Expression and Functional Evaluation of ABC and SLC Transporters in Human Choroid Plexus Papilloma (HIBCPP) Cells: A Human Blood-Cerebrospinal Fluid Barrier Model.

Introduction
The choroid plexus (CP) produces cerebrospinal fluid (CSF), which fills the ventricles and subarachnoid space and protects the brain through buoyancy. Additionally, the CSF functions as a source of ions and water that contribute to the maintenance of homeostasis within the brain. The epithelial cells that form the CP are polarized with apical (AP) and basolateral (BL) membranes that face the CSF and plasma, respectively. These cells prevent free exchange of substances between the blood and CSF by tightly sealing cell junctions, thus forming the blood-cerebrospinal fluid barrier (BCSFB). The BCSFB controls the supply of nutrients to the brain and the removal of waste products produced in the brain [1]. This function, together with the blood–brain barrier (BBB) formed by brain capillary endothelial cells, plays a crucial role in protecting the brain.
The transport of drugs and nutrients across the BCSFB and BBB is characterized by vectorial transport via transporters [2]. The functional characteristics of various solute carrier (SLC) and ATP-binding cassette (ABC) transporters in BCSFB have been clarified in studies using experimental animals, such as rodents [3–5]. In contrast, transporter proteins expressed in BCSFB show differences between humans and rats [6]. The functions and roles of the transporters in human BCSFB are not fully understood. This is due to the difficulty of conducting clinical research in humans and the unavailability of alternative human BCSFB model cells. Primary cultured cells, which are the only human models available, are difficult to evaluate accurately for transport functions owing to changes in cell morphology and fragile tight junctions [7]. Therefore, an appropriate model human BCSFB cell line that reproduces the regulation of substance transport is urgently required.

Human choroid plexus papilloma (HIBCPP) cells, derived from a CP papilloma in the lateral ventricle of a Japanese woman, were established as a functional model cell of the human BCSFB [8, 9]. This model has been used to analyze bacterial and viral meningitis [9–12]. In addition, HIBCPP cells show anatomical and functional characteristics of BCSFB, including microvillus formation and sealed junctions between cells, and are considered useful model cells for human BCSFB [9, 12–14]. Analysis of HIBCPP cells should provide important information concerning the physiological functions of the human BCSFB, the mechanisms underlying the onset of brain diseases, and strategies for developing related therapeutic drugs.
Therefore, this study aimed to analyze the expression of transporters in HIBCPP cells using quantitative proteomics and quantitative reverse transcription-PCR (qRT-PCR) and to evaluate the functional characterization of transporter-mediated transport in HIBCPP cells.

Materials and Methods

Reagents
The reagents used in this study were purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan), Sigma-Aldrich (St. Louis, MO, USA), and Tokyo Chemical Industry (Tokyo, Japan), unless otherwise stated. JPH203 was purchased from Selleck Chemicals (Houston, TX, USA). Radioisotopes sucrose [14C(U)] ([14C]sucrose, 600 mCi/mmol), arginine L-[2,3,4-3H] ([3H]L-arginine, 40 Ci/mmol), glutamic acid L-[2,3,4-3H] ([3H]L-glutamate, 60 Ci/mmol), methyl-D-glucose 3-O-[methyl-3H] ([3H]3-O-methylglucose, 90 Ci/mmol), and lactic acid L-[14C(U)] Na salt ([14C]L-lactate, 150 mCi/mmol) were purchased from American Radiolabeled Chemicals (St. Louis, MO). Digoxin [3H(G)] ([3H]digoxin, 23.8 Ci/mmol) was purchased from PerkinElmer (Waltham, MA, USA). The details of the products and buffer components used in this study are listed in the Supplementary Information.

Cell Culture
HIBCPP cells were established from a human CP papilloma that was surgically removed as described by Ishiwata et al. [8]. All experiments were performed using HIBCPP cells between passages 35 and 40. The cells were cultured in Dulbecco's Modified Eagle Medium/Ham’s F-12 with L-glutamine, phenol red, HEPES, and sodium pyruvate (FUJIFILM Wako) supplemented with penicillin–streptomycin (10,000 U/mL; Thermo Fisher Scientific, Waltham, MA, USA) and 10% fetal bovine serum (FBS; Thermo Fisher Scientific) in a 10-cm dish (Corning, NY, USA). The culture medium was replaced daily with fresh medium. For passage of HIBCPP cells, the cells were washed with PBS(-) and then detached by treatment with 0.05% trypsin–EDTA for five minutes at 37°C. Then, single cells were obtained by centrifuging the detached cells (200 g, 5 min) using a cell strainer filter (pore size 10 μm). After cell counting, single cells were seeded at a density of 10 × 105 cells/dish in a new 10-cm dish.
Human primary choroid plexus epithelial cells (HCPEpiC) and culture-related reagents were purchased from ScienCell Research Laboratories (Carlsbad, CA, USA). HCPEpiC on 10-cm dishes (Corning) coated with poly-L-lysine were cultured in epithelial cell medium (EpiCM, ScienCell) containing the supplied 1% EpiCGS, 100 units/mL, 100 μg/mL penicillin–streptomycin, and 2% FBS following the product manual. In the passage of HCPEpiC, the cells were washed with PBS(-) and then detached by treatment with 0.05% trypsin–EDTA at 37°C. Following the addition of trypsin neutralization solution (ScienCell), the cell suspension was centrifuged (200 g, 5 min) and resuspended in culture medium to obtain single cells. After cell counting, single cells were seeded at a density of 10 × 105 cells/dish in a new 10-cm dish coated with poly-L-lysine.
Cells were cultured at 37 °C in an atmosphere of 5% CO2 and 95% air.

Transporter Protein and mRNA Expression in HIBCPP Cells and HCPEpiC
Protein expression was determined using quantitative proteomics in whole-cell lysates of HIBCPP and HCPEpiC. Whole-cell lysates were prepared by dissolving cells in extraction buffer followed by sonication to disrupt the cells using a Sonicator (Ohtake Works, Tokyo, Japan). Cell lysate pretreatment was conducted using the SP3 method as previously described [15, 16]. Briefly, the samples were reduced using 25 mM Tris (2-carboxyethyl)phosphine hydrochloride and alkylated using 36 mM chloroacetamide for 30 min at room temperature. The alkylated proteins were cleaned using a Sera-Mag SpeedBead in 80% ethanol (Cytiva, Marlborough, MA, USA). The protein samples were digested using lysyl endopeptidase (FUJIFILM Wako) and trypsin (Promega, Madison, WI, USA) for 3 and 16 h, respectively. The digested samples were desalted using GL-Tip SDB (GL Sciences, Tokyo, Japan) and dried. Dried samples were reconstituted with 0.1% trifluoroacetic acid and analyzed using liquid chromatography-tandem mass spectrometry. The samples were analyzed using the data-independent acquisition method on a ZenoTOF 7600 (Sciex, Framingham, MA, USA) interfaced with an Ultimate3000 RSLCnano (Thermo Fisher Scientific). For protein identification and quantification, the data were analyzed using DIA-NN 1.8.1 and the human reference proteome [17].
Total RNA was extracted from HIBCPP cells and HCPEpiC using NucleoSpin RNA Plus (Macherey–Nagel, Düren, Germany), according to the manufacturer’s protocol. Total RNA was treated with a Ribonuclease Inhibitor (TaKaRa Biomedicals, Shiga, Japan) and SuperScript III Reverse Transcriptase (Thermo Fisher Scientific) for reverse transcription. PCR was performed using a mixture of 10 ng cDNA, 5 pmol sense/antisense primers, and SYBR Select Master Mix (Thermo Fisher Scientific) on a 7500 Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). Detailed methods for reverse transcription and PCR are described in the Supplementary Information.
The relative protein and mRNA expression of each transporter in HIBCPP cells and HCPEpiC was corrected using the expression of TATA-binding protein (TBP; a housekeeping gene). Expression analysis in HIBCPP cells and HCPEpiC was performed using cells cultured for 5 days in a 10-cm dish.

Cellular Uptake Assessment
HIBCPP cells were seeded at a density of 1.0 × 105 cells/well on a 24-well plate (Corning). A cellular uptake assessment was conducted using confluent HIBCPP cells 5 days after seeding. After removing the culture medium, the HIBCPP cells were washed with incubation buffer (IB) and equilibrated in IB for 20 min. Unless otherwise stated, the pH of IB was adjusted to 7.4. After equilibration, IB was replaced with the test drug or compound, and uptake by the cells was allowed for the designated time. Temperature-dependent uptake was evaluated by monitoring substrate uptake at 4°C. In the inhibition analysis, uptake was measured in the presence of inhibitors at specified concentrations. After uptake, cells were washed three times with ice-cold IB and collected in 200 μL of water via scraping. Uptake was evaluated by comparing the cell-to-medium ratios (μL/mg protein). This ratio was calculated using the amount of substrate uptake in the cells, the concentration of the test solution, and the cell amount per well. The number of cells was calculated using a Micro BCA™ Protein Assay Kit (Thermo Fisher Scientific). The measurement of the sample is described in the Supplementary Information.

Transcellular Transport Assessment
HIBCPP cells were seeded at a density of 3.0 × 105 cells/insert on PET membrane inserts (0.4-μm pore size; Corning) placed upside down. The insert was placed upside down, an optional ring was attached, and the cells were attached using 500 μL of culture medium. The next day, after removing the medium in the ring, the insert was placed in 1 mL of medium in a 24-well plate, and 500 μL of medium was added to the insert, and the inverted culture of HIBCPP cells was initiated. Inverted culture of HIBCPP cells has been reported to show several advantages, including increased transepithelial electrical resistance (TEER) values [9, 18]. The inverted culture was incubated for approximately 5 days until the cells reached confluence. After equilibration for 20 min using the transport (TR) buffer, HIBCPP cells were used for transcellular transport evaluation. TR buffer was added to the upper part (BL side) and the lower part (AP side) in volumes of 200 and 900 μL, respectively. The test drug or compound was added to one part and allowed to permeate the other part, and transport in the AP-to-BL and BL-to-AP directions was evaluated. The portion of the test drug or compound that permeated the cell layer was collected within a designated time period. The measurement of the sample is described in the Supplementary Information.
The apparent permeability surface product (Papp) was calculated using the following equation, and the directional transport of each transporter was evaluated:where PS, dQ/dt, and D0 are the permeability surface area product, the transport velocity of the test drug or compound, and the concentration of the test solution, respectively.where A denotes the surface area of the PET membrane (0.33 cm2). Each Papp value was corrected (Papp,correct) using the permeability of the paracellular marker lucifer yellow (LY), which has a molecular weight close to that of small-molecule drugs and indicates transcellular transport.
The efflux ratio (ER) was calculated using Papp,correct in the AP-to-BL and BL-to-AP directions, as follows:
HIBCPP cells with a TEER of 150 Ω × cm2 or more were used for transcellular transport evaluation. The resistance of the HIBCPP cells was measured using an ENDOHM-6G chamber (WPI, Sarasota, FL, USA) and a Millicell ERS-2 Volt-Ohm meter (Merck Millipore, Burlington, MA, USA), and TEER was calculated according to the following equation:

Statistical Analysis
Statistical analyses were performed using Student’s t-test and one-way analysis of variance with Dunnett’s test.

Results

Evaluation of Tight Junction Formation in HIBCPP Cells
HIBCPP cell layers cultured on inserts showed a TEER of over 150 Ω × cm2. The Papp values for the paracellular markers [14C]sucrose (molecular weight [MW]: 342.30), LY (MW: 457.25), and fluorescein isothiocyanate (FITC)-dextran (average MW: 4000) in the BL-to-AP direction were 6.42 × 10–6, 3.27 × 10–6, and 0.550 × 10–6 cm/sec, respectively. These values were similar to those in the AP-to-BL direction (Fig. 1). The Papp value decreased depending on the MWs of the paracellular markers.

Protein and mRNA Expression of SLC and ABC Transporters in HIBCPP Cells
The protein and mRNA expression of 34 SLC transporters and 7 ABC transporters involved in the transport of drugs and nutrients in HIBCPP cells and HCPEpiC are listed in Table 1.

The protein expression of the energy transporters glucose transporter 1 (GLUT1/SLC2A1), GLUT3 (SLC2A3), and monocarboxylate transporter 1 (MCT1/SLC16A1) in HIBCPP cells was 432, 11.5, and 29.5-times higher than that of TBP, respectively. Among the amino acid transporters tested, L-type amino acid transporter 1 (LAT1/SLC7A5) showed the highest protein expression, followed by cationic amino acid transporter 1 (CAT1/SLC7A1) and glutamate transporter (GLAST/SLC1A3). The expression of other transporters involved in nutrient transport, such as equilibrative nucleoside transporter 1 (ENT1/SLC29A1), ENT2 (SLC29A2), choline transporter-like protein 1 (CTL1/SLC44A1), and CTL2 (SLC44A2) was 6.41, 1.16, 6.84, and 17.1-times higher than that of TBP, respectively. Among the SLC transporters, organic anion-transporting polypeptide 3A1 (OATP3A1/SLCO3A1) and SLC35F2 were detected. Among the ABC transporters, the expression of breast cancer resistance protein (BCRP/ABCG2) was 2.62-times higher than that of TBP, whereas P-glycoprotein (P-gp/multidrug resistance protein 1/ABCB1) expression was not detected in HIBCPP cells. Multidrug resistance-associated proteins (MRP) were highly expressed in the following order: MRP1 (ABCC1), MRP3 (ABCC3), MRP4 (ABCC4), MRP2 (ABCC2), and MRP5 (ABCC5). The mRNA and protein expression of transporters in HIBCPP cells was correlated with that of HCPEpiC (Fig. 2a and b).
In addition, OATP1B3 (SLCO1B3) and OATP2B1 (SLCO2B1) mRNA expression was 3.41- and 1.28-times greater than that of TBP, respectively. The mRNA expression of the organic cation transporter organic cation/carnitine transporter 2 (OCTN2/SLC22A5) was comparable with that of TBP. The mRNA expression of plasma membrane monoamine transporter (PMAT/SLC29A4) and P-gp were lower than those of TBP. The mRNA expression of these transporters was confirmed in HIBCPP cells; however, the protein expression was below the limit of detection.

Effects of Inhibitors and Low Temperature on Transporter Substrate Uptake by HIBCPP Cells
To evaluate the transport activity in HIBCPP cells, the effects of inhibitors and low temperature on substrate uptake were examined (Table 2).

Uptake of gabapentin (LAT1 substrate), [3H]L-arginine (CAT1 substrate), [3H]L-glutamate (GLAST substrate), [3H]3-O-methylglucose (GLUT1 substrate), [14C]L-lactate (MCT1 substrate), YM155 (SLC35F2 substrate), 1-methyl-4-phenylpyridinium (MPP+, PMAT substrate), and telmisartan (OATP1B3 substrate) by HIBCPP cells occurred in a time-dependent manner (Supplemental Figure), and these effects were evaluated in the initial uptake (Table 2). JPH203 (LAT1 specific inhibitor) and L-ornithine (CAT1 substrate) inhibited the uptake of gabapentin and [3H]L-arginine by 13.1 and 18.7%, respectively. The uptake of both decreased by > 90% at low temperatures. The uptake of [3H]L-glutamate and [3H]3-O-methylglucose was substantially reduced to 65.3 and 54.7%, respectively, in the presence of L-aspartic acid (GLAST substrate) and glucose (GLUT1 substrate). These uptakes decreased to 17.7 and 14.0% at low temperatures, respectively. [14C]L-Lactate uptake decreased to 28.0% in the presence of AZD3965 (MCT1 specific inhibitor) and decreased to 17.7% at low temperatures. The uptake of YM155, MPP+, and telmisartan was moderately inhibited by 40, 48.6, and 43.6%, respectively, in the presence of famotidine (SLC35F2 substrate), serotonin (PMAT substrate), and cyclosporin A (OATP1B3 inhibitor). Similarly, their uptake decreased at low temperatures, thereby indicating a temperature dependence.

Transcellular Transport Across the HIBCPP Cell Layer
Transcellular transport rates across the HIBCPP cell layer were measured in both the AP-to-BL and BL-to-AP directions (Fig. 3). The Papp values were calculated using the permeability across the HIBCPP cell layer (Eqs. 1, 2). Furthermore, Papp,correct was estimated based on LY permeability (Eq. 3). Finally, ER was estimated to quantify directional transport across the HIBCPP cell layer (Eq. 4) (Table 3).
The Papp,correct values in the AP-to-BL direction for gabapentin, [3H]L-arginine, and YM155 were 10.8 × 10–6, 1.64 × 10–6, and 3.36 × 10–6 cm/sec, respectively. The Papp,correct values in the AP-to-BL direction exceeded those in the BL-to-AP direction, which resulted in the ER values of these substrates to exceed 1 (Fig. 3a, b, and h). In contrast, the Papp,correct values in the BL-to-AP direction for [3H]L-glutamate and telmisartan were 2.33 × 10–6 and 10.5 × 10–6 cm/sec, respectively, which resulted in the ER values of these substrates to fall below 1 (Fig. 3c and j). The permeabilities of [3H]3-O-methylglucose and MPP+ showed little difference in either direction (Figs. 3d and i). Similarly, the Papp,correct values of [14C]L-lactate were comparable in both directions (Fig. 3e). In contrast, the Papp,correct values in the AP-to-BL and BL-to-AP directions for [14C]L-lactate changed to 27.3 × 10–6 and 1.44 × 10–6 cm/sec, respectively, due to pH adjustment (AP side: pH 6.0, BL side: pH 7.4; Fig. 3f). In addition, the Papp,correct value in the BL-to-AP direction (AP side: pH 7.4, BL side: pH 6.0) was 8.20 × 10–6 cm/sec (Fig. 3g).
In the ABC transporter, the Papp,correct values in the AP-to-BL and BL-to-AP directions for dantrolene (BCRP substrate) were 5.98 × 10–6 and 11.5 × 10–6 cm/sec (Fig. 3k), respectively, and the ER value fell below 1. The Papp values in the AP-to-BL and BL-to-AP directions for [3H]digoxin (P-gp substrate) were 1.54 × 10–6 and 1.80 × 10–6 cm/sec (Fig. 3l), respectively, which did not exceed the permeability of LY.

Discussion
HIBCPP cells show promising potential as model cells that facilitate the transport of substances across the human BCSFB. However, the expression of transporters responsible for transport in HIBCPP cells has not yet been fully characterized.
TEER is an indicator of tight junction integrity and was approximately 200 Ω × cm2 in HIBCPP cells. Furthermore, the permeability of paracellular markers ([14C]sucrose, LY, FITC-dextran) decreased with increasing MW (Fig. 1). These results are consistent with those of previous reports [19, 20]. The TEER value of the HCPEpiC layer was 30–50 Ω × cm2 [21]. However, the integrity of tight junctions in human BCSFB has not been clearly characterized in vivo. To evaluate the contribution of transporters at the epithelial membrane of BCSFB to transport between the blood and CSF sides, transport analyses were performed using HIBCPP cells, which exhibit higher TEER values than HCPEpiC.
Protein and mRNA expression of SLC and ABC transporters in HIBCPP cells were measured using quantitative proteomics and qRT-PCR, respectively (Table 1). A significant correlation was observed between the protein expression levels of transporters commonly expressed in HIBCPP cells and HCPEpiC (r = 0.76; Fig. 2a). Similarly, a correlation was observed in the mRNA expression between the two cell types (r = 0.62; Fig. 2b). These results suggest that HIBCPP cells could be used to analyze the transport of nutrients and drugs between the blood and CSF, making them a useful alternative model for human BCSFB. In addition, the expression of transporters was previously observed in HIBCPP cells [20, 22]; however, their functional evaluation remains insufficient. Therefore, in this study, the characteristics of transporters whose protein and/or mRNA expression was confirmed in HIBCPP cells were investigated.
It has been reported that microvilli are formed on the membrane facing the culture medium in HIBCPP cells cultured on PET membrane insert [9, 10]. Therefore, the AP (CSF side) and BL (blood side) membranes were in contact with the culture medium and PET membrane, respectively. The uptake of gabapentin and [3H]L-arginine into HIBCPP cells was reduced by more than 80% at 4 °C or in the presence of respective inhibitors (Table 2). In addition, directional transport of these amino acid transporters was observed. Transport of gabapentin and [3H]L-arginine in the AP-to-BL direction greatly exceeded that in the opposite direction (Fig. 3a, b, and Table 3). In particular, the corrected permeability of gabapentin in the AP-to-BL direction was 50-times greater than that in the BL-to-AP direction, indicating a LAT1- or CAT1-mediated efflux increase from the CSF side to the blood side. These results are supported by previous studies indicating that the concentrations of neutral or basic amino acids in the CSF, which are substrates for LAT1 and CAT1, are lower than those in the human blood [23]. This was probably due to asymmetric or unidirectional transport from the CSF to the blood through the CP membrane. In contrast to the pronounced transport of gabapentin and [3H]L-arginine in the AP-to-BL direction, the transport of [3H]L-glutamate in the BL-to-AP direction was approximately 2-times greater than that in the AP-to-BL direction in terms of Papp,correct (Fig. 3c and Table 3). The transport of the excitatory amino acid glutamate has shown previously that the transport from the blood to the brain occurs in human BBB model cells [24]; therefore, the dynamics of glutamate in the human CNS need to be evaluated in more detail.
The uptake of [3H]3-O-methylglucose was considerably reduced at 4 °C or in the presence of glucose (Table 2). Transcellular transport investigations revealed symmetrical transport (Fig. 3d and Table 3). In addition, GLUT1 was abundantly expressed in HIBCPP cells (Table 1), suggesting that GLUT1 efficiently transports glucose across these cells. Similarly, cellular uptake (Table 2), transcellular transport (Fig. 3e, f, and g), and proteomics analyses (Table 1) showed that MCT1 is the primary [14C]L-lactate transporter in HIBCPP cells. The proton-coupled co-transport phenomena (Fig. 3e, f, g, and Table 3) are consistent with previous reports [25]. Given that glucose and lactate levels in the CSF are used to determine the severity of meningitis [26, 27], the functions of GLUT1 and MCT1 in HIBCPP cells could be a major advantage as model cells for human BCSFB. However, it has been reported that multiple isoforms of GLUTs are involved in glucose transport across the BCSFB [28]. In the present study, GLUT1 and GLUT3 showed relatively high expression levels, suggesting that both transporters may be involved in glucose transport in HIBCPP cells.
This study revealed that some SLC transporters involved in drug transport exhibit abundant expression or notable functions in HIBCPP cells. SLC35F2, which specifically transports YM155, was abundantly expressed in HIBCPP cells at both the protein and mRNA levels (Table 1). In the transcellular transport analysis, YM155 was asymmetrically transported from the AP to the BL side, suggesting that SLC35F2 may be involved in efflux transport from the CSF to the blood (Fig. 3h and Table 3). PMAT is a transporter with low affinity and high capacity for catecholamine transport compared with that of monoamine transporters with high affinity, such as dopamine and serotonin transporters. Moreover, PMAT is involved in the transport of excess catecholamines in the brain [29, 30]. PMAT exhibited bidirectional transport of MPP+ in HIBCPP cells (Fig. 3i and Table 3). These results indicate that PMAT may function bidirectionally and supplementally in the transport of monoamines between the CSF and blood. However, the involvement of serotonin transporter (SERT) and organic cation/carnitine transporter 2 (OCTN2), which were confirmed to be expressed as mRNA in HIBCPP cells, also needs to be investigated. Interestingly, the Papp,correct for telmisartan was markedly greater in the BL-to-AP direction than in the AP-to-BL direction (Fig. 3j and Table 3). This suggests that OATP1B3 is predominantly involved in transport from the blood to the CSF. OATP1B3 mRNA expression in HIBCPP cells was high (Table 1), consistent with previous reports [22]. However, the OATP1B3 protein expression in HIBCPP cells was below the limit of detection. Therefore, the involvement of multiple organic anion transporters, including OATP1B3, in telmisartan transport cannot be ruled out. However, these results are important for drug delivery to the brain, and the mechanisms underlying this transport system need to be evaluated in more detail.
ABC transporters, such as P-gp and BCRP, are expressed in the CSF side of CP epithelial cells and have different functions from those of the BBB which prevents drugs from entering the brain [2, 4]. However, the functions of the ABC transporters in the human BCSFB are not fully understood. The Papp,correct of dantrolene in the BL-to-AP direction was 2-times greater than that in the opposite direction, indicating substantial transport to the CSF side (Fig. 3k and Table 3). These results suggest that BCRP in HIBCPP cells functions in the opposite direction to the BBB which prevents the penetration of substances into the brain. The importance of BCRP in the BCSFB has been reported [31], and HIBCPP cells have been suggested as useful tools for evaluating BCRP activity in the human BCSFB. The expression of P-gp in HIBCPP cells was lower than that of BCRP (Table 1), which is consistent with previous reports [20]. Considering the extremely low Papp of [3H]digoxin (Fig. 3l and Table 3), the contribution of P-gp in HIBCPP cells was minor. In HIBCPP cells, MRP1 showed the highest protein expression among the ABC transporters, followed by MRP3 and MRP4. The expression profiles of these ABC transporters were consistent with those of previous western blot analyses [20]. Among these transporters, the function of MRP1 in HIBCPP cells has been reported [20, 22]. In human BCSFB, MRP2 and MRP3 expression has also been reported [32]. The finding that MRP2 and MRP3 protein expression was confirmed only in HIBCPP cells provides a significant advantage in utilizing HIBCPP cells as a human BCSFB model. Interestingly, MRPs in the BCSFB exhibit unique expression patterns with different polarities depending on the molecular species [4]. Considering that HIBCPP cells express various MRPs, future comprehensive evaluation of these functions is necessary.

Conclusion
In this study, protein and mRNA expressions of SLC and ABC transporters in HIBCPP cells were quantified using quantitative proteomics and qRT-PCR, respectively. Cellular uptake and transcellular transport analyses demonstrated that the nutrient and drug transporters LAT1, CAT1, GLAST, GLUT1, MCT1, SLC35F2, and BCRP were functionally expressed in HIBCPP cells. These findings suggest that HIBCPP cells show promising potential as human BCSFB model cells to evaluate nutrient and drug transport between the CSF and blood, analyze the physiological functions of the CSF, monitor changes in pathological conditions, and develop new CNS-acting drugs.

Supplementary Information
Below is the link to the electronic supplementary material.