Diurnal expression of MRP4 in bone marrow cells underlies the dosing-time dependent changes in the oxaliplatin-induced myelotoxicity

The expression and function of some xenobiotic transporters varies according to the time of day, causing the dosing time-dependent changes in drug disposition and toxicity. Multidrug resistance-associated protein-4 (MRP4), an ATPbinding cassette (ABC) efflux transporter encoded by the Abcc4 gene, is highly expressed in bone marrow cells (BMCs) and protects them against xenobiotics, including chemotherapeutic drugs. In this study, we demonstrated that MRP4 was responsible for the extrusion of oxaliplatin (L-OHP), a platinum (Pt)-based chemotherapeutic drug, from BMCs of mice, and that the efflux transporter expression exhibited significant diurnal variation. Therefore, we investigated the relevance of the diurnal expression of MRP4 in BMCs for L-OHP-induced myelotoxicity in mice maintained under standardized light/dark cycle conditions. After intravenous injection of L-OHP, the Pt content in BMCs varied according to the injection time. Lower Pt accumulation in BMCs was detected in mice after injection of L-OHP at the mid-dark phase, during which the expression levels of MRP4 increased. Consistent with these observations, the myelotoxic effects of L-OHP were attenuated when mice were injected with L-OHP during the dark phase. This dosing schedule also alleviated the L-OHP-induced reduction of the peripheral white blood cell count. The present results suggest that the myelotoxicity of L-OHP is attenuated by optimizing the dosing schedule. Diurnal expression of MRP4 in BMCs is associated with the dosing time-dependent changes in L-OHP-induced myelotoxicity.


Results
Role of MRP4 in the extrusion of L-OHP from BMCs. In addition to MRP4, the mRNA expression of major xenobiotic ABC transporters, P-gp (encoded by Abcb1a or Abcb1b), BCRP (Abcg2), and MRP2 (Abcc2), was detected in BMCs (Fig. 1a). To investigate the role of these ABC transporters in the extrusion of L-OHP from BMCs, we prepared BMCs from mouse femora, and cells were treated with 50 μM L-OHP in the presence of selective inhibitors for P-gp, BCRP, MRP2, or MRP4. After treatment of BMCs with L-OHP in the presence of the selective MRP4 inhibitor ceefourin 30 , intracellular Pt content was significantly higher than that in vehicle (0.001% DMSO)-treated cells (Fig. 1b). Treatment of BMCs with ceefourin also resulted in the accumulation of 6-mercaptopurine (6-MP), a typical substrate of MRP4 (P < 0.01, Fig. 1c). On the other hand, no significant accumulation of Pt was observed when BMCs were treated by L-OHP together with the P-gp inhibitor PSC833 31 , BCRP inhibitor Ko143 32 , or MRP2 inhibitor bromosulfalein 33 (Fig. 1b).
To further investigate the role of MRP4 in the extracellular efflux of L-OHP, we prepared mouse MRP4expressing NIH3T3 cells (Fig. 1d), which were also treated with 50 μM L-OHP in the presence or absence of PSC833, Ko143, bromosulfalein, or ceefourin. After treatment with L-OHP in the absence of ABC transporter inhibitors, intracellular Pt accumulation in MRP4-expressing cells was significantly lower than that observed in mock-transduced cells (P < 0.01, Fig. 1e). The low Pt accumulation in L-OHP-treated MRP4-expressing cells was abrogated by sodium orthovanadate, a Na + /K + -ATPase inhibitor ( Supplementary Fig. 1), suggesting ATPdependent extrusion of L-OHP by MRP4. After treatment of MRP4-expressing cells with L-OHP in the presence of ceefourin, intracellular Pt content was significantly increased compared with that in vehicle-treated MRP4expressing cells (P < 0.01, Fig. 1e). As observed in BMCs, treatment of MRP4-expressing cells with ceefourin also caused significant accumulation of 6-MP (P < 0.05, Fig. 1f). On the other hand, PSC833, Ko143, and bromosulfalein had negligible effects on Pt content in L-OHP-treated MRP4-expressing cells (Fig. 1e). These results suggest that MRP4 is responsible for the extracellular efflux of L-OHP.
Dosing time-dependent changes in L-OHP disposition in mice. Next, we explored the possibility that the expression and function of MRP4 vary according to the time of day. To achieve this, mice were maintained under standardized light/dark cycle conditions (Zeitgeber time 0 [ZT0], lights on; ZT12 lights off). The mRNA and membrane fractions of BMCs were prepared at six different time points. Accumulation of Pt in BMCs after L-OHP treatment was also assessed by preparing cells from mice at ZT6 and ZT18. The mRNA levels of Abcc4 in BMCs exhibited a significant diurnal oscillation, with a peak level during the light phase (P < 0.05, Fig. 2a). Significant diurnal oscillation was also detected in the protein levels of MRP4 in BMCs, but the oscillation in the MRP4 protein levels was delayed by approximately 12 h relative to the Abcc4 mRNA rhythm (P < 0.01, Fig. 2b). The expression of MRP4 at ZT18 was approximately twofold higher than that at ZT6. Indeed, the accumulation of Pt in BMCs prepared at ZT18 was significantly lower than that in cells prepared at ZT6 (P < 0.01, Fig. 2c). The time-dependent difference in the Pt accumulation in L-OHP-treated BMCs disappeared when cells were treated with the MRP4 inhibitor ceefourin. Taken together, these results suggest that diurnal expression of MRP4 induces the time-dependent change in the Pt accumulation in L-OHP-treated BMCs.
To further explore the relevance of diurnal expression of MRP4 in BMCs for the disposition of L-OHP, mice were injected with a single dose of L-OHP (17.5 mg/kg, i.v.) at ZT6 or ZT18. Although there was no significant dosing time-dependent difference in the serum Pt concentration after L-OHP injection (Fig. 2c), that in BMCs varied according to the injection time of L-OHP. A lower Pt content in BMCs was detected in mice after injection of L-OHP at ZT18 (P < 0.05, Fig. 2d), during which the expression level of MRP4 increased (Fig. 1b). These results suggest a close relationship between the diurnal expression of MRP4 and Pt accumulation in BMCs after injecting L-OHP. www.nature.com/scientificreports/ Dosing time-dependent changes in L-OHP-induced myelotoxicity. To examine whether the dosing time-dependent difference in Pt accumulation in BMCs after the administration of L-OHP is also associated with its myelotoxicity, the number of viable BMCs was assessed after injection of L-OHP (20 mg/kg, i.v.) www.nature.com/scientificreports/ at ZT6 or ZT18. At 72 h after the injection of L-OHP, a significant reduction of the number of viable BMCs was detected at both dosing times (P < 0.01, Fig. 3a). However, the myelocytotoxic effects of L-OHP in mice injected with L-OHP at ZT18 were significantly alleviated compared with those in mice injected with the drug at ZT6 (P < 0.05). Lineage is a mature blood cell marker to differentiate mature hemocytes, including T cells, B cells, monocytes, macrophages, granulocytes, and neutrophils 34 . After injecting L-OHP (20 mg/kg, i.v.) at ZT6 or ZT18, the number of lineage-positive and -negative BMCs was assessed as differentiated and undifferentiated cells, respectively. The number of viable differentiated cells was not significantly reduced after L-OHP injection at either dosing time ( Fig. 3b right panel). Although a significant reduction of the number of viable undifferentiated cells was detected after L-OHP injection at both dosing times, the survival rate of undifferentiated cells after L-OHP injection at ZT18 was significantly higher than that after the drug administration at ZT6 (P < 0.05, Fig. 3b left panel). The expression of MRP4 slightly decreased in differentiated BMCs (Fig. 3c), but still exhibited significant diurnal variations in both differentiated and undifferentiated cells (P < 0.05 respectively, Fig. 3c). Therefore, the myelotoxic effects of L-OHP may be associated with the decrease in the number of undifferentiated BMCs.
Dosing time-dependent difference in L-OHP-induced leukopenia. As hematopoietic stem cells in bone marrow proliferate and differentiate repeatedly to produce peripheral blood cells 35 , chemotherapeutic drug-induced myelotoxicity is often associated with leukopenia 36,37 . We thus examined whether L-OHP-induced leukopenia changed depending on the dosing time. The number of peripheral WBCs in mice decreased 3 days after L-OHP injection, and the peripheral WBC count varied according to the time of day in both humans and experimental rodents 17,18,38 . Therefore, mice were injected with a single dose of L-OHP (20 mg/kg, i.v.) at ZT6 www.nature.com/scientificreports/ or ZT18, and then the number of peripheral WBCs was assessed from 72 to 92 h (4-h interval) after L-OHP injection (Fig. 4a).
The number of WBCs in saline-treated mice exhibited significant 24-h oscillation, with a peak during the light phase (Fig. 4b). A significant reduction of the number of peripheral WBCs was observed in mice injected with L-OHP at both time points, but the L-OHP-induced reduction of the peripheral WBC count was attenuated by injection of the drug at ZT18 (Fig. 4b). Indeed, the value of the area under the curve (AUC) of the time course of the peripheral WBC count (AUC 72-92 of WBCs) after L-OHP injection at ZT6 was markedly lower than that after the drug injection at ZT18 (Fig. 4c). This demonstrated that L-OHP-induced leukopenia also varies according to its administration time. The dosing time-dependent difference in the myelotoxic effects of L-OHP may affect the severity of leukopenia.

Discussion
Diurnal oscillations in the activity of drug metabolism enzymes and xenobiotic transporters are often associated with the dosing time-dependent changes in drug disposition 1,2,38 . Such dosing time-dependency of drug pharmacokinetics also affects their efficacy and safety. In this study, we demonstrated that the expression of MRP4 in In addition to MRP4, the mRNA expression of P-gp (encoded by Abcb1a or Abcb1b), BCRP (Abcg2), and MRP2 (Abcc2) was detected in BMCs of mice. However, no significant accumulation of Pt was observed in BMCs after treatment with L-OHP in the presence of inhibitors. Treatment of BMCs by L-OHP together with ceefourin resulted in a 5-times increase in Pt accumulation. Ceefourin was originally reported as a human MRP4 inhibitor 29 , but it also induced significant Pt accumulation in mouse MRP4-expressing cells after L-OHP treatment. Ceefourin may exert its inhibitory action on both human and mouse MRP4. Collectively, these findings suggest that MRP4 functions as a major efflux transporter of L-OHP in mouse BMCs.
In contrast to efflux transporters, human organic cation transporter 2 and 3 (hOCT2 and hOCT3), human novel organic cation transporter 1 and 2 (hOCTN1 and hOCTN2), and human apical multidrug and toxin extru-sion1 and 2-K (hMATE1 and hMATE2-K) have been suggested to incorporate L-OHP into cells [39][40][41] . Although the mRNA expression of OCTN1 (encoded by Slc22a4), but not of OCT2 (Slc22a2), OCT3 (Slc22a3), OCTN2 (Slc22a5), or MATE1 (Slc47a1), was detected in BMCs of mice, the protein levels of OCTN1 failed to exhibit significant diurnal oscillation ( Supplementary Fig. 2). In the mouse small intestine, OCTN1 is expressed in a diurnal time-dependent manner 5 . The oscillation of intestinal expression of OCTN1 is dependent on food-entrainable bile acid stimulation; however, it is unlikely that bile acid reaches bone marrow and affects the expression of OCTN1. This may explain the reason for the lack of diurnal expression of OCTN1 in BMCs. Although other xenobiotic transporters may also participate in the influx and efflux of L-OHP in BMCs, diurnal expression of MRP4 is, at least in part, associated with the dosing time-dependent difference in Pt accumulation in BMCs after treatment with L-OHP. This is also supported by the present findings that the time-dependent change in Pt accumulation in L-OHP-treated BMCs disappeared after inhibition of MRP4 activity.
Diurnal expression of some types of ABC transporters are regulated by the molecular circadian clock 10,11,13 . Gene products of Clock and Bmal1 form a heterodimer that activates the transcription of Period (Per) and Cryptochrome (Cry) genes through E-box (CAC GTG ) or E-box-like (CAC GTT ) enhancer sequences 42 . Once PER and CRY proteins have reached a critical concentration, they attenuate CLOCK/BMAL1-mediated transactivation 43,44 , thereby generating diurnal oscillation in their own transcription. CLOCK and BMAL1 also activate their target genes, but this transactivation is periodically repressed by PER and CRY, resulting in diurnal oscillation in the downstream events 45,46 . Proline-and acid-rich (PAR) basic leucine zipper (bZIP) proteins, hepatic leukemia factor (HLF), thyrotroph embryonic factor (TEF), and D-site binding protein (DBP) are examples of such output mediators 47,48 . Previous studies have demonstrated that renal expression of Abcc4 mRNA is decreased in PAR bZIP protein-deficient mice 49 . As the circadian clock genes and clock-controlled output mediators are also expressed in BMCs of mice 50 , these molecular components may generate diurnal expression of Abcc4 mRNA and MRP4 protein in BMCs.
The dosing time-dependency of the pharmacological effects of drugs is attributable not only to the diurnal changes in their pharmacokinetics, but also to the sensitivity of cells to drugs 1,35 . Intracellularly incorporated L-OHP forms Pt-DNA adducts and activates apoptotic signals, resulting in cell death. One possible signal of L-OHP-induced apoptosis is the blockage of RNA polymerases by Pt-DNA adducts, causing transcription cessation and cell death through p53-dependent pathways 51 . The expression of p53 in mouse BMCs exhibits diurnal oscillation, with a peak during the light phase 52 . As severe myelotoxic effects of L-OHP were observed when mice were injected with the drug during the light phase, the temporal p53 accumulation in BMCs may also be associated with dosing time-dependent changes in L-OHP-induced myelotoxicity. www.nature.com/scientificreports/ WBCs are produced and derived from multipotent cells in the bone marrow known as hematopoietic stem cells. Diurnal release and accumulation of hematopoietic stem cells in the circulating blood is important for the regeneration of the stem cell niche in bone marrow 53 , resulting in diurnal changes in the number of circulating WBCs. Diurnal release of hematopoietic stem cells from bone marrow is regulated by CXCL12 whose microenvironmental expression is controlled by core components of the molecular clock through circadian noradrenaline secretion by the sympathetic nervous system. The cellular sensitivity to chemotherapeutic drugs differs between undifferentiated and differentiated hematopoietic stem cells 54 . The cytotoxic effects of L-OHP were only detected in undifferentiated BMCs, even though diurnal expression of MRP4 was noted in both undifferentiated and differentiated BMCs. Marked cytotoxicity of L-OHP in undifferentiated BMCs was observed when mice were injected with the drug during the light phase. This may also explain the dosing time-dependent difference in L-OHP-induced leukopenia.
Significant dosing time-dependent differences in myelosuppression were detected in experimental animals when they were administered irinotecan or doxorubicin 17,18 . Although the active metabolites of irinotecan SN-38 and doxorubicin have been reported as substrates of MRP4 55 , there was no significant time-dependent accumulation of these drugs in mouse BMCs (Supplementary Fig. 3). In addition, accumulation of SN-38 and doxorubicin in mouse BMCs were not affected by the MRP4 inhibitor ceefourin. The extracellular extrusion of SN-38 and doxorubicin may be caused by other xenobiotic transporters whose expression does not exhibit diurnal oscillation in mouse BMCs. On the other hand, dosing time-dependent differences in myelosuppression by SN-38 and doxorubicin may be mainly associated with diurnal changes in the sensitivity of BMCs to drugs rather than their pharmacokinetics. The activity of topoisomerase, a target enzyme of SN-38 and doxorubicin, has been reported to exhibit diurnal oscillation in mouse BMCs 18 .
In humans, leukopenia induced by L-OHP also varies with the administration time 28 . However, the underlying mechanism is not fully understood. Our present study suggests a possible mechanism of the dosing timedependent difference in L-OHP-induced myelotoxicity and leukopenia. Although many chemotherapeutic drugs often cause severe leukopenia, they are still administered without considering the time of day. Identification of the mechanism underlying the dosing time-dependent toxicity of chemotherapeutic drugs is expected to improve pharmacotherapy for the treatment of cancer.

Animals and treatments. All animal experiments were conducted in accordance with the Guidelines for
Animal Experiments of Kyushu University and were approved by the Institutional Animal Care and Use Committee of Kyushu University (approved protocol ID #A30-060-0). Five-week-old male ICR mice (Charles River) were housed 6 per cage in a temperature-controlled (24 ± 1 °C) room under a 12-h light/dark cycle (ZT0, lights on; ZT12 lights off) with food and water ad libitum. During the dark period, a dim red light was used to aid animal treatment.
Preparation of mouse MRP4-expressing cells. To  Quantitative Rt-pcR analysis. The mRNA levels of ABC transporter were quantified as described in our previous report 56 . Total RNA was extracted using RNAiso PLUS reagent (Takara, Osaka, Japan). cDNA was synthesized using 1 µg of RNA and the ReverTra Ace qPCR RT kit (Toyobo, Osaka, Japan). The cDNA was amplified by a real-time PCR system (Applied Biosystems, Life Technologies, Carlsbad, CA) using specific primer sets. Sequences of primers are listed in Supplementary Table 1.
Measurement of pt content. Cells were incubated in transport buffer (130 mM NaCl, 6.0 mM KCl, 12 mM D-glucose, 2.0 mM CaCl 2 , 1.2 mM MgCl 2 , 0.1 mM CH 3 COONa, and 6.0 mM HEPES) for 15 min. After pre-incubation, cells were incubated in transport buffer containing 50 μM L-OHP at 37 °C in the presence or absence of ABC transporter inhibitors, PSC833, Ko143, bromosulfalein, or ceefourin, at 37 °C for 1 h. After washing with ice-cold PBS for two times, the amount of Pt in cells was measured by inductively coupled plasma mass spectrometry (ICP-MS; Agilent7500c; GenTech Science Inc., Frankfurt, Germany). The intracellular Pt content was normalized by protein concentrations.

Measurement of 6-MP content.
After pre-incubation of cells with transport buffer for 15 min, cells were incubated in transport buffer containing 1 mM 6-MP at 37 °C in the presence or absence of ceefourin at 37 °C for 1 h. After washing with ice-cold PBS for two times, cells were homogenized with 1 mL of acetonitrile containing 50 µg of 6-thioguanine (internal standard) and 10 nmol of dithiothreitol. The mixture was centrifuged at 12,000 rpm for 10 min, and the organic phase was collected and added to 1.5 mL of dichloromethane. The mixture was centrifuged for 10 min. The supernatant was subjected to HPLC after evaporation to dryness. The mobile phase of phosphate buffer (pH 2.5; 10 mM)-acetonitrile (96:4, v/v) containing 0.015% dithiothreitol was eluted at 1.0 mL/min through a C18-MS-II column (4.6 × 150 mm; Nacalai Tesque, Kyoto, Japan) using an LC-20AD pump (Shimadzu Corporation, Kyoto, Japan). The separated analyte was detected at a wavelength of 323 nm using SPD-20A (284 nm; Shimadzu). www.nature.com/scientificreports/ Western blot analysis. The levels of MRP4 protein were quantified as described in our previous report 57 .

Measurement of the number of white blood cells in peripheral blood.
To investigate the dosing time-dependent differences in L-OHP-induced myelotoxicity, mice were injected with a single dose of 20 mg/kg (i.v.) of L-OHP at ZT6 or ZT18. Peripheral blood was collected from the mouse tail vein at ZT6 and ZT18, and 20-µL blood samples were incubated with 1 mL of hemolytic buffer (0.75% NH 4 Cl and 17 mM Tris, pH = 7.65, 37 °C) for 15 min. After centrifugation, white blood cells (WBCs) were collected and fixed with 4% paraformaldehyde for 10 min. After nuclear staining with 4′,6-diamidino-2-phenylindole (Dojindo Laboratories, Kumamoto, Japan), the numbers of WBCs were measured using an automatic cell counting instrument (Thermo Fisher Scientific, San Jose, CA).
Pharmacokinetic analysis of L-OHP. Mice were injected with a single dose of L-OHP (17.5 mg/kg, i.v.) at ZT6 and ZT18. Serum and BMCs were collected at 1, 4, and 24 h after L-OHP administration. Pt concentrations in serum and BMCs were measured by ICP-MS as described above.
Measurement of BMC viability. BMCs were collected from mouse femora at 72 h after the administration of L-OHP. After hemolysis, as described above, cell suspensions were prepared in 1 mL of PBS containing 1% bovine serum albumin (BSA). The number of cells was counted using an automatic cell counting instrument. Thereafter, the number of cells was adjusted to 1.0 × 10 6 /mL and they were treated with 50 μM calcein-AM (Live/ Dead Cell Staining Kit II; PromoKine, Heidelberg, Germany) for 45 min at room temperature under light shielding. Calcein-stained cells were counted using a flow cytometer (FACS; BD Biosciences). Statistical analysis. The significance of differences among groups was analyzed by ANOVA, followed by the Tukey-Kramer post-hoc test. The unpaired t-test was used for the comparison of data between two groups.

Measurement of the number of differentiated and undifferentiated
Equal variances were not formally tested. A 5% level of probability was considered to be significant.

Data availability
All data supporting the results of the present study are included in the article, either in the main figures or supplementary information files.