Article Cite This: Mol. Pharmaceutics XXXX, XXX, XXX−XXX
Drug Clearance from Cerebrospinal Fluid Mediated by Organic Anion Transporters 1 (Slc22a6) and 3 (Slc22a8) at Arachnoid Membrane of Rats Zhengyu Zhang, Masanori Tachikawa, Yasuo Uchida, and Tetsuya Terasaki* Division of Membrane Transport and Drug Targeting, Graduate School of Pharmaceutical Sciences, Tohoku University, Sendai 980-8578, Japan S Supporting Information *
ABSTRACT: Although arachnoid mater epithelial cells form the blood− arachnoid barrier (BAB), acting as a blood−CSF interface, it has been generally considered that the BAB is impermeable to water-soluble substances and plays a largely passive role. Here, we aimed to clarify the function of transporters at the BAB in regulating CSF clearance of watersoluble organic anion drugs based on quantitative targeted absolute proteomics (QTAP) and in vivo analyses. Protein expression levels of 61 molecules, including 19 ATP-binding-cassette (ABC) transporters and 32 solute-carrier (SLC) transporters, were measured in plasma membrane fraction of rat leptomeninges using QTAP. Thirty-three proteins were detected; others were under the quantification limits. Expression levels of multidrug resistance protein 1 (Mdr1a/P-gp/Abcb1a) and breast cancer resistance protein (Bcrp/Abcg2) were 16.6 and 3.27 fmol/μg protein (51.9- and 9.82-fold greater than in choroid plexus, respectively). Among those organic anion transporters detected only at leptomeninges, not choroid plexus, organic anion transporter 1 (oat1/Slc22a6) showed the greatest expression (2.73 fmol/μg protein). On the other hand, the protein expression level of oat3 at leptomeninges was 6.65 fmol/μg protein, and the difference from choroid plexus was within two-fold. To investigate oat1’s role, we injected paraaminohippuric acid (PAH) with or without oat1 inhibitors into cisterna magna (to minimize the contribution of choroid plexus function) of rats. A bulk flow marker, FITC-inulin, was not taken up from CSF up to 15 min, whereas uptake clearance of PAH was 26.5 μL/min. PAH uptake was completely blocked by 3 mM cephalothin (inhibits both oat1 and oat3), while 17% of PAH uptake was inhibited by 0.2 mM cephalothin (selectively inhibits oat3). These results indicate that oat1 and oat3 at the BAB provide a distinct clearance pathway of organic anion drugs from CSF independently of choroid plexus. KEYWORDS: blood−cerebrospinal fluid barrier, blood−arachnoid barrier, pharmacokinetics, organic anion transporters, oat1, oat3, leptomeninges
■
INTRODUCTION It has long been believed that the drug transport between CSF and blood is regulated by choroid plexus epithelial cells, which form the blood−cerebrospinal fluid barrier (BCSFB) and express a range of drug transporters.1,2 On the other hand, considering that the CSF volume in ventricles accounts for only about 15% and 5% of the total CSF volume in human and rat, respectively,3 and choroid plexuses exist in ventricles, it seems likely that other regulatory mechanisms contribute to BCSFB barrier function, especially in the subarachnoid space. It has been reported that the concentrations of homovanillic acid (HVA), a major metabolite of dopamine, in ventricle and cisterna magna of dog are significantly different,4 and the concentrations of methotrexate in ventricle and lumber spinal canal after CSF perfusion in human are also significantly different.5 Furthermore, multidrug resistance protein 1 (Mdr1a/P-gp/Abcb1a) and breast cancer resistance protein (Bcrp/Abcg2) are expressed at subapical and apical sides of choroid plexus epithelium.6,7 Previous work has demonstrated that penetration of topotecan, a substrate of mdr1 and bcrp into © XXXX American Chemical Society
the ventricular CSF, was decreased in Mdr1a/b(−/−)Bcrp1(−/−) mice compared to wild-type mice, whereas penetration into the brain parenchymal extracellular fluid (ECF) was increased.8 This result suggested that mdr1a and bcrp serve as influx transporters at the BCSFB, transporting substrates from blood to CSF. On the other hand, Kodaira et al. showed that the CSF-to-plasma concentration ratios (Kp,CSF) of mdr1 and bcrp substrates are significantly increased in mdr1a/1b(−/ −)and bcrp(−/−) knockout mice, respectively, compared with wild-type mice.9 We focused on the difference in sampling site of CSF in those studies; that is, CSF was collected from ventricle in the former case but from cisterna magna (subarachnoid space) in the latter case. Therefore, we hypothesized that a choroid plexus-independent transport Received: Revised: Accepted: Published: A
September 28, 2017 December 20, 2017 February 13, 2018 February 13, 2018 DOI: 10.1021/acs.molpharmaceut.7b00852 Mol. Pharmaceutics XXXX, XXX, XXX−XXX
Article
Molecular Pharmaceutics
Figure 1. Schematic illustration of (A) intraventicular administration and (B) intracisternal administration. Arrows, flow direction of injected substance. CP: choroid plexus.
dark cycle in a temperature-controlled environment with free access to food and water. All experiments were approved by the Institutional Animal Care and Use Committee in Tohoku University and performed in accordance with the guidelines in Tohoku University. Isolation of Leptomeninges from Rat. Animals were decapitated under anesthesia induced with isoflurane, and the leptomeninges were freshly isolated from Wistar rat brains (about 289 mg of wet tissue pooled from 12 males) as reported.15 Briefly, as the connection between the dura mater and the arachnoid membrane is loose,16 the dura mater is generally removed with the skull during brain extraction. The isolated brain was chilled in ice-cold normal saline for 5 min, then turned upside down, and the middle cerebral arteries (MCA) and the anterior artery were gripped with small bent-tip forceps and pulled gently in an anterior-dorsal direction to lift the leptomeninges covering the ventral anterior cortex. Then the brain was turned to a 45° to 90° angle to the Petri dish. The lateral, more anterior regions of the leptomeninges surrounding the lateral cortical regions were freed from the cortex by gripping the MCA and associated arteries and gently pulling dorsally along the cortex. To remove the most dorsal regions of the leptomeninges surrounding the primary somatosensory and motor cortex, the brain was turned dorsal side up. Finally, the leptomeninges covering the cerebellum were also gently removed with the forceps. Quantification of Protein Expression Levels. Unlabeled (standard) and stable-isotope labeled peptides (internal standard) used to quantify targeted proteins were designed by employing the in-silico peptide selection criteria reported previously17 and synthesized by Thermo Fisher Scientific (Sedanstrasse, Germany) with >95% peptide purity. The concentrations of peptide solutions were determined by quantitative amino acid analysis using an HPLC-UV system with postcolumn ninhydrin derivatization (LaChrom Elite, Hitachi, Tokyo, Japan). Other chemicals were commercial products of analytical grade. Preparation of Plasma Membrane Fraction of Rat Leptomeninges and Lysyl Endopeptidase and Trypsin Digestion. Plasma membrane fraction was prepared as previously described.18 The recovery rate of plasma membrane fraction was 437 μg protein/g wet tissue of isolated leptomeninges. The plasma membrane fractions were stored at −80 °C. The plasma membrane fraction of isolated leptomeninges (50 μg protein) was subjected to LysC and trypsin digestion as described previously.18 Clean-up of Peptide Samples for LC−MS/MS Analysis. Samples of calibration curve and digested leptomeninges for trap and elute LC−MS/MS injection were cleaned up by using GL-tip GC and SDB (GL Science Inc., Tokyo, Japan),
mechanism(s) plays a role in the uptake of substrates from CSF, especially in the subarachnoid space. Anatomical studies with electron microscopy have revealed that arachnoid mater epithelial cells form the blood−arachnoid barrier (BAB), which is a CSF-to-blood interface with complex tight junctions.10 However, it has been believed that “arachnoid membrane is impermeable to water-soluble substances and its role in forming blood−CSF barrier is largely passive”.1 Yasuda et al. recently showed that several transporters are expressed at human arachnoid mater and mouse leptomeninges at the mRNA level. In addition, P-gp and bcrp were immunohistochemically detected in mouse arachnoid mater cells but not other meningeal tissue.11 Thus, the BAB could be a functional barrier regulating drug pharmacokinetics in CSF, especially in the subarachnoid space. Recently, we have established a protein quantification method using liquid chromatography−tandem mass spectrometry (LC−MS/MS), called quantitative targeted absolute proteomics (QTAP), which enables us to selectively determine the absolute protein expression levels of multiple transporters simultaneously.12 This technique helps us to understand drug pharmacokinetics in terms of the protein expression levels of functional proteins such as transporters, receptors, and enzymes. We have found that the enzyme activities of CYP enzymes were better correlated with the protein expression levels than with the mRNA expression levels.13 Also, protein expression of organic cation/carnitine transporter 1 (OCTN1/ SLC22A4) and multidrug resistance-associated protein 1 (MRP1/ABCC1) in plasma membrane fraction correlated well with the measured transport activity.14 Therefore, quantification of protein levels of transporters, rather than mRNA levels, is needed to functionally understand drug distribution and metabolism. The purpose of the present study was to verify that transporters on the BAB are actively involved in regulating drug pharmacokinetics in CSF, independently of choroid plexus, by means of QTAP and in vivo analyses in rats. We applied the QTAP technique to plasma membrane fraction of rat leptomeninges to measure the absolute protein expression levels of transporters at the BAB. For in vivo functional analysis, we focused on water-soluble organic anion transporters since water-soluble organic anions show low passive diffusion and low binding to cell membranes. To avoid the influence of choroid plexus, we employed intracisternal administration instead of intraventricular administration (i.c.v) (Figure 1).
■
MATERIALS AND METHODS Animals. Male adult Wistar rats (10 weeks of age, 220−280 g body weight) were purchased from Japan SLC, Inc. (Shizuoka, Japan). They were maintained on a 12-h light/ B
DOI: 10.1021/acs.molpharmaceut.7b00852 Mol. Pharmaceutics XXXX, XXX, XXX−XXX
Article
Molecular Pharmaceutics
Table 1. Protein Expression Levels of Transporters and Marker Protein in Rat Brain Leptomeninges and Choroid Plexus protein expression level (fmol/μg protein)a molecule name Abc Transporter Abca1/Abc1 Abca2/Abc2 Abca8a Abca9 Abcb1a/Mdr1a Abcc1/Mrp1 Abcc4/Mrp4 Abcc6/Mrp6 Abcc7/Mrp7 Abcg2/Bcrp Slc Transporter Slc1a2/Eaat2 Slc2a1/Glut1 Slc3a2/4f2hc Slc4a2/Ae2 Slc6a8/Crt1 Slc7a8/Lat2 Slc15a2/Pept2 Slc16a1/Mct1 Slc16a8/Mct3 Slc19a1/Rfc1 Slc21a5/Oatp1a4 Slc21a7/Oatp1a5 Slc21a11/Oatp3a1 Slc21a14/Oatp1c1 Slc22a6/Oat1 Slc22a8/Oat3 Slc38a2/Ata2 Others CD147 Membrane Marker Na+/K+-ATPase Tight Junction Marker Claudin5 Occludin Astrocyte Marker GFAP Pericyte Marker NG2 Neuron Marker SYP Endothelial Cell Marker γ-gtp
fold difference
leptomeninges
choroid plexusb
leptomeninges/choroid plexus
0.262c ± 0.020 0.620c ± 0.033 1.70 ± 0.17 1.58c ± 0.41 16.6 ± 0.2 0.671 ± 0.107 0.510 ± 0.045 0.165c ± 0.037 0.118 ± 0.013 3.27 ± 0.10
ULQ (0.201 9.82
18.8 ± 1.1 56.6 ± 0.4 2.45 ± 0.05 0.726 ± 0.050 0.381 ± 0.018 ULQ (
