Role of l-Type Amino Acid Transporter 1 at the ... - ACS Publications

Gabapentin is an antiseizure drug that is known to also have beneficial effects on the retinal cells. To use gabapentin in retinal pharmacotherapy, it...
0 downloads 0 Views 1MB Size
Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

Role of L-type amino acid transporter 1 at the inner bloodretinal barrier in blood-to-retina transport of gabapentin Shin-ichi Akanuma, Atsuko Yamakoshi, Takeshi Sugouchi, Yoshiyuki Kubo, Anika M.S. Hartz, Bjoern Bauer, and Ken-ichi Hosoya Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.8b00179 • Publication Date (Web): 24 Apr 2018 Downloaded from http://pubs.acs.org on April 25, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Role of L-type amino acid transporter 1 at the inner blood-retinal barrier in blood-to-retina transport of gabapentin

AUTHOR NAMES Shin-ichi Akanuma†, Atsuko Yamakoshi†, Takeshi Sugouchi†, Yoshiyuki Kubo†, Anika M.S. Hartz§‡, Björn Bauer£, Ken-ichi Hosoya*†



Department of Pharmaceutics, Graduate School of Medicine and Pharmaceutical

Sciences, University of Toyama, Sugitani 2630, Toyama, Japan §

Sanders-Brown Center on Aging, University of Kentucky, Lexington, KY, USA



Department of Pharmacology and Nutritional Sciences, College of Medicine,

University of Kentucky, Lexington, KY, USA £

Department of Pharmaceutical Sciences, College of Pharmacy, University of Kentucky,

Lexington, KY, USA

1 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

*Corresponding author: Professor Ken-ichi Hosoya, Ph.D. Department of Pharmaceutics, Graduate School of Medicine and Pharmaceutical Sciences, University of Toyama, Sugitani 2630, Toyama, 930-0194, Japan Phone: +81-76-434-7505; Fax: +81-76-434-5172 Email: [email protected]

2 ACS Paragon Plus Environment

Page 2 of 50

Page 3 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Abbreviations b0,+AT1, b(0,+)-type amino acid transporter 1; BBB, blood-brain barrier; BCH, 2-aminobicyclo[2.2.1]heptane-2-carboxylic acid; BRB, blood-retinal barrier; C/M, cell/medium; DC, n-octanol/buffer (pH 7.4) distribution coefficient; GABA, γ-aminobutyric acid; MeAIB, methylaminoisobutyric acid; OCT, organic cation transporter; OCTN, organic cation/carnitine transporter; PBS(-), phosphate-buffered saline without Ca2+ and Mg2+; rBAT, related to b(0,+) amino acid transport protein; RPE, retinal pigment epithelial; RPE-J cells, a conditionally-immortalized rat RPE cell line; RUI, retinal uptake index; siRNA, small interfering RNA; TEA, tetraethylammonium; TR-iBRB cells, a conditionally-immortalized rat retinal capillary endothelial cell line; y+LAT, y+L amino acid transporter 1.

3 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Table of Contents/Abstract Graphic

4 ACS Paragon Plus Environment

Page 4 of 50

Page 5 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Abstract Gabapentin is an anti-seizure drug that is known to also have beneficial effects on the retinal cells. To use gabapentin in retinal pharmacotherapy, it is critical to understand gabapentin distribution into the retina. The purpose of this study was to clarify the kinetics of gabapentin influx transport across the inner and outer blood-retinal barrier (BRB), which regulates the exchange of compounds/drugs between the circulating blood and the retina. In vivo blood-to-retina gabapentin transfer was evaluated by the rat carotid artery injection technique. In addition, gabapentin transport was examined using in vitro models of the inner (TR-iBRB2 cells) and outer BRB (RPE-J cells). In vivo [3H]gabapentin transfer to the rat retina across the BRB was significantly reduced in the presence of unlabeled gabapentin, suggesting transporter-mediated blood-to-retina distribution of gabapentin. Substrates of the Na+-independent L-type amino acid transporter 1 (LAT1), such as 2-aminobicyclo[2.2.1]heptane-2-carboxylic acid (BCH), also significantly inhibited the in vivo [3H]gabapentin transfer. [3H]Gabapentin uptake in TR-iBRB2 and RPE-J cells exhibited Na+-independent and saturable kinetics with a Km of 735 µM and 507 µM, respectively. Regarding the effect of various transporter substrates/inhibitors on gabapentin transport in these cells, LAT1 substrates significantly inhibited [3H]gabapentin uptake in TR-iBRB2 and RPE-J cells. In addition, preloaded [3H]gabapentin release from TR-iBRB2 and RPE-J cells was trans-stimulated by LAT1 substrates through the obligatory exchange mechanism as LAT1. Immunoblot analysis indicates the protein expression of LAT1 in TR-iBRB2 and RPE-J cells. These results imply that LAT1 at the inner and outer BRB takes part in gabapentin transport between the circulating blood and retina. Moreover, treatment of LAT1-targeted small interfering RNA to TR-iBRB2 cells significantly reduced both the level of LAT1 protein 5 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

expression and [3H]gabapentin uptake activities in TR-iBRB2 cells. In conclusion, data from the present study indicate that LAT1 at the inner BRB is involved in retinal gabapentin transfer, and also suggest that LAT1 mediates gabapentin transport in the RPE cells.

6 ACS Paragon Plus Environment

Page 6 of 50

Page 7 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Keywords Blood-retinal barrier, inner blood-retinal barrier, outer blood-retinal barrier, gabapentin, L-type amino acid transporter, LAT1

7 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Introduction Gabapentin is an analog of the inhibitory neurotransmitter γ-aminobutyric acid (GABA) and is used for the treatment of epilepsy and neuropathic pain since it has been reported that gabapentin .1, 2 Despite gabapentin’s structural similarity to GABA, it does not act at the GABAA receptor, but instead binds to the α2δ1 subunit that is associated with the voltage-gated calcium channel, thereby inhibiting neuronal glutamate release, and attenuating the epileptic seizure and pain reaction.3 In addition, data from ex vivo experiments using isolated rat retina showed that gabapentin inhibits the retinal de novo synthesis of L-glutamine and L-glutamate,4, 5 which, at high concentrations, contribute to retinal ganglion cell toxicity.6 In this regard, Farrell et al. have demonstrated that gabapentin reduces voltage-gated calcium channel-mediated excitotoxicity in cultured retinal ganglion cells, thereby reducing cell death.7 Apoptosis of retinal ganglion cells has been observed in retinal ischemia and glaucoma, resulting in a progressive loss of vision.8-10 A more detailed knowledge of the kinetics of retinal gabapentin distribution is necessary to apply gabapentin for the treatment of retinal diseases. However, the gabapentin distribution processes that occur at the retina are not fully understood. Drug distribution from the circulating blood to the retina is regulated by the inner and outer blood-retinal barrier (BRB). Non-selective drug exchange between the circulating blood and retina across the inner and outer BRB is restricted by the tight junctions of the retinal capillary endothelial cells and the retinal pigment epithelial (RPE) cells, respectively.11,

12

However, examples for facilitative blood-to-retina

transport at the inner BRB are known. It has been reported that L-carnitine and branched-chain amino acids such as L-leucine are transported from the circulating blood to the retina through the organic cation/carnitine transporter 2 (OCTN2/solute carrier 8 ACS Paragon Plus Environment

Page 8 of 50

Page 9 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

(SLC) 22A5) and the L-type neutral amino acid transporter 1 (LAT1/SLC7A5).13, 14 Our functional analyses have revealed that the branched-chain amino acid transporters, LAT1 and LAT2 (SLC7A8), are involved in L-leucine transport in the human RPE cell line, ARPE-19.15 In addition to these LATs, Nakauchi et al. have reported that the y+L amino acid transporter 1 (y+LAT1/SLC7A7) is also expressed in another human RPE cell line, hTERT-RPE cells.16 In summary, transporters at the inner and outer BRB are considered responsible for the active transfer of compounds from the circulating blood to the retina. To specifically understand gabapentin distribution to the retina, gabapentin transport and that of any contributing molecules at the inner and outer BRB need to be identified. Several reports describe plasma membrane transporter-mediated permeation of gabapentin across membranes into mammalian tissues. For example, Nguyen et al. have demonstrated that gabapentin transport at the apical membrane of the rat intestine involves the system b0,+,17 which is composed of the b(0,+)-type amino acid transporter 1 (b0,+AT1/Slc7a9) and related to the b(0,+) amino acid transport protein (rBAT/Slc3a1).18 For the organic cation/carnitine transporters (OCTN), some reports question whether gabapentin is a substrate of OCTN2/SLC22A5.19,

20

OCTN1/SLC22A4 has been shown to accept gabapentin as a substrate21,

However, 22

and, in

individuals carrying the OCTN1-L503F polymorphism, active tubular secretion of gabapentin in the kidney is reduced or absent.21 Another SLC family member, LAT1, has been reported to accept gabapentin as a substrate,19 whereas LAT2 does not transport gabapentin.20 Furugen et al. using human placental choriocarcinoma cells showed that LAT1 is involved in gabapentin uptake.23 Furthermore, in rat brain astrocytes and synaptosomes, gabapentin uptake was saturable and was strongly 9 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

inhibited by L-leucine, whereas cationic amino acids, such as L-ornithine and L-lysine, had no effect on gabapentin uptake kinetics.24 With regard to gabapentin distribution from blood to brain, Wang et al. performed microdialysis studies to determine gabapentin influx and efflux kinetics at the rat blood-brain barrier (BBB), which separates brain from blood.25 In addition, researchers found that gabapentin distribution from the blood in vivo in rat is saturable26 and data from experiments with in vitro human BBB models show LAT1-mediated gabapentin uptake.19 Taken together, gabapentin transport in tissues and the BBB is regulated by SLC family transporters that accept gabapentin as a substrate. Despite all this evidence of gabapentin transport in various organs and barrier tissues, there is no literature report on gabapentin transport between the circulating blood and the retina. To examine the details of gabapentin distribution from the circulating blood to the retina across the BRB, we examined in vivo rat blood-to-retina gabapentin transport with the retinal uptake index (RUI) method using the carotid artery injection technique.27 In addition, we used a conditionally-immortalized rat retinal capillary endothelial cell line #2 and a conditionally-immortalized rat RPE cell line, TR-iBRB2 and RPE-J, to identify the molecular mechanism(s) of gabapentin transport across the inner and outer BRB.28, 29

10 ACS Paragon Plus Environment

Page 10 of 50

Page 11 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Experimental Section Reagents Butanol, n [1-14C] ([14C]n-butanol, 2 mCi/mmol) and gabapentin [3H(G)]([3H]gabapentin, 256 Ci/mmol) were purchased from American Radiolabeled Chemicals (St. Louis, MO, USA) and PerkinElmer (Boston, MA, USA), respectively. All other reagents were analytical grade and commercial products.

Animals All animal experiments conducted in this study have been approved by the Animal Care Committee of the University of Toyama (#A2017PHA-6; principle investigator, Hosoya K.). Adult male Wistar rats (150-180 g) were obtained from Japan SLC (Hamamatsu, Japan) and maintained in a controlled environment (temperature, ~23 °C; humidity, 40-50 %; dark/light cycle, 12 h).

In vivo injection into rat carotid artery This in vivo injection of 6.0 µCi [3H]gabapentin and 0.35 µCi [14C]n-butanol without (control) or with 10 mM of transporter substrates/inhibitors was performed as described previously, and the detail has been given in the Supporting Information.30 In this study, [14C]n-butanol has not been identified as a marker of retinal blood flow exactly, but known as a freely-diffusible internal reference compound.27, 30 It has been reported that the relationship between the RUI value, which is obtained by using [14C]n-butanol as an internal reference, and the n-octanol/buffer (pH 7.4) distribution coefficient (DC) of compounds transported across the BRB by passive diffusion has been obtained.30 Hence, the RUI result by using [14C]n-butanol leads to the discussion of gabapentin transport 11 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 50

manners across the BRB. [3H]Gabapentin distribution into the retina relative to that of [14C]n-butanol was expressed as the retinal uptake index (RUI) using Eq. 1. [3H]Gabapentin/[14C]n-butanol in the retina × 100

RUI (%) =

(1)

[3H]Gabapentin/[14C]n-butanol in injectate

Culture of TR-iBRB2 and RPE-J cells and in vitro transport analysis TR-iBRB2 and RPE-J cells were maintained as described in previous reports.31 For [3H]gabapentin transport studies, TR-iBRB2 and RPE-J cells were plated at a density of 0.5 × 105 cells/cm2 onto Corning Biocoat Collagen-I 24-well plates (Corning, Kennebunk, ME, USA). The detail procedure of gabapentin uptake and efflux studies13, 14

has been included in the Supporting Information.13, 14

Data analysis of in vitro transport studies [3H]Gabapentin cellular uptake was expressed as the distribution volume, namely the cell/medium (C/M) ratio (eq. 2). [3H]Gabapentin amount per cell protein (dpm/mg protein) C/M ratio (µL/mg protein) =

(2) [3H]Gabapentin concentration in buffer (dpm/µL)

From the data of time-course of [3H]gabapentin uptake by these cultured cells, the initial uptake clearance and initial distribution volume, which reflects the surface binding of compounds to the cells and rapid compound uptake within the short time, were obtained by using equation 3, using MULTI, a program for nonlinear least-square regression analysis.32 C/M ratio (µL/mg protein)

=

Initial uptake clearance (µL/(min•mg protein))

×

Time (min)

12 ACS Paragon Plus Environment

+

Initial distribution volume (µL/mg protein)

(3)

Page 13 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

For the concentration-dependent analysis of gabapentin cellular uptake, the value for the maximal uptake rate (Vmax) and the value for the Michaelis-Menten constant (Km) of gabapentin uptake by TR-iBRB2 cells, as well as the values for Vmax, Km, and the non-saturable uptake clearance (Kd) of gabapentin uptake by RPE-J cells were obtained from equations 4 and 5, respectively, using MULTI, V = (Vmax × [S]) / (Km + [S])

(4)

V = (Vmax × [S]) / (Km + [S]) + Kd × [S]

(5)

In these equations, V and [S] are the uptake rate of gabapentin and the gabapentin concentration in buffer, respectively. [3H]Gabapentin efflux from cells was expressed as a percentage of the fractional outflow from the intracellular compartment into the incubation buffer (eq. 6). [3H]Gabapentin in incubation buffer (dpm) (6)

Outflow (%) = [3H]Gabapentin in the cells (dpm) + [3H]gabapentin in incubation buffer (dpm)

Immunoblot analysis TR-iBRB2 and RPE-J cells which were cultured onto Collagen I-coated 100 mm dishes at ~100 % confluency were rinsed twice by phosphate-buffered saline without Ca2+ and Mg2+ (PBS(-); 137 mM NaCl, 8.1 mM Na2HPO4, 2.7 mM KCl, 1.5 mM KH2PO4) and collected. The cells were treated with hypotonic lysis buffer (10 mM Tris-HCl, 10 mM NaCl, 1.5 mM MgCl2, pH 7.4) at 4 °C for 1 h and then homogenized. The homogenate was centrifuged (10,000 × g, 4 °C, 15 min), and the supernatant was again centrifuged (100,000 × g, 4 °C, 60 min). The resulting pellets were suspended in a suspension buffer (10

mM

Tris-HCl,

250

mM

sucrose,

1

mM

O,O'-bis(2-aminoethyl)ethyleneglycol-N,N,N',N'-tetraacetic acid-NaOH, pH 7.4) and 13 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

used as a crude membrane fraction. The protein concentration of this fraction was determined using a detergent-compatible protein assay kit (BIO-RAD). The crude membrane fractions (5 µg/lane) were separated on 7.5 % sodium dodecyl sulfate (SDS)-polyacrylamide gel and then electroblotted onto a polyvinylidene difluoride (PVDF) membrane (Amersham Hybond P PVDF 0.45; GE healthcare, Chalfont St. Giles, UK). Following incubation with Tris-buffered saline (TBS; 25 mM Tris-HCl, 125 mM NaCl, pH 7.4) containing 0.1 % Tween-20 and 4 % non-fat dry milk for 14 h at 4 °C, the membranes were incubated with rabbit anti-LAT1 (0.1 µg/mL; Trans Genic, Fukuoka, Japan) or mouse anti-Na+, K+-ATPase α1 antibodies (0.1 µg/mL; EMD Millipore, Darmstadt Germany) for 3 h at ~15 °C. After the subsequent treatment of horseradish peroxidase-conjugated anti-rabbit/mouse IgG, the bands were visualized with ECL Prime Western Blotting Detection System (GE healthcare).

Immunoblot and gabapentin transport analyses by TR-iBRB2 cells after LAT1-targeted small interfering RNA treatment For the LAT1 protein expression analyses, TR-iBRB2 cells were seeded onto 100 mm Collagen I-coated plastic dishes at a density of 0.1 × 105 cells/cm2. Since the cellular proliferation of TR-iBRB2 cells is attenuated in the procedure of small interfering RNA (siRNA) transfection, the cells were cultivated until 80-90 % confluency. The cells were rinsed twice with Opti-MEM I Reduced Serum media (ThermoFisher Scientific, Waltham, MA, USA). The cells were treated with siRNAs targeting to rat LAT1 (s132356, ThermoFisher Scientific), which was used in the previous report,33 or Silencer select Negative control No.1 siRNA (ThermoFisher Scientific) using Lipofectamine RNAiMAX (ThermoFisher Scientific) following the manufacture’s 14 ACS Paragon Plus Environment

Page 14 of 50

Page 15 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

protocol. After 48-h cultivation, the crude membrane fraction of TR-iBRB2 cells was prepared as described in the “Immunoblot analysis” subsection. This fraction (10 µg/lane) was separated 10 % SDS-polyacrylamide gel and then electroblotted onto the PVDF membrane. Following incubation with TBS containing 0.1 % Tween-20 and 4 % non-fat dry milk for 14 h at 4 °C, the membranes were incubated with the rabbit anti-LAT1 (0.4 µg/mL) or the mouse anti-Na+, K+-ATPase α1 antibodies (0.1 µg/mL) for 3 h at ~12 °C. After the subsequent treatment of horseradish peroxidase-conjugated anti-rabbit/mouse IgG, the bands were visualized with ECL Prime Western Blotting Detection System (GE healthcare). The band intensities were quantified by Image J (U.S. National Institutes of Health, Bethesda, Maryland, USA). The protein expression level of LAT1 normalized by that of Na+, K+-ATPase α1 was obtained from an equation 7. Intensity of a band at ~40 kDa obtained by anti-LAT1 antibodies

LAT1 protein level

=

(7) Intensity of a band at ~110 kDa obtained by anti-Na+, K+,ATPase α1 antibodies

For the transport studies, TR-iBRB2 cells were seeded onto 6-well Collagen I-coated microplate (AGC Techno Glass, Shizuoka, Japan) at a density of 0.1 × 105 cells/cm2. The LAT1 siRNAs and Silencer select Negative control No.1 siRNA were treated as described above. The [3H]gabapentin uptake by TR-iBRB2 cells was performed as described in the subsection of “Culture of TR-iBRB2 and RPE-J cells and in vitro transport analysis”. The [3H]gabapentin uptake by negative control siRNA- and LAT1 siRNA-transfected TR-iBRB2 cells was expressed as C/M ratio (eq. 2).

Statistical analysis 15 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

All kinetic parameters, such as initial uptake clearance, initial distribution volume, Vmax, Km, and Kd, are expressed as mean ± SD. Other data are expressed as mean ± SEM. The normality of data in the group was tested by the Kolmogorov-Smirnov test. Regarding the test, 5 % as a p-value was utilized to judge the normal distribution, and it was shown that all data in this study were normally distributed. Statistically significant differences between the means of two or more than three groups were determined using the unpaired two-tailed Student’s t-test or one-way ANOVA followed by Dunnett’s test, respectively. A p-value of less than 5 % was considered statistically significant.

16 ACS Paragon Plus Environment

Page 16 of 50

Page 17 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Results In vivo transfer of gabapentin to rat retina across the BRB Table 1 summarizes the data for [3H]gabapentin uptake into the rat retina after administration into the common carotid artery. The [3H]gabapentin RUI was 143 % compared with the [14C]n-butanol control, which means that transfer to the retina of [3H]gabapentin is 1.43-fold greater than that of [14C]n-butanol. In addition, retinal uptake of [3H]gabapentin was significantly reduced by 58 % (p = 0.005) and 53 % (p = 0.006) in the presence of 10 mM unlabeled gabapentin and the LAT inhibitor 2-aminobicyclo[2.2.1]heptane-2-carboxylic acid (BCH), respectively. In contrast, retinal [3H]gabapentin uptake was not significantly altered by L-alanine (p = 0.0621; substrate of neutral and small amino acid transporters), acetyl-L-carnitine (p = 0.117; OCTN substrate),

GABA

(p

=

0.230;

substrate

of

GABA

transporters),

and

tetraethylammonium (TEA; p = 0.357; substrate of typical organic cation transporters including OCTN).

[3H]Gabapentin uptake in TR-iBRB2 and RPE-J cells Since using the compound in the in vivo inhibition studies has been limited, we investigated [3H]gabapentin uptake TR-iBRB2 (Figures 1 and 2) and RPE-J cells (Figures 3 and 4) to characterize the transport of gabapentin at the inner and outer BRB, respectively. [3H]Gabapentin uptake into TR-iBRB2 cells at 37 °C exhibited a time-dependent manner for at least 3 min with initial distribution volume of 9.27 ± 2.25 µL/mg protein (Figure 1A, open circle). At 1 min, the [3H]gabapentin uptake at 4 °C was obtained to be 2.62 ± 0.22 µL/mg protein (Figure 1A, closed square), and significantly reduced by 94 % compared with that at 37 °C. Since the initial distribution 17 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 50

volume at 37 °C is greater than the [3H]gabapentin uptake at 4 °C, it is implied that the rapid uptake of [3H]gabapentin within 20 sec, which is the shortest time in this study (Figure 1A, open circle), in addition to the surface binding of gabapentin onto TR-iBRB2 cells. Hence, there is a possibility that an initial uptake clearance of gabapentin in TR-iBRB2 cells, which was obtained to be 29.1 ± 1.4 µL/(min•mg protein) from the data between 20 sec to 3 min, is underestimated. In addition, gabapentin

uptake

into

TR-iBRB2

cells

at

1

min

exhibited

saturable,

concentration-dependent characteristics with a Vmax and Km of 13.1 ± 2.6 nmol/(min•mg protein) and 735 ± 198 µM, respectively (Figure 1B). [3H]Gabapentin uptake by TR-iBRB2 cells was not altered in the absence of extracellular Na+ (Figure 2A, Na+-free) or Cl- (Figure 2A, Cl--free). In addition, replacing Na+ in the uptake buffer with K+ had little effect on [3H]gabapentin uptake in TR-iBRB2 cells (Figure 2A, K+-replacement), indicating that the resting membrane potential does not affect gabapentin uptake by these cells. To examine if H+/substrate co-transport or anti-port is involved in [3H]gabapentin uptake by TR-iBRB2 cells, we performed experiments at a different extracellular pH. At pH 6.4 and 8.4, [3H]gabapentin uptake by TR-iBRB2 cells was not significantly changed relative to the control pH 7.4 (Figure 2B). Figure 3A shows that [3H]gabapentin uptake by RPE-J cells was time-dependent (open circle) with an initial uptake clearance and initial distribution volume of 20.0 ± 0.5 µL/(min•mg protein) and 1.89 ± 1.40 µL/mg protein, respectively. At 1 min, [3H]gabapentin uptake at 4 °C (Figure 3A, closed square) was significantly reduced by 90 % compared with that at 37 °C. Given that gabapentin distribution into RPE-J cells was linear for at least 5 min, all subsequent uptake studies in RPE-J cells were conducted for a period of 3 min. In the concentration-dependent experiment shown 18 ACS Paragon Plus Environment

Page 19 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

in Figure 3B, gabapentin uptake into RPE-J cells exhibited both saturable and non-saturable kinetics with Vmax, Km, and Kd values of 12.4 ± 1.6 nmol/(min•mg protein), 507 ± 71 µM, and 3.81 ± 0.36 µL/(min•mg protein), respectively. [3H]Gabapentin uptake by RPE-J cells was not altered in the absence of extracellular Na+ (Figure 4A, Na+-free) or Cl- (Figure 4A, Cl--free). In addition, replacing Na+ with K+ in the uptake buffer had little effect on the [3H]gabapentin uptake by RPE-J cells (Figure 4A, K+-replacement), indicating that the resting membrane potential did not affect gabapentin transport into these cells. [3H]Gabapentin uptake by RPE-J cells was also determined at different extracellular pH values (Figure 4B). [3H]Gabapentin uptake by RPE-J cells at pH 6.4 was 1.15-fold greater compared with pH 7.4 (control), which was statistically significant, whereas [3H]gabapentin uptake at pH 8.4 was not significantly changed relative to control.

[3H]Gabapentin uptake is blocked by transporter inhibitors To identify molecules that contribute to gabapentin transport across the inner and outer BRB, we tested the effect of several SLC transporter substrates on [3H]gabapentin uptake by TR-iBRB2 and RPE-J cells (Table 2). The concentration of all substrates/inhibitors for transporters was settled to be 1 mM to observe the inhibitory effect on interested molecules. [3H]Gabapentin uptake by TR-iBRB2 cells was significantly reduced by more than 37 % in the presence of unlabeled gabapentin as well as several L-type neutral amino acids, such as L-phenylalanine, L-leucine, L-tryptophan, L-isoleucine, L-methionine, L-tyrosine, L-valine, and L-cysteine. L-Glutamine, L-alanine, L-aspargine, L-serine, and L-arginine did not have a significant effect on [3H]gabapentin uptake by TR-iBRB2 cells. In contrast, the LAT inhibitor BCH 19 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 50

produced significant inhibition (81 %) of [3H]gabapentin uptake by TR-iBRB2 cells. In addition, D-leucine and D-phenylalanine, which are D-type neutral amino acids and substrates of LAT1 and LAT3, significantly inhibited [3H]gabapentin uptake by 65-68 %. However, [3H]gabapentin uptake by TR-iBRB2 cells was not significantly inhibited by methylaminoisobutyric acid (MeAIB; a substrate of sodium-dependent neutral amino acid transporter), L-carnitine (a substrate of OCTN), TEA (a substrate of typical organic cation transporters), and GABA.

Stimulation of [3H]gabapentin release from the intercellular compartment LAT and system b0,+ are obligatory substrate exchangers.34,

35

To evaluate the

contribution of these SLC transporters to the distribution processes at the inner and outer BRB, we measured [3H]gabapentin efflux from TR-iBRB2 and RPE-J cells. [3H]Gabapentin efflux from both TR-iBRB2 and RPE-J cells was significantly increased by 1 mM unlabeled gabapentin (Table 3), indicating the presence of a gabapentin/substrate exchanger in the inner and outer BRB. [3H]Gabapentin efflux from TR-iBRB2 cells was increased at least 2.1-fold by adding 1 mM D-leucine, BCH, L-leucine, and L-phenylalanine, which are all LAT substrates. In contrast, extracellular addition of 1 mM L-arginine (a substrate of system b0,+), L-carnitine (a substrate of OCTN), TEA (a substrate of various organic cation transporters), and GABA did not stimulate [3H]gabapentin efflux (Table 3). In RPE-J cells, [3H]gabapentin efflux was increased at least 1.7-fold by adding 1 mM D-leucine, BCH, L-leucine, and L-phenylalanine. The addition of 1 mM L-arginine increased [3H]gabapentin efflux from RPE-J cells 1.2-fold, whereas 1 mM L-carnitine, TEA, and GABA had little effect on [3H]gabapentin efflux (Table 3). 20 ACS Paragon Plus Environment

Page 21 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

LAT1 protein expression in TR-iBRB2 and RPE-J cells Among LATs, LAT1 is known to accept gabapentin as a substrate.19 By the immunoblot analysis using anti-LAT1 antibodies, the protein expression of LAT1 in the BRB model cells was examined. A single band at ~40 kDa was detected in the crude membrane fraction prepared from TR-iBRB2 cells (Figure 5A, left panel). The position of the band was consistent with that in the previous report.14, 36 In addition, a band at ~40 kDa was detected in the crude membrane fraction prepared from RPE-J cells (Figure 5B, left panel).

LAT1 protein expression level and [3H]gabapentin transport activities in TR-iBRB2 cells after LAT1 RNA interference To elucidate the contribution of LAT1 at the inner BRB to gabapentin transport across the BRB, we examined the alteration of gabapentin uptake by TR-iBRB2 cells after the treatment of siRNAs which specifically target to rat LAT1. Since a single band at ~110 kDa was detected in the fraction of TR-iBRB2 (Figure 5A, right panel) and RPE-J cells (Figure 5B, right panel) using anti-Na+, K+-ATPase α1 antibodies, it is indicated that the protein expression of Na+, K+-ATPase α1 is able to apply the internal standard to examine the alteration of LAT1 protein expression level under this gene-knockdown study. Regarding the protein expression level of LAT1 in TR-iBRB2 cells after the treatment of LAT1 siRNAs, the intensity of the LAT1-derived band at ~40 kDa in LAT1 siRNA-treated TR-iBRB2 cells was significantly decreased by 60 % compared with that in negative control siRNA-treated TR-iBRB2 cells (Figure 6A). The [3H]gabapentin uptake by LAT1 siRNA-treated TR-iBRB2 cells was significantly reduced by 36 % 21 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

compared with that by negative control siRNA-treated TR-iBRB2 cells (Figure 6B). Since LAT1-mediated transport activity of the substrate is considered to be decreased by at most 60 % under the condition of this LAT1 siRNA transfection, the LAT1 contribution ratio of gabapentin uptake by TR-iBRB2 cells was obtained from dividing the degree of the decrease of gabapentin uptake by TR-iBRB2 cells under this condition (= 0.36) by that of LAT1 protein expression level (= 0.60). From this calculation, this LAT1 contribution ratio was calculated to be at least 60 % (= 0.36/0.60 × 100).

22 ACS Paragon Plus Environment

Page 22 of 50

Page 23 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Discussion The data obtained in this RUI study (Table 1) indicate active blood-to-retina transport of gabapentin across the BRB. In our previous study, we analyzed the relationship between the RUI and the DC of compounds transported across the BRB by passive diffusion, and expressed the relationship as “RUI = 46.2 × exp(0.515 × log DC)”.30 Using the experimentally determined logarithmic DC for gabapentin (-1.21),19 we estimated the RUI from this equation to be 24.8 %. This estimated RUI value is 5.8-fold lower than the actual RUI value of gabapentin (143 %). Moreover, in vivo [3H]gabapentin retinal transfer was significantly reduced by co-administration of unlabeled gabapentin. These findings suggest carrier-mediated transport of gabapentin from the circulating blood, across the BRB, to the retina. Using cell lines that mimic the inner and outer BRB, we analyzed the in vitro transport of [3H]gabapentin. Since [3H]gabapentin uptake by TR-iBRB2 and RPE-J cells exhibited temperature-dependence and saturable kinetics (Figures 1 and 3), the involvement of carrier-mediated processes in gabapentin transport at the inner and outer BRB is suggested. In RPE-J cells, concentration-dependent gabapentin transport had both a saturable and a non-saturable component (Figure 3B). The uptake clearance of the saturable process of gabapentin uptake by RPE-J cells was found to be 24.5 µL/(min•mg protein) (= Vmax/Km), which is 87 % of the total gabapentin uptake processes (= Vmax/Km ÷ (Vmax/Km + Kd) × 100). The therapeutic range of gabapentin has been reported to be 11.7-117 µM.37 With the maximal concentration of this therapeutic range (= 117 µM), the gabapentin uptake rate of the saturable process (= Vmax × [S] ÷ (Km + [S])) was obtained to be 2.33 nmol/(min•mg protein), which is 5.2-fold greater than that of non-saturable processes (= Kd × [S]; 0.446 nmol/(min•mg protein)). Taking 23 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

these points into consideration, the non-saturable transport processes play a minor role in gabapentin transport at the outer BRB. Moreover, it is suggested that carrier-mediated blood-to-retina gabapentin transport across the inner and outer BRB is not saturated by gabapentin itself in patients receiving gabapentin therapeutically since this the gabapentin therapeutic range is below the Km values of gabapentin uptake by TR-iBRB2 (735 µM; Figure 1B), RPE-J cells (507 µM, Figure 3B). Consequently, these data suggest that carrier-mediated influx transport plays an important role in blood-to-retina gabapentin transport across both the inner and outer BRB. The in vivo rat blood-to-retina [3H]gabapentin transfer was significantly reduced by BCH, but other transporter substrates, such as small neutral amino acids and the organic cationic compounds had no effect (Table 1). This result indicates that BCH-selective molecular mechanism(s) are involved in gabapentin influx transport across the BRB. Based on data from previous reports and our own data on gabapentin transport, we proposed that LAT, OCTN, and b0,+AT1 are candidates for gabapentin transport at the BRB. Regarding the expression of transporters for the candidates, mRNAs of LAT1-2 and OCTN1 have been detected in rat retinal capillary endothelial cells,13,

14

whereas b0,+AT1 mRNA has not been found in TR-iBRB2 cells.38 Using

TR-iBRB2 cells, we observed that [3H]gabapentin uptake was insensitive to changes in extracellular Na+, Cl-, pH, and the resting membrane potential (Figure 2). Of the transporters that accept gabapentin as a substrate, LAT and b0,+AT transport is insensitive to extracellular Na+ and Cl- changes,39-43 whereas the antiporter OCTN1 and the H+/substrate antiporter are sensitive to changes in extracellular Na+.44-46 Although the Km of gabapentin uptake by TR-iBRB2 cells (735 µM; Figure 1B) is similar to that by OCTN1-expressing cells (417 µM),22 the data suggest that the contribution of 24 ACS Paragon Plus Environment

Page 24 of 50

Page 25 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

OCTN1 to gabapentin transport at the inner BRB is minor. Consistent with this, [3H]gabapentin uptake by TR-iBRB2 cells was not inhibited by the OCTN1 substrates L-carnitine and TEA (Table 2). Data for both b0,+AT1 and LAT1-2 are consistent with that of an obligatory substrate exchanger.34, 35 In this regard, our data showing that [3H]gabapentin efflux from TR-iBRB2 cells was increased by unlabeled gabapentin suggest that gabapentin transport at the BRB is consistent with that of an obligatory substrate exchanger. b0,+AT1 recognizes cationic amino acids, such as L-arginine and L-lysine, as substrates/inhibitors, whereas LAT does not.39-43, 47 [3H]Gabapentin uptake by TR-iBRB2 cells was not reduced in the presence of L-arginine (Table 2) and [3H]gabapentin release from cells was not increased by adding L-arginine (Table 3). In addition, the b0,+AT1 mRNA expression levels were extremely low.38 Based on these data, it is likely that b0,+AT1 does not contribute to gabapentin transport at the inner BRB. Our data also suggest that the cation/amino acid transport systems do not play a role in gabapentin uptake by RPE-J cells. In the human RPE cell line, ARPE-19 cells, mRNA expression of LAT1-2 and protein expression of LAT1 have been reported, both of which are candidates of transporters for gabapentin.15 In contrast, there are no reports of OCTN1 expression in RPE cells. Kaneko et al. detected mRNA expression of b0,+AT in another human RPE cell line, hTERT-RPE cells.48 Since y+LAT1 expression at the outer BRB has been also reported, this transporter could, in addition to LAT, OCTN, and b0,+AT1, be a potential candidate for gabapentin transport at the outer BRB.16 Like b0,+AT1, y+LAT1 accepts cationic and neutral amino acids as a substrate and/or inhibitor. Although the cationic amino acid L-arginine had a weak trans-stimulation effect on [3H]gabapentin release from RPE-J cells (Table 3), [3H]gabapentin uptake by RPE-J 25 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

cells was not altered in the presence of L-arginine (Table 2). Thus, the contribution of b0,+AT1 and y+LAT1 to gabapentin transport at the outer BRB seems to be minor. The Km of gabapentin uptake by RPE-J cells (507 µM) is similar to that of OCTN1-mediated gabapentin transport.22 However, gabapentin uptake by RPE-J cells was independent of changes in Na+, Cl-, the resting membrane potential (Figure 4A), and was only slightly increased by changes in extracellular pH to 6.4 (Figure 4B). Moreover, gabapentin uptake was not reduced in the presence of L-carnitine and TEA, both of which are substrates of OCTN1-2.44-46 Taking the above results into consideration, LAT plays a role in gabapentin transport at the inner and outer BRB. As shown in Table 2, [3H]gabapentin uptake by TR-iBRB2 cells was increased in the presence of L-serine and L-arginine, whereas this uptake by RPE-J cells were not. In contrast, [3H]gabapentin uptake by RPE-J cells was promoted by co-existence of L-asparagine, MeAIB, and TEA, although L-asparagine, MeAIB, and TEA had little effect on [3H]gabapentin uptake by TR-iBRB2 cells. There are no clear reports that transport mechanism(s) including LAT are promoted by these compounds. Thus, further studies to understand the mechanism(s) of promoting gabapentin uptake by these cells should be needed, and could help us to understand the detail of gabapentin transport properties at the BRB. All LAT1-4 (SLC7A5, SLC7A8, SLC43A1, and SLC43A2) recognize several L-type large neutral amino acids and do not transport small neutral amino acids or their derivatives.39-42 As shown in Table 2-3, we observed strong inhibition of [3H]gabapentin uptake by TR-iBRB2 and RPE-J cells as well as trans-stimulation of [3H]gabapentin release from these cells in the presence of L-phenylalanine, L-leucine, and the LAT selective substrate/inhibitor, BCH.39,

42, 47

In addition, [3H]gabapentin uptake by

26 ACS Paragon Plus Environment

Page 26 of 50

Page 27 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

TR-iBRB2 and RPE-J cells was significantly reduced in the presence of L-isoleucine, L-methionine, L-valine and L-cysteine (Table 2). Of the LAT1-4 transporters, LAT2 has been reported not to recognize D-type branched-chain neutral amino acids such as D-leucine.39, 47 In TR-iBRB2 and RPE-J cells, the inhibition of [3H]gabapentin uptake by D-leucine and D-phenylalanine and trans-stimulation of [3H]gabapentin efflux in the presence of D-leucine were observed (Table 2-3). These results suggest minor contribution of LAT2 to gabapentin transport across the inner and outer BRB. As the other LATs, LAT3 and LAT4 have been known. It has been reported that L-tyrosine and L-tryptophan are weak inhibitors of LAT3-mediated substrate transport by less than 50 % at 10 mM41 and LAT4-mediated substrate transport by less than 30 % at 20 mM.42 At a concentration of 1 mM, L-tryptophan and L-tyrosine significantly reduced [3H]gabapentin uptake into TR-iBRB2 and RPE-J cells by more than 58 % (Table 2). This strong inhibitory effect implies that other transport systems except for LAT3-4 play a major role in gabapentin transport at the inner and outer BRB. Regarding the contribution of LAT1 to the gabapentin transport, the Km of LAT1-mediated gabapentin transport has been reported to be 217 µM,19 which is consistent with gabapentin uptake by TR-iBRB2 (735 µM; Figure 1B) and RPE-J cells (507 µM; Figure 3B). In addition, our immunoblot analyses indicate the protein expression of LAT1 at the inner (Figure 5A) and outer BRB (Figure 5B). Taking these lines of evidence into consideration, it is suggested that LAT1 is involved in gabapentin transport across the inner and outer BRB, at least in part. To elucidate the detail of the gabapentin transport at the inner and outer BRB, the expressional and functional analyses using the monolayer prepared from TR-iBRB2 and RPE-J cells could be helpful and needed as further studies. On the other hand, our 27 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

previous study has indicated that the uptake capacity of compounds in TR-iBRB2 cells is correlated with the in vivo blood-to-retina transport activity of the compounds.49 Thus, it is considered that molecular mechanism(s) in the inner BRB play an important role in the retinal transfer of gabapentin. From the results of LAT1 protein expression level and [3H]gabapentin transport studies after the transfection of LAT1-targetted siRNA into TR-iBRB2 cells (Figure 6), the major contribution (~60 %) of LAT1 to gabapentin transport at the inner BRB is suggested. To explain the remaining contribution ratio (40 %), further studies about LAT3-4 expression and function at the inner BRB and LAT3/4-mediated gabapentin transport are needed since the contribution of LAT3 and/or 4 to gabapentin transport at the inner BRB cannot be completely excluded as described above. Nevertheless, because LAT1 contribution ratio of gabapentin transport at the inner BRB is more than 50 %, it is suggested that LAT1 plays an important role in gabapentin transport at the inner BRB. LAT1 is known as an obligatory substrate exchanger. The RUI value of gabapentin means that retinal transfer of gabapentin is 1.43-fold greater than that of n-butanol, which is a highly- and freely-diffusible marker compound (Table 1). Since the in vivo uptake time in this carotid artery injection technique has been settled as 15 sec, the effect of gabapentin binding to the cells in the retina is considered to be low for the great retinal transfer of gabapentin. One possible explanation of the greater transfer of gabapentin compared with that of n-butanol is that gabapentin transport across the inner BRB via LAT1 is actively driven by the endogenous large neutral amino acids in the retinal interstitial fluid and/or retinal capillary endothelial cells. Indeed, this facilitative transfer was reported to be also observed in the case of several large neutral amino acids, such as L-phenylalanine, L-leucine, and L-Dopa.30 28 ACS Paragon Plus Environment

Page 28 of 50

Page 29 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

In conclusion, this study, for the first time, demonstrates carrier-mediated blood-to-retina transport of gabapentin across the BRB in vivo. Our gabapentin transport studies using TR-iBRB2 and RPE-J cells suggest that LAT subfamily is involved in gabapentin transport at both the inner and outer BRB. The gene knockdown study indicates that LAT1 plays a major role in gabapentin transport across the inner BRB. Since it has been reported that gabapentin has protective effects in retinal ganglion cells7 and apoptosis of these cells leads to a progressive loss of vision,8-10 it is possible that the further mechanistic elucidation of LAT1 up-regulation at the inner BRB under the conditions, such as retinal ischemia which have been demonstrated previously,50 could lead to the efficient delivery of gabapentin to the retina. In addition, this fundamental knowledge in this study could apply the mechanistic elucidation about the possibility of alteration of inner BRB-mediated gabapentin transfer under several conditions, such as the retinal diseases and gender difference as further studies. Like this, we expect this finding to increase our understanding of gabapentin pharmacokinetics and that it will have an impact on the systemic application of gabapentin for the treatment of retinal disease.

29 ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Supporting information The Supporting Information is available free of charge on the ACS Publications website, http://pubs.acs.org. This information contains the detail experimental section about in vivo injection into rat carotid artery and culture of TR-iBRB2 and RPE-J cells and in vitro transport analysis, and original immunoblot images used in Figure 5 of this manuscript (PDF).

Acknowledgement We thank Y. Kinoshita and Y. Tanno (University Toyama) for technical advice on the in vivo and in vitro studies, respectively.

Funding sources This research was supported by Japan Society for the Promotion of Science (JSPS) KAKENHI [Grant Numbers JP16H05110 and JP16K08365].

Conflict of Interest The authors declare no conflicts of interest.

30 ACS Paragon Plus Environment

Page 30 of 50

Page 31 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Figure legends Figure 1 Time-, temperature-, and concentration-dependence of [3H]gabapentin uptake by TR-iBRB2 cells A, [3H]Gabapentin uptake (0.1 µCi/200 µL) by TR-iBRB2 cells was performed at 37 °C (open circles) or 4 °C (a closed square) for indicated times. Each point represents the mean ± SEM (n=3). **p