Recent advances regarding the role of ABC subfamily C member 10 (ABCC10) in the efflux of antitumor drugs

Abstract

ABCC10, also known as multidrug-resistant protein 7 (MRP7), is the tenth member of the C subfamily of the ATP-binding cassette (ABC) superfamily. ABCC10 mediates multidrug resistance (MDR) in cancer cells by preventing the intracellular accumulation of certain antitumor drugs. The ABCC10 transporter is a 171-kDa protein that is localized on the basolateral cell membrane. ABCC10 is a broad-specificity transporter of xenobiotics, including antitumor drugs, such as taxanes, epothilone B, vinca alkaloids, and cytarabine, as well as modulators of the estrogen pathway, such as tamoxifen. In recent years, ABCC10 inhibitors, including cepharanthine, lapatinib, erlotinib, nilotinib, imatinib, sildenafil, and vardenafil, have been reported to overcome ABCC10-mediated MDR. This review discusses some recent and clinically relevant aspects of the ABCC10 drug efflux transporter from the perspective of current chemotherapy, particularly its inhibition by tyrosine kinase inhibitors and phosphodiesterase type 5 inhibitors.

One known cause of multidrug resistance (MDR) to cancer chemotherapy is the overexpression of ATP-binding cassette (ABC) transporters on the membranes of cancer cells[1]. ABC transporters mediate an energy-dependent efflux of structurally and functionally unrelated chemotherapeutic compounds out of cancer cells, significantly decreasing the probability of successful treatment[2]. In mammals, ABC transporters have been divided into subfamilies A-G based on genome sequence similarities[3]. Studies to date have consistently focused on the three major ABC transporters involved in producing MDR in most cancer cells: ABCB1 (MDR1, also known as P-glycoprotein or P-gp), ABCC1 (also known as multidrug-resistant protein 1 or MRP1), and ABCG2 (also known as breast cancer-resistant protein or BCRP)[3]^–[7]. However, ABCC10 (MRP7) has only been recently classified as a member of the ABC superfamily that confers the MDR phenotype to cancer cells. This review briefly discusses the emergence of ABCC10 as an important mediator of MDR and the compounds that overcome ABCC10-induced resistance to specific antitumor drugs (Table 1).

Table 1. A timeline of the major discoveries associated with ABCC10.

Year	Major discovery
2001	Discovery of ABCC10 as a new member of ABCC subfamily[9]
2002	cDNA cloning and genomic organization of the murine Abcc10A and Abcc10B[57]
2003	ABCC10 is a lipophilic anion transporter involved in phase III of detoxification[12]
2004	First report of ABCC10 as a resistance factor to anticancer drugs like paclitaxel, docetaxel, vincristine, and vinblastine[10]
	ABCC10 gene expression found to be the highest in the pancreas[14]
2005	Peptide derived from ABCC10 reported as an immunoregulator[58]
2007	ABCC10 is a resistance factor for docetaxel in salivary gland adenocarcinoma[21]
2008	ABCC10 is established as a biomarker for paclitaxel resistance in non-small cell lung cancer[18]
2009	ABCC10 is a resistance factor for vinorelbine in non-small cell lung cancer[17]
	ABCC10 is a resistance factor for epothilone B, and ABCC10 transport does not involve glutathione, unlike ABCB1 and ABCC1[22]
	Cepharanthine, a herbal extract, reverses ABCC10-mediated paclitaxel resistance[24]
	BCR-Abl tyrosine kinase inhibitors, imatinib and nilotinib, inhibit efflux function of ABCC10 efflux transporter[28]
2010	Epidermal growth factor receptor tyrosine kinase inhibitors, erlotinib and lapatinib, inhibit efflux function of ABCC10 efflux transporter[33]
2011	Docetaxel intermittently increases simultaneous ABCC10 and ABCB1 gene expression[59]
	ABCC10 transcript was detected in acute myeloid leukemia cell lines[19]
	ABCC10 termed as "endogenous resistance factor" for taxanes that confer paclitaxel resistance in vivo in Abcc10-knockout mice[12]
	ABCC10 single nucleotide polymorphisms (rs9349256 and rs2125739) and their haplotypes are associated with kidney tubular dysfunction[48]
2012	ABCC10 single nucleotide polymorphism rs2125739 is associated with nevirapine-induced hepatotoxicity[54]
	Phosphodiesterase 5 inhibitors, such as sildenafil and vardenafil, reverse MDR mediated by ABCC10[41]
	First murine ABCC10-paclitaxel resistance xenograft model developed[60]
	Localization of ABCC10 was found on basolateral cell membrane[11]
2013	Tariquidar, a third-generation ABCB1 inhibitor, reverses ABCC10-mediated MDR[61]
	Tandutinib, an FMS-like tyrosine kinase 3 inhibitor, blocks efflux function of ABCC10 efflux transporter[47]

ABC, ATP-binding cassette superfamily divided into A-G subfamilies; MDR, multidrug resistance.

ABCC10/MRP7 Transporter

The human ABCC10 gene is located on chromosome 6p12[8]^,[9]. ABCC10 is a 171-kDa protein that contains three membrane-spanning domains (MSDs) and two nucleotide-binding domains (NBDs) (Figure 1)[10]. ABCC10 belongs to the class of long ABCCs, such as ABCC1, ABCC2, ABCC3, and ABCC6, and is located on the basolateral cell surface[10]^–[13]. Using reverse transcription-polymerase chain reaction (RT-PCR), a low level of ABCC10 transcript expression has been found in the skin, testes, spleen, stomach, colon, kidneys, heart, and brain[8]^,[9]. In addition, the ABCC10 transcript is expressed (in order of highest to lowest) in the pancreas, liver, placenta, lungs, kidneys, brain, ovaries, lymph nodes, spleen, heart, leukocytes, and colon[14]. ABCC10 mRNA is highly expressed in various tissues, including the kidneys, brain, and colon, suggesting that it is involved in the transport of drugs and other endogenous molecules[15]. Kao et al.[16] have discovered a splice variant of ABCC10 that is truncated at its NH₂ terminus and has a 15-amino acid deletion between MSD2 and NBD2.

Figure 1. Structure of ABCC10.

ABCC10 is located on the basolateral cell membrane and consists of 3 membrane-spanning domains (MSDs) and 2 nucleotide-binding domains (NBDs).

Chen et al.[12] have reported that ABCC10 transports leukotriene C4, but not glycocholic acid, taurocholic acid, methotrexate, folic acid, cyclic adenosine monophosphate, or cyclic guanosine monophosphate, which are substrates for other MRP family members. However, the transport of estradiol-17β-D-glucuronide (E₂17βG), a glucuronide conjugate and the only physical element other than leukotriene C4 to be transported by ABCC10, is competitively inhibited by amphiphiles, such as leukotriene C4, glycolithocholate 3-sulfate, and MK571, as well as lipophilic compounds, such as cyclosporine A[12]. The transfection of HEK293 cells with the ABCC10 gene confers resistance to various chemotherapeutic drugs, including docetaxel, paclitaxel, vincristine, vinblastine, cytarabine, gemcitabine, 2′,3′-dideoxycytidine, 9-(2-phosphonyl methoxyethyl)adenine (PMEA), and epothilone B[10]^,[13]. Specifically, the presence of ABCC10 is significantly associated with vinorelbine, and paclitaxel resistance in non-small cell lung cancer (NSCLC)[17]^,[18]. In acute myeloid leukemia (AML) cell lines, ABCC10 protein expression was detected (in highest to lowest order) in ML-2, NB4, MV4, and Kasumi-1 cell lines[19]. The ABCC10 transcript has been found in breast, lung, colon, ovarian, and pancreatic tumor samples, although the interpretation of these studies may be limited due to their small sample size[13]^,[14]. ABCC10 transcript has been detected in the HepG2 liver cancer cell line and two prostate cancer cell lines, CWR22Rv1 and TSU-PR1[20]. ABCC10 transcript up-regulation has also been shown in salivary gland adenocarcinoma[21].

The ectopic expression of ABCC10 confers resistance to taxanes, which is of particular interest because aside from ABCB1, none of the established cellular efflux pumps produce resistance to clinically used taxanes[22]. Indeed, the role of ABCC10 in taxane resistance is noticeable, as ABCC10 produces high levels of resistance to paclitaxel and docetaxel (116- and 46-fold, respectively) in ABCB1-deficient fibroblasts[22]. In another study, fibroblasts from Abcc10-knockout mice have been shown to be taxane-resistant[13]. In the same study, the mortality of the Abcc10-knockout mice was found to be significantly increased due to neutropenia and marked bone marrow toxicity after paclitaxel treatment[13]. These results suggest that ABCC10 protects cells against paclitaxel toxicity. Recently, it was reported that ABCC10 and ABCB1 gene expression is induced in chemoresistant and chemosensitive tumors by intermittent docetaxel treatment[23], implying that the dosing schedule of chemotherapy affects the development of resistance.

ABCC10 Modulators

To circumvent ABCC10-induced MDR, various modulators that could significantly reverse the resistance mediated by ABCC10 by increasing the accumulation and decreasing the efflux of antitumor drugs have been tested (Table 2). Various compounds that function as ABCC10 modulators, albeit with different mechanisms of action, will be subsequently discussed (Figure 2).

Table 2. Tyrosine kinase inhibitors (TKIs) and ABCC10 modulators.

Compound	Concentrations	Effect on ABCC10 substrate drug	Effect on accumulation of [3H]-paclitaxel	Effect on efflux of [3H]-paclitaxel	Effect on ABCC10 protein expression
Cepharanthine[24]	2 µmol/L	+	↑	↓	↔
Imatinib[28]	1, 2.5, 5 µmol/L	+	↑	↓	↔
Nilotinib[28]	1, 2.5, 5 µmol/L	+	↑	↓	↔
Lapatinib[33]	0.625, 1.25, 2.5 µmol/L	+	↑	↓	↔
Erlotinib[33]	0.625, 1.25, 2.5 µmol/L	+	↑	↓	↔
Sorafenib[19]	5 µmol/L	+	N/A	N/A	N/A
Sildenafil[41]	1.25, 2.5, 5 µmol/L	+	↑	↓	↔
Vardenafil[41]	1.25, 2.5, 5 µmol/L	+	↑	↓	↔
Tariquidar[61]	0.1, 0.3 µmol/L	+	↑	↓	↓ (>24 h treatment)
Tandutinib[47]	5, 10 µmol/L	+	↑	↓	↔

(+): this compound shows reversal activity for the ABCC10 transporter; (-): this compound shows no reversal activity for the ABCC10 transporter; (↑): increase or up-regulate; (↓): decrease or down-regulate; (↔): no significant alterations; (N/A): not applicable.

Figure 2. Modulation of ABCC10.

Overexpression of ABCC10 leads to efflux of antitumor drugs out of cancer cells. Modulation of ABCC10 by different classes of compounds, such as lapatinib, erlotinib, imatinib, nilotinib, sorafenib, tandutinib, sildenafil, and vardenafil, leads to increased accumulation and decreased efflux of antitumor drugs, thus rendering them more efficacious. BCR-Abl, breakpoint cluster region-Abelson; EGFR, endothelial growth factor receptor; FLT3, FMS-like tyrosine kinase 3; PDE5, phosphodiesterase type 5.

Cepharanthine

Cepharanthine [6′12′-dimethoxy-2,2′-dimethyl-6,7-[methylenebis (oxy) oxyacan-than] is isolated from the herbal plant Stephania cepharantha Hayata. Cephanranthine, at 2 µmol/L, has been shown to completely reverse the paclitaxel resistance of ABCC10-transfected HEK293 cells[24]. In contrast, cepharanthine does not significantly alter paclitaxel transport in empty vector-transfected HEK293 cells. Accumulation and efflux studies have indicated that cepharanthine significantly increases the intracellular accumulation of [³H]-paclitaxel and inhibits the efflux of [³H]-paclitaxel from ABCC10-transfected cells, but not in the cells lacking the ABCC10 transporter[24]. The transport of E₂17βG is competitively inhibited by cepharanthine with a Ki value of 4.86 µmol/L[24].

Imatinib and nilotinib

Imatinib and nilotinib are inhibitors of the tyrosine kinase (TK) breakpoint cluster region-Abelson (BCR-Abl) protein and stem cell factor receptor (c-kit), a class III receptor TK[25]. The abnormal translocation of the BCR-Abl gene is associated with a deregulation of TK function, and its expression subsequently leads to chronic myeloid leukemia (CML)[26]. Previous results from our laboratory suggest that nilotinib significantly inhibits the drug efflux functions of ABCB1 and ABCG2[27]. Subsequently, it has been reported that imatinib and nilotinib reverse ABCC10-mediated MDR[28]. Western blotting analysis has indicated that both imatinib and nilotinib do not significantly affect ABCC10 expression. However, imatinib and nilotinib have been shown to enhance the sensitivity of ABCC10-transfected HEK293 cells to two established ABCC10 substrates, paclitaxel and vincristine, in a dose-dependent fashion[28]. Imatinib and nilotinib significantly increase the accumulation of [³H]-paclitaxel in ABCC10-transfected cells almost to the level of control cells in the absence of imatinib and nilotinib. The efflux of [³H]-paclitaxel from ABCC10-transfected cells is also significantly inhibited by imatinib and nilotinib. In conclusion, imatinib and nilotinib reverse ABCC10-mediated MDR through inhibition of the drug efflux function. If these results can be clinically translated, they suggest that nilotinib and imatinib could be used in combination with paclitaxel to surmount MDR in certain types of cancers[28].

Lapatinib and erlotinib

Lapatinib is a reversible, small-molecule inhibitor of endothelial growth factor receptor (EGFR/HER1) and HER2 receptor approved by the Food and Drug Administration (FDA) in the United States for the treatment of HER2-positive metastatic breast cancers[29]. Erlotinib, another selective small-molecule inhibitor of EGFR, is used clinically for the treatment of chemotherapy resistance in patients with advanced NSCLC as well as advanced pancreatic cancer in combination with gemcitabine[30]. Previously, it has been reported that lapatinib and erlotinib reverse ABCB1- and ABCG2-mediated MDR[31]^,[32] and that they significantly reduce ABCC10-mediated MDR[33]. For example, lapatinib and erlotinib produce a dose-dependent increase in the sensitivity of ABCC10-transfected HEK293 cells to docetaxel, paclitaxel, vinblastine, and vinorelbine[33]. Furthermore, lapatinib and erlotinib significantly increase the intracellular accumulation of [³H]-paclitaxel by inhibiting the efflux of [³H]-paclitaxel in HEK293/ABCC10 cells[33]. Lapatinib produces a significantly greater inhibitory effect on ABCC10-mediated MDR than erlotinib, which may result from the action of lapatinib on the TKs of EGFR and HER2 receptors, whereas erlotinib only inhibits EGFR[34]. Additionally, Western blotting analysis has indicated that ABCC10 protein expression is not significantly altered by lapatinib or erlotinib. In conclusion, lapatinib and erlotinib reverse ABCC10-mediated MDR through inhibition of the drug efflux function of ABCC10. These findings suggest that these compounds could be used as adjuvant chemotherapeutic drugs in clinical practice[33], although this remains to be verified.

Sorafenib

Sorafenib is an inhibitor of C-rapidly accelerated fibrosarcoma (C-raf), B-raf, c-kit, FMS-like tyrosine kinase 3 (FLT3), platelet-derived growth factor receptor-α and -β, and vascular endothelial growth factor receptor-1, -2, and -3[35]. At a clinically achievable plasma concentration of 10 µmol/L, sorafenib significantly inhibits multiple additional receptor TKs and intracellular kinases[35]. Sorafenib is approved for the treatment of advanced renal cell carcinoma and hepatocellular carcinoma (HCC) and is being evaluated for the treatment of AML[36]. Sorafenib inhibits the cellular efflux function of ABC transporters[37] and re-sensitizes cells to cytarabine, which is a substrate of ABCC10 and ABCC11 transporters[22]. Consequently, we can make the assumption that sorafenib may inhibit the transporter-mediated efflux of cytarabine from AML cells and enhance the accumulation of cytarabine-phosphorylated metabolites, leading to an increase in the sensitivity of AML blast cells to cytarabine. Indeed, sorafenib significantly increases the cellular accumulation of cytarabine and its metabolites, resulting in additive or synergistic anti-leukemic activity[37]. To determine if the expression of ABCC10 and ABCC11 was responsible for the sorafenib-induced cellular retention of cytarabine, gene and protein expression were assessed in 10 human AML cell lines; Western blotting demonstrated that the ABCC10 protein is expressed in 4 of the 10 AML cell lines[19].

Sildenafil and vardenafil

Sildenafil and vardenafil are selective, competitive inhibitors of the phosphodiesterase type 5 (PDE5) enzyme and inhibit the biotransformation of cyclic guanosine monophosphate (cGMP)[38]. These compounds are used to treat erectile dysfunction and certain types of pulmonary hypertension[38]. Sildenafil and vardenafil increase the intracellular accumulation of cGMP, leading to the relaxation of vascular smooth muscle[38]. Currently, several groups have evaluated the anticancer activities of PDE5 inhibitors. Previous results from our laboratory indicate that sildenafil and vardenafil significantly reverse ABCB1-mediated MDR[39]^,[40]. Furthermore, sildenafil and vardenafil produce a significant dose-dependent reversal of ABCC10-mediated MDR by enhancing the sensitivity of HEK293/ABCC10 cells to paclitaxel, docetaxel, and vinblastine[41]. Sildenafil and vardenafil, at 5 µmol/L, significantly increase the accumulation of [³H]-paclitaxel in HEK/ABCC10 cells. Incubating HEK293/ABCC10 cells with sildenafil or vardenafil (5 µmol/L) also significantly inhibits the efflux of [³H]-paclitaxel. Immunoblotting and immunofluorescence analyses showed that ABCC10 protein expression and localization in HEK293/ABCC10 cells are not significantly altered by 5 µmol/L of sildenafil or vardenafil for up to 72 h. In conclusion, sildenafil and vardenafil reverse ABCC10-mediated paclitaxel resistance, and this is most likely due to the inhibition of the efflux of paclitaxel via ABCC10. These findings suggest that sildenafil or vardenafil, in combination with other antitumor drugs, may be useful in the treatment of certain MDR cancers[41].

Tariquidar

Tariquidar is a third-generation ABCB1 inhibitor with high affinity (K_d = 5 nmol/L), low toxicity, and increased selectivity for ABCB1[42]. Tariquidar is 10- to 30-fold more potent than second-generation ABCB1 inhibitors, such as PSC833[43], and is 100- to 1,000-fold more potent than first-generation ABCB1 inhibitors, such as cyclosporine A and verapamil[44]. Tariquidar is an inhibitor of ABCG2 but does not interact with ABCC1, indicating that it may not be as specific as previously thought[45].

Currently, only a few inhibitors of ABCC10 have been identified, and none of them have progressed to clinical trials. A recent in vitro study reported that tariquidar produces a significant dose-dependent increase in the sensitivity of ABCC10-transfected HEK293 cells to the ABCC10 substrates paclitaxel, docetaxel, vincristine, vinblastine, and vinorelbine. In contrast, tariquidar does not significantly alter the efficacy of the aforementioned substrates in empty vector-transfected HEK293 cells. Tariquidar completely reversed ABCC10-mediated MDR at 0.3 µmol/L, making it a potent ABCC10-reversing agent identified thus far. Furthermore, accumulation assays have demonstrated that tariquidar significantly increases the intracellular accumulation of [³H]-paclitaxel, as well as boron-dipyrromethene (BODIPY)-paclitaxel, in ABCC10-transfected HEK293 cells. The enhanced intracellular accumulation of [³H]-paclitaxel is partly due to the rapid and direct inhibition of ABCC10-mediated drug efflux by tariquidar. Tariquidar also produces a down-regulation of ABCC10 protein expression after 24 h of incubation. Incubating cells with tariquidar for 24 h down-regulates ABCC10 protein expression in a time- and dose-dependent manner. Consequently, the reversal effect of tariquidar on ABCC10-mediated MDR in HEK293/ABCC10 cells is due to the inhibition of ABCC10 efflux function and down-regulation of ABCC10 protein expression. Importantly, the down-regulation of ABCC10 protein expression was not caused by the reduction in ABCC10 mRNA levels or the cellular translocation of ABCC10. In conclusion, tariquidar could be used in combination with specific anti-cancer drugs to treat certain types of cancer, although this remains to be proven.

Tandutinib

Tandutinib is a novel quinazoline-based inhibitor of FLT3 (a transmembrane receptor in the tyrosine kinase family), the platelet-derived growth factor receptor, and c-kit. Tandutinib is approved for the treatment of AML and is currently in phase II clinical trials[46]. A recent study showed that tandutinib reverses ABCC10-mediated MDR[47]. For example, tandutinib significantly sensitizes ABCC10-expressing cells to paclitaxel and vincristine[47]. Moreover, accumulation and efflux experiments have indicated that tandutinib significantly enhances the intracellular accumulation of [³H]-paclitaxel and inhibits the efflux of [³H]-paclitaxel from HEK293/ABCC10 cells[47]. However, Western blotting analysis has indicated that tandutinib does not significantly affect ABCC10 protein expression. These findings suggest that clinical studies should be considered to test the efficacy of tandutinib to reverse ABCC10-mediated MDR in patients[47].

ABCC10 Genetic Variants

Currently, relatively few studies have been published regarding the effects of genomic ABCC10 mutations on the interactions of the ABCC10 protein with various compounds. Pushpakom et al.[48] have found some common variants in ABCC10 in human immunodeficiency virus (HIV)-positive patients receiving tenofovir-containing regimens. This study is the first genetic association analysis reported for ABCC10[48]. Two single nucleotide polymorphisms (SNPs) and their haplotype were significantly associated with renal tubular dysfunction: rs9349256 located in intron 4 and rs2125739, a nonsynonymous SNP, located in exon 12 that results in an amino acid change (Ile920Thr)[48]. Although the functional effects of the polymorphisms identified in ABCC10 are not known, a bioinformatic approach using FastSNP software (http://fastsnp.ibms.sinica.edu.tw/) found rs2125739 in a putative splice site[48]. Previously, splice site polymorphisms have been shown to affect pre-mRNA splicing and may cause the splicing apparatus to use nearby cryptic splice sites or skip exons, leading to altered ABCC10 expression[49]. Based on HapMap data (www.hapmap.org), the minor allele frequency for rs2125739 in various populations was reported as follows: Northern and Western European ancestry (26.7%), Sub-Saharan African (34.2%), Han Chinese (3.4%), and Japanese (8.0%)[48]. ABCC10 rs9349256 is associated with urine phosphorus wasting and β2-microglobulinuria, which are markers for tubular dysfunction[48]^,[50]^,[51]. This observation is interesting, as urine phosphorus wasting is among the 3 criteria defining Fanconi syndrome and tenofovir is the drug that is most associated with this syndrome[52]^,[53], which leads to bone demineralization and osteomalacia[52]. Liptrott et al.[54] have investigated the contribution of ABCC10 SNPs to variability in nevirapine plasma concentration. An exonic SNP in ABCC10, rs2125739, was significantly associated with nevirapine plasma concentration in Caucasians; the variant allele was significantly more prevalent in those with lower nevirapine plasma concentrations[54]. The minor allele frequency for this SNP was similar in both the Caucasians and African-Americans represented in this cohort. Indeed, the same allele was also more prevalent in African-Americans with lower nevirapine plasma concentration, although this association was not statistically significant[54]. This finding is of interest as lower nevirapine plasma concentrations have been reported to be associated with reduced viral suppression[55]^,[56]. Hence, these SNPs can provide insight for developing modulation strategies of ABCC10 expression to augment the efficacy of therapeutic drugs.

Conclusions

Despite previous reports showing the widespread tissue expression of ABCC10[11]^,[14], it remains one of the least-characterized ABC family members. It would be of great interest to characterize the mechanism of regulation of ABCC10 expression. Several compounds have recently been shown to interact with ABCC10; thus, ABCC10 can confer resistance to these compounds, both at the tumor cell level and by decreasing their oral bioavailability[28]. Furthermore, the role of ABCC10 needs to be elucidated in both cancer cells and cancer stem cells. Although efflux pump inhibitors have yet to be established as a clinical strategy for sensitizing tumors to chemotherapeutic substances, genetic variations should be considered in the design of new therapeutic strategies.