Uremic serum residue decreases SN-38 sensitivity through suppression of organic anion transporter polypeptide 2B1 in LS-180 colon cancer cells

Pharmacokinetics of SN-38 in patients with end-stage kidney disease (ESKD) is partially varied because of fluctuations in transporters expression and/or function by high protein bound-uremic toxins concentration. The fluctuations may induce variations in anticancer drugs sensitivity to cancer cells. We aimed to clarify the variations in sensitivity of SN-38 to cancer patients with ESKD and investigate this mechanism, by human colon cancer cells exposed to uremic serum residue. LS180 cells were exposed to normal or uremic serum residue (LS/NSR or LS/USR cells) for a month. IC50 values of SN-38 in LS/NSR or LS/USR cells were calculated from viability of each cells treated SN-38. mRNA expression and intracellular SN-38 accumulation was evaluated by RT-PCR and HPLC-fluorescence methods, respectively. The IC50 value in LS/USR cells was higher than that in LS/NSR cells. Organic anion transporter polypeptide (OATP) 2B1 mRNA expression was lower in LS/USR cells than in LS/NSR cells, and SN-38 accumulation in LS/USR cells was lower than that in LS/NSR cells. Only co-treatment baicalin, which is OATP2B1 inhibitor, almost negated the difference in SN-38 accumulation between LS/NSR and LS/USR. Anticancer effects of substrates of OATP2B1, such as SN-38, were reduced in ESKD patients at the same plasma substrate concentration.

acid and indole-3-acetic acid inhibited the uptake of estron-3-sulfate through OATP1B1 and OATP1B3 and the uptake of losartan into rat hepatic cells. They also indicated that the serum of patients with ESKD inhibited the uptake of losartan into human hepatic cells 5 . In addition, we reported that the simultaneous treatment of serum and digoxin in patients with ESKD inhibits the uptake of digoxin into isolated human hepatocytes 6 , and that pretreatment of human hepatocellular carcinoma Hep3B cells with serum from patients with ESKD decreases the initial uptake of pravastatin and the expression of OATP2B1 mRNA 7 . These reports indicate that the variation in hepatic clearance was partially induced by fluctuations in transporter expression and/or function in patients with ESKD.
The functional variation in transporters also induces the resistance of cancer cells to anticancer drugs. For example, it has been reported that high expression of efflux transporters, such as multidrug resistance protein (MDR) 1 and breast cancer resistance protein (BCRP), induces resistance to several chemotherapy 8 . In addition, Yang L et al. 9 showed that indoxyl sulfate enhanced the expression of BCRP, one of the determinant factors of the anticancer effects of SN-38, in Caco-2 cells. In contrast, it was reported that the high expression of uptake transporter OATP1B3 improved the 5-year survival rate for patients with colorectal cancer 10 . Therefore, there is a concern that the functional variation of transporters in patients with ESKD affects not only pharmacokinetics, but also sensitivity to anticancer drugs. Some reports about OATP2B1 and SN-38 have already been reported. Fujita et al. have shown that SN-38 is a substrate of OATP2B1 11 . We have previously demonstrated that uremic serum residue down-regulates OATP2B1 expression on Hep3B 7 and uremic serum residue decline functions of OATP1B1 when to carry SN-38 into HEK293 expressing OATP1B1 3 . However, these reports do not demonstrate directly that uremic serum affect SN-38 sensitivity in colon cancer. Our hypothesis is that these fluctuations of transporter function also occur in cancer cells, and that the anticancer effects of SN-38 can vary in patients with ESKD. Nevertheless, the evaluation of uremic serum, including the effect of uremic toxins on SN-38 sensitivity, has not been conducted.
The aim of this study was to clarify the effects of variations in SN-38 sensitivity in patients with ESKD and cancer; therefore, we evaluated the variations in SN-38 sensitivity and the mechanisms in human colon cancer cells exposed to uremic serum residue.

Difference in SN-38 uptake between LS/NSR and LS/USR cells.
The uptake of SN-38 was increased over time up to 5 min (Fig. 2a) and was dependent on SN-38 concentration up to 10 µM in both cells (Fig. 2b). In addition, these uptakes were saturated at high SN-38 concentration (Fig. 2b), and the uptake of SN-38 into LS/ USR cells was significantly lower than that in LS/NSR cells under any conditions (Fig. 2).

Difference in expression of drug transporters between LS/NSR and LS/USR cells.
There was no difference in OATP1B1 and OATP1B3 mRNA between LS/NSR and LS/USR cells (Fig. 3a,b). In contrast, BCRP and MDR1 mRNAs in LS/USR cells were expressed more highly than in LS/NSR cells (Fig. 3d,e), and OATP2B1 mRNA in LS/USR cells was less strongly expressed in LS/NSR cells (Fig. 3c).

Discussion
In this study, it has been shown that the long-term exposure to USR significantly reduced the sensitivity to SN-38 in LS180 cells (Fig. 1), and the mechanism may be the decreased uptake on SN-38 owing to downregulation of the uptake transporter (Figs 2 and 5). These results suggested that the sensitivity of cancer cells to SN-38 may be decreased in patients with ESKD. Therefore, it is important to be concerned about the decline of renal and  . Difference in mRNA expression of drug transporters between LS180 cells were exposed to normal serum residue (LS/NSR) and uremic serum residue (LS/USR) and LS180 cells were exposed to uremic serum residue (LS/USR cells). LS/NSR (□) and LS/USR cells (■) were seeded at 3.0 × 10 5 cells/well into 6-well plates. The expression of organic anion transporting polypeptide (OATP)1B1 (a), OATP1B3 (b), OATP2B1 (c), breast cancer resistance protein (BCRP) (d), and multidrug resistance protein (MDR) 1 mRNA (e) was quantified by real-time RT-PCR. Each column represents the mean ± S.E. (n = 3-4). The significance of any differences between the mean values were determined by unpaired Student's t-test (*p < 0.05, **p < 0.01).
www.nature.com/scientificreports www.nature.com/scientificreports/ nonrenal clearance, but also the decrease in sensitivity to anticancer drugs; therefore, there is a need for careful monitoring of the antitumor effects in patients with ESKD.
The expression of not only OATP1B1 and 1B3, but OATP2B1 mRNA in LS/USR cells was lower than in LS/ NSR cells (Fig. 3). In addition, not rifampicin, an OATP1B1 and OATP1B3 inhibitor, but baicalin, an OATP2B1 inhibitor, almost negated the difference in SN-38 uptake between LS/NSR and LS/USR cells (Fig. 5). As SN-38 is a known substrate of OATP2B1 11 , the decline in sensitivity to SN-38 in LS/USR cells is caused by the downregulation of OATP2B1 expression. These results suggested that the sensitivity to anticancer drugs, such as substrates of OATP2B1 like SN-38, may be decreased in patients with ESKD. This is the first study to show the possibility of decreased sensitivity to anticancer drugs in patients with ESKD.
In this study, it was shown that USR decreased OATP2B1 mRNA expression (Fig. 3c). It has been reported that OATP2B1 expression is partially regulated by thyroid hormone receptor (TR), one of the transcription factors. Henriette et al. 12 reported that triiodothyronine and thyroxine induced the expression of OATP2B1 mRNA. The promoter region of OATP2B1 has direct repeats spaced by four nucleotides (DR-4) and a TR bound sequence. In addition, Santos et al. 13 reported that serum before dialysis significantly decreased luciferase activity on the cDNA-inserted DR-4 domain, compared with serum after dialysis, in patients after ESKD. Therefore, our results suggested that serum compounds, which can be eliminated by dialysis, downregulated the expression of OATP2B1 mRNA.
Although it was shown that MDR1 mRNA expression was higher in LS/USR cells than in LS/NSR cells, although the MDR1 inhibitor verapamil did not affect the difference in SN-38 uptake of both cells (Fig. 4). Jansen et al. 14 have reported that the overexpression of MDR1 did not affect SN-38 sensitivity. Therefore, it cannot be denied that MDR1 is induced in patients with ESKD, although the induction of MDR1 mRNA expression did not affect the decreased sensitivity to SN-38 in patients with ESKD.
Ko143, a BCRP inhibitor, did not affect the difference in SN-38 uptake in LS/NSR cells and LS/USR cells, even though BCRP mRNA in LS/USR cells was significantly higher than that in LS/NSR cells. These results indicated that the induction of BCRP did not contribute to the decrease in SN-38 sensitivity caused by USR. In general, the  www.nature.com/scientificreports www.nature.com/scientificreports/ overexpression of BCRP decreased SN-38 sensitivity 9 . Mutsaers et al. 15 reported that indoxyl sulfate and hippuric acid noncompetitively inhibited BCRP-mediated estron-3-sulfate uptake into inside-out vesicles from cells overexpressing BCRP, and K i and IC 50 of indoxyl sulfate are 500 µM and 770 ± 120 µM and K i and IC 50 of hippuric acid are 3670 ± 170 µM and 4000 ± 100 µM, respectively. We have previously shown uremic toxins concentration, indoxyl sulfate (total: 163.8 ± 6.2 µM, unbound: 26.3 ± 2.4 µM) and hippuric acid (total: 208.7 ± 2.6 µM, unbound 106.1 ± 1.5 µM) (Values represent the mean ± S.E.) 16 . These K i and IC 50 are higher than concentration of uremic toxins in our uremic serum, in other words, BCRP seems not to be inhibited by sole uremic toxin, but uremic toxins and other components in uremic serum may co-inhibit that. Therefore, future studies should be conducted to consider the interaction between SN-38, uremic toxins and other components, and the induction and inhibition of BCRP.
This study has showed that USR decreased the expression of OATP2B1 mRNA (Fig. 3cs) and the sensitivity to SN-38, the substrate of OATP2B1 (Fig. 1). Because some anticancer drugs, such as erlotinib, are a substrate of OATP2B1 17 , the sensitivity to these anticancer drugs may be decreased in patients with ESKD. However, OATP2B1 is reported to be expressed in both cancer cells and hepatocytes 18 ; in addition, we have reported that USR decreased the expression of OATP2B1 mRNA in Hep3B cells 7 . The plasma concentration of these anticancer drugs may increase as the hepatic uptake is decreased by the downregulation of OATP2B1 expression in hepatocytes, and the increase most likely offsets the decreased susceptibility of patients with ESKD.
This study has some limitations. First, as we did not conduct in vivo research, there is a need for further research using animal models with ESKD and cancer. The in vivo research should provide more information such as the difference of sensitivity in relation to the calculated renal clearance or to concentration of uremic toxin, hormone or waste in blood. Second, this study did not research other factors, such as UDP-glucuronosyltransferase (UGT) 1A1. UGT1A1 is expressed in colorectal cancer, and p-cresol and indoxyl sulfate inhibit the function of UGT1A1 19,20 ; therefore, further research is needed to evaluate the effects of USR on the function of UGT1A1. Third, we demonstrated different function of OATP2B1 in LS/NSR and LS/USR in uptake study, however we did not confirm protein level in each cells. It is necessary to discover the mechanism in molecular biology level in future study. Lastly, we used only LS180 to demonstrate SN-38 sensitivity in colorectal cancer patients with ESKD. Because of restriction to use uremic serum in our study due to ethical concern and the amount, we could do our research by only one cell line, LS180. Even though LS180 is a good model for our study since some clinical samples are reportedly recognized OATP2B1 expression on themselves 21 , it is needed to demonstrate by other colorectal cancer cell line or tissue to confirm reproducibility in future study.

Conclusion
In conclusion, long-term exposure to USR decreased the sensitivity of LS180 cells to SN-38 through the downregulation of OATP2B1 mRNA and a reduction in the uptake of SN-38. These results suggested that the anticancer effects of substrates of OATP2B1, such as SN-38, decreased in patients with ESKD at the same plasma concentration. Therefore, not only the increase in SN-38 concentration, but also the decrease in the antitumor effects of SN-38 in patients with ESKD, should be considered.
Ko143 was purchased from Sigma Aldrich Co. LLC (St. Louis, MO, USA). Verapamil and rifampicin were purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan). Baicalin was purchased from AdooQ Bioscience (Irvine, CA, USA), and Cell Quanti-Blue TM Cell Viability Assay Kit was purchased from Bio Assay Systems (Hayward, CA, USA). All other reagents were of high-purity analytical or high-performance liquid chromatography (HPLC)-grade. Pooled human serum from healthy subjects (normal serum) was purchased from Merck Millipore Ltd. (Billerica, MA, USA), and pooled human serum from patients with ESKD (uremic serum) was obtained from more than 400 patients with ESKD undergoing hemodialysis at Shirasagi Hospital (Osaka, Japan). Blood samples were collected immediately before hemodialysis in a non-invasive manner. Normal and uremic serum were deproteinized by 3-times volume of methanol as much as serum and evaporated under a nitrogen stream at 50 °C; the resulting serum residues, normal serum residue (NSR) and uremic serum residue (USR), were used in the study 3 . The present study was approved by the ethics committees of Shirasagi Hospital (IRB number J2012013) and Kyoto Pharmaceutical University (IRB number 08-04).

Cell culture. A stock of the intestinal human colon adenocarcinoma LS180 cell line was purchased from
American Type Culture Collection (Manassas, VA, USA). LS180 cells were cultured in Dulbecco's modified Eagle's medium (DMEM, Invitrogen Co. Grand Island, NY, USA) supplemented with 10% fetal bovine serum (Lot. AWD12933 and AXM56561, Thermo Fischer Scientific, Waltham, MC, USA), 100 U/mL penicillin and 100 μg/mL streptomycin (Invitrogen Co. and Nacalai Tesque, Inc., Kyoto, Japan), and 0.1 mM non-essential amino acids (Invitrogen Co. and Nacalai Tesque, Inc.). NSR and USR were dissolved in DMEM as much as 10 times volume of serum and passed through a 0.22 µm membrane filter to sterilize. LS180 cells were cultured for 1 month in DMEM with NSR (NSR-DMEM) or USR (USR-DMEM); these cells were as named LS/NSR cells and LS/USR cells, respectively. For evaluation of sensitivity to SN-38 in LS180 cells, LS/NSR cells and LS/USR cells were seeded into 96-well plates (1 × 10 3 cells/well) for a day. For uptake and efflux studies, both of cells were seeded into 24-well plates (1.2-1.5 × 10 5 cells/well) for 3-4 days. For real-time RT-PCR, both of cell were seeded into 6-well plates (3 × 10 5 cells/well) for 3-4 days. All of culturing conditions were in a humidified atmosphere containing 5% CO 2 at 37 °C.
Evaluation of sensitivity to SN-38 in LS180 cells. After removal of DMEM, each well was exposed to SN-38 (0.1-25 µM) dissolved in NSR-DMEM or USR-DMEM for 6 days. After 6 days, the cell viability was Real-time RT-PCR. Total RNA was extracted from the cells by RNAzol ® RT Reagent (Molecular Research Center, Inc., Cincinnati, OH, USA). The absorbance was measured by using DU ® 730 (Beckman Coulter, Tokyo, Japan), and the concentration of RNA was calculated from the equation Total RNA (ng/mL) = A 260 × 40. cDNA was synthesized from 1 µg of total RNA by a reverse transcriptase (RT) reaction using ReverTra Ace ® qPCR (Toyobo Co., Ltd., Osaka, Japan) and i-Cycler iQ (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Total RNA was heated at 37 °C for 15 min (RT reaction), and then 98 °C for 5 min (inactivation of RT), and then cooled at 4 °C. Polymerase chain reaction (PCR) was performed by a LightCycler ® Nano System (Roche Diagnostics K.K., Tokyo, Japan). From the prepared cDNA, a 2.0 µL sample was mixed with sterilized distilled water (7.16 µL), THUNDERBIRD TM SYBR ® qPCR Mix (Toyobo) (10 µL), 10 µM sense primer (0.4 µL), 10 µM anti-sense primer (0.4 µL), and 50 × ROX reference dye (Toyobo) (0.04 µL). The initial denaturation period was 95 °C for 1 min, followed by 45 cycles of amplification (95 °C for 10 s, and 60 °C for 30 s). The sequences of the primers are shown in Table 1.
The fluorescence of the SYBR green dye was determined as a function of the number of PCR cycles, and the threshold cycle (C T ), the cycle number at which the amplification reached a significant threshold, was determined. The C T values were used to quantify the PCR products; in other words, the relative expression of the target gene was expressed as 2 −ΔCT , and ΔC T was calculated by ΔC T = C T tar − C T β2M , where C T tar and C T β2M are the C T of the target gene and C T of β2M, respectively. Statistical analysis. Measured and IC 50 values were expressed as the mean ± standard error (S.E.) and median (95% confidence intervals), respectively. Non-overlapping confidence intervals of the IC 50 were considered statistically significant; and significant differences between two groups were determined by using an unpaired Student's t-test; p values of less than 0.05 was considered statistically significant.  Table 1. Sequences of primers used to amplify different genes by real-time PCR. * We synthesized these by using Primer Express ® software.