obtain additional biochemical information on the interactions between SKLB610 and ABCG2 as ATP hydrolysis is known to be associated with the transport activity of ABCG2 [71]. SKLB610 stimulated Vi-sensitive ATPase activity of ABCG2 in a concentration-dependent manner, with a half-maximal effective concentration (EC₅₀) value of approximately 98 nM and a maximum stimulation of almost 300% higher than the basal activity of 58.1 ± 11.2 nmole P_i/min/mg protein (Fig. 5). These results suggest that SKLB610 interacts with ABCG2 at the substrate-binding site.

Fig. 5.

SKLB610 stimulates the ATPase activity of ABCG2. The effect of SKLB610 (0–5 μM) on ABCG2-mediated ATP hydrolysis was measured in membrane vesicles prepared from ABCG2 baculovirus-infected High-Five insect cells and recorded as vanadate (Vi)-sensitive ATPase activity as described previously [43]. Points, mean from at least three independent experiments; bars, S.D.

3.7. Docking of SKLB610 in the drug-binding pocket of ABCG2

To further understand the binding of SKLB610 with ABCG2, the potential site of interaction between SKLB610 and ABCG2 was determined by docking SKLB610 into the drug-binding pocket within the transmembrane domain of the human ABCG2 structure (PDB: 6VXH) [49]. The best ligand binding pose was chosen with the calculated binding energy of − 73 kcal/mol. As shown in Fig. 6, hydrophobic interactions between the aromatic rings of SKLB610 and the residues Met⁵⁴⁹ and Val⁵⁴⁶ as well as interaction of the trifluoromethyl moiety with LEU⁴⁰⁵ were predicted within the transmembrane domain of ABCG2.

Fig. 6.

Docking of SKLB610 in the drug-binding pocket of ABCG2. Binding modes of SKLB610 with the protein structure of ABCG2 protein structure (PDB: 6VXH) were predicted by Accelrys Discovery Studio 4.0 software as described in Section 2. SKLB610 (yellow color) is presented in stick representation and the atoms for interacting amino acid residues are colored carbon-gray, hydrogen-light gray, oxygen-red, nitrogen-blue and fluorine-cyan. Proposed interactions are presented as dotted lines.

4. Discussion

The development of MDR associated with the overexpression of drug transporters such as ABCB1 or ABCG2 remains a significant obstacle in cancer chemotherapy [3,19]. Patients with multidrug-resistant cancers display reduced responses to conventional cytotoxic anticancer drugs. Unfortunately, the progress of developing selective and potent synthetic inhibitors of these drug transporters has been hindered by unforeseen toxicity [3,25–28]. We and others have discovered in recent years that the overexpression of ABCB1 or ABCG2 could be responsible for reduced susceptibility of cancer cells to certain TKIs [46,54,72–74]. Some of the TKIs were found to reverse MDR mediated by ABCB1 and/or ABCG2 in cancer cells [50,58,61,75–78]. Most recently, studies have demonstrated that branebrutinib [78], almonertinib [77], erdafitinib [61], and sitravatinib [50] inhibit drug efflux mediated by ABCB1, whereas rociletinib [79] and cabozantinib [80] inhibit drug efflux mediated by ABCG2. Notably, results of several combination therapy trials have shown the advantages of combination therapy of TKIs with conventional anticancer drugs over monotherapy in patients with advanced pancreatic cancer [81,82] or human epidermal growth factor receptor 2 (HER2)-positive advanced breast cancer [83,84]. More importantly, TKIs could potentiate the antiproliferative effect of cytotoxic anticancer drugs against multidrug-resistant cancers [85]. Collectively, these findings indicate that further studies are warranted to investigate combination therapies of TKIs and cytotoxic anticancer drugs to overcome ABC transporter-mediated MDR.

In this study, we investigated the potential interactions between the multi-targeted TKI SKLB610, ABCB1 and ABCG2 in multidrug-resistant cancer cell lines overexpressing human ABCB1 or ABCG2. First, we found that SKLB610 was cytotoxic in all tested cancer cell lines (Table 1), with IC₅₀ values in the range of 10–32 μM, which are comparable to the previously reported IC₅₀ values against 12 other cancer cell lines [29]. More importantly, the drug-sensitive parental cancer cell line and the multidrug-resistant sublines overexpressing ABCB1 or ABCG2 were equally sensitive to SKLB610, indicating that SKLB610 is not rapidly transported out of cancer cells by either ABCB1 or ABCG2. Our results suggest that the overexpression of ABCB1 or ABCG2 is unlikely to be one of the key contributing factors in the development of SKLB610 resistance in patients, but the clinical effects of prolonged treatment of patients with SKLB610 remain to be determined. Next, we examined the potential chemosensitizing effect of SKLB610 on drug resistance mediated by ABCB1 and ABCG2 in cells overexpressing ABCB1 or ABCG2. We found that SKLB610 had no significant effect on ABCB1-mediated resistance to colchicine, vincristine, or doxorubicin in ABCB1-overexpressing cancer cells or in HEK293 cells transfected with human ABCB1 (Table 2). In contrast, SKLB610 significantly sensitized ABCG2-overexpressing multidrug-resistant H460-MX20 and S1-MI-80 cancer cells, as well as HEK293 cells transfected with human ABCG2, to mitoxantrone, SN-38, and topotecan in a concentration-dependent manner (Fig. 1), signifying that the activity of SKLB610 is more selective to ABCG2 as compared to ABCB1. Interestingly, although studies have identified TKIs that are capable of re-sensitizing ABCG2-overexpressing multidrug-resistant cancer cells to anticancer drugs, most of these TKIs are not ABCG2-selective due to a substantial overlapping substrate specificity between ABCB1 and ABCG2 [44,50,60,66,86,87]. It should be noted that Cao et al. reported that SKLB610 at 10 μM inhibited the activity of VEGFR2, FGFR2, and PDGFR by 97%, 65%, and 55% in biochemical kinase assays [29], while at the same concentration, we demonstrated that SKLB610 inhibited ABCG2-mediated efflux of a fluorescent substrate drug (Fig. 2D) and enhanced the extent of drug-induced apoptosis (Fig. 4) in ABCG2-overexpressing multidrug-resistant cancer cells. Furthermore, SKLB610 had no significant effect on the total protein level of ABCG2 (Fig. 3) in ABCG2-overexpressing cancer cells, suggesting that SKLB610 resensitizes ABCG2-overexpressing cancer cells to drug-induced apoptosis (Fig. 4) and cytotoxicity (Table 3) by attenuating the drug transport function of ABCG2. Interaction of SKLB610 at the drug-binding pocket of ABCG2 was confirmed by its stimulatory effect on the ATPase activity in a concentration-dependent manner (Fig. 5), and the in silico docking analysis of SKLB610 binding to the inward-open conformation of human ABCG2 (Fig. 6). Our data indicate that SKLB610 interacts with multiple residues within the substrate-binding pocket of ABCG2 and competes with the binding of substrate drugs at the same site, thus inhibiting the transport function of ABCG2 (Fig. 7).

Fig. 7.

Simplified schematic illustration of SKLB610 re-sensitizing ABCG2-overexpressing multidrug-resistant cancer cells to cytotoxic anticancer drugs by blocking the drug efflux function of ABCG2. The intracellular concentration of ABCG2 substrate drugs (represented by blue circles) in ABCG2-overexpressing cells is actively reduced by ABCG2-mediated drug transport. In contrast, in the presence of SKLB610 (red triangle), ABCG2-mediated drug efflux is attenuated by SKLB610 outcompeting the binding of ABCG2 substrate drug at the same drug-binding pocket, thus restoring the intracellular accumulation of cytotoxic drugs in ABCG2-overexpressing multidrug-resistant cancer cells.

In summary, we demonstrated that SKLB610 could selectively sensitize ABCG2-overexpressing multidrug-resistant cancer cells by modulating the drug transport activity of ABCG2. While undesirable responses or altered pharmacokinetics of concomitant anticancer drugs may occur in combination therapies [3,88–91], our findings revealed an additional pharmacological activity of SKLB610 that could potentially be used in combination therapy with cytotoxic anticancer drugs against multidrug-resistant cancers associated with the overexpression of ABCG2.

Abbreviations:

ABC

ATP-binding cassette

MDR

multidrug resistance

VEGFR

vascular endothelial growth factor

Vi

sodium orthovanadate

FR

fold-reversal