Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Drug transport mechanisms in the rabbit corneal epithelium: In vitro-in vivo correlation
(USC Thesis Other)
Drug transport mechanisms in the rabbit corneal epithelium: In vitro-in vivo correlation
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
DRUG TRANSPORT MECHANISMS IN THE RABBIT CORNEAL EPITHELIUM: IN VITRO-IN VIVO CORRELATION by Joan-En Chang-Lin A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (PHARMACEUTICAL SCIENCES) August 2003 Copyright 2003 Joan-En Chang-Lin Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 3116676 Copyright 2003 by Chang-Lin, Joan-En All rights reserved. INFORMATION TO USERS The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. ® UMI UMI Microform 3116676 Copyright 2004 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 90089-1695 This dissertation, written by J o a n -e n C h a n g -L in under the direction o f h e r dissertation committee, and approved by all its members, has been presented to and accepted by the Director o f Graduate and Professional Programs, in partial fulfillment of the requirements fo r the degree o f DOCTOR OF PHILOSOPHY Vi: Director Date J u n e 1 3 , 2003 Dissertation Committee Chair Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DEDICATION To my loving parents, sisters, the love of my life, Dennis, the mother I never got to know on earth, and our Gracious Heavenly Father Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS It has been a great privilege to be guided into the field of ocular drug delivery research by my untiring advisor, Professor Vincent H.L. Lee. I would like to express my deepest gratitude to him for his excellent guidance, understanding, caring, scientific freedom/stimulation, financial support and his friendship throughout my graduate career at USC. Dr. Robert T. Koda - the first professor I carried out a research project with at USC. I had the privilege of being one of Dr.Koda’s students for a research project in Advance Pharmaceutical Analysis course. It was a privilege and a great honor to have known him. My sincere thanks to my guidance committee for their years of help and support on my thesis project: Dr. Wei-Chiang Shen, my temporary advisor for the first semester at USC, you are like a father to me. Dr. Kwang-Jin Kim, thank you for your critical analysis of my experimental designs and interpretations. Dr. Ian Haworth and Dr. Curtis Okamoto, thank you for your encouragement and helpful suggestions on my thesis project. Thank you all for your time and patience. My graduate life would not have been the same without the friendship and help of my current and previous lab members as well as post-docs and visiting scholars. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I would also like to thank Daisy, Liz, Ruth and Josie for their extra mile of friendship, encouragement and help throughout my graduate life. Thanks to Wade and Linda for the years of administration help especially during the thesis editing process. To my past mentors: I thank Mr. Brent Shenton for his excellent chemistry teaching at Valencia High School that inspired me into science. I would like to thank my undergraduate mentors, Drs. Katherine A. Kantardjieff and Gary Fujii for inspiring me into biophysical research. Their love for research is contagious. To my loving sisters (JH, JJ, JY, JU): it goes without saying how much you mean to me. Thank you for your love, understanding and help especially during my graduate years. I thank my wonderful husband, Dennis, for his love, patience, support and help with everything from taking care of household chores, to staying up with me all night in the lab, to editing my thesis and delivering it back and forth to USC during the thesis editing process. You’re great! Last but not least, I’d like to thank my parents for their unconditional love and support throughout my life. They have never failed to inspire me in every aspect of life. This dissertation is as much the fulfillment of their dream as is mine. iv Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS DEDICATION......................................................................................................... ii ACKNOWLEDGEMENTS................................................................................... H i LIST OF TABLES...................................................................................................xi LIST OF FIGURES.................................................................................................xiii LIST OF SCHEMES...............................................................................................xvi ABBREVIATIONS......................................................... . ........................................xvii ABSTRACT............................................................................................................. xix I. INTRODUCTION .............................................................................................1 A. BARRIERS TO OCULAR DRUG ABSORPTION...................................4 1. Limitations to Topically Applied Ophthalmics................................................ 5 1.1. Precomeal Area and Drug L oss................................................................ 5 1.2. Systemic Absorption .................................................................................. 8 2. Significance of the Comeal R oute..................................................................... 10 2.1. Structure and Function of the Cornea........................................................12 2.2. Resistance to Comeal Drug Penetration....................................................14 B. MEANS TO IMPROVE CORNEAL DRUG ABSORPTION.................. 15 1. Models for Comeal Drug Delivery.................................................................... 16 1.1. In Vivo Animal Models ..............................................................................16 1.2. Ex Vivo Tissue Models ...............................................................................17 1.3. In Vitro Cell Culture M odels.................................................................... 18 2. Solute Transport Across the Comeal Epithelial Cell Layers............................22 2.1. Ion Transport ...............................................................................................30 2.2. Paracellular Transport ................................................................................33 2.3. Passive Transcellular.................................................................................. 34 2.4. Carrier-Mediated Active Transport.......................................................... 35 2.4.1. Amino Acid Transporters................................................................ 36 2.4.2. Monocarboxylate Transporters.........................................................39 V Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.4.3. Nucleoside Transporters....................................................................42 2.4.4. Oligopeptide Transporters.................................................................44 2.4.5. Organic Cation Transporters .............................................................46 2.4.6. Efflux Transporters............................................................................ 48 II. STATEMENT OF THE PROBLEM..............................................................51 A. GOALS AND OBJECTIVE.............................................................................52 B. SPECIFIC AIMS ................................................................................................52 III. MATERIALS AND METHODS................................................................... 58 A. MATERIALS......................................................................................................59 1. Animal M odel........................................................................................................59 2. Cell Culture ...........................................................................................................59 3. Chemicals .............................................................................................................. 61 B. METHODS..........................................................................................................61 1. Preparation of Bicarbonate Ringer’s Solutions .................................................. 61 2. Primary Culture of Rabbit Corneal Epithelial Cell Layers (RCrECL)..............61 2.1. Primary Corneal Epithelial Cell Culture Procedure ................................. 61 2.2. Bioelectric Parameter Measurements.........................................................63 2.3. Transmission and Scanning Electron Microscopy of RCrECL................63 2.4. Immunohistochemical Staining of Cytokeratin.........................................64 2.5. Paracellular Permeability Assessment....................................................... 65 2.6. Active Ion Transport Processes Across RCrECL.......................................66 2.6.1. Ussing-Type Chamber Set U p ..........................................................66 2.6.2. Pharmacological Assessment ...........................................................68 2.6.3. Sodium and Chloride Flux ...............................................................69 3. In Vitro Drug Transport Processes Through the RCrECL ................................ 70 3.1. Radiometry Analysis ....................................................................................70 3.2. HPLC Analysis (P-Blockers) ...................................................................... 71 4. In Vitro Drug Uptake Studies ...............................................................................71 4.1. Gly-Sar U ptake............................................................................................ 71 4.2. Functional and Molecular Evidence of Nucleoside Transporters ........... 74 4.2.1. Uridine Uptake ..................................................................................74 vi Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4.2.2. Concentration Dependent Uridine Uptake .....................................75 4.2.3. RNA Isolation and RT-PCR.............................................................76 5. In Vivo Ocular Absorption ..................................................................................77 5.1. In Vivo Topical Dosing .............................................................................77 5.2. Radiometry Analysis .................................................................................78 5.3. (3-Blockers (HPLC Analysis) ................................................................... 79 C. CALCULATIONS AND STATISTICAL ANALYSIS.............................. 81 IV. RESULTS..........................................................................................................82 A. DEVELOPMENT AND CHARACTERIZATION OF A CORNEAL EPITHELIAL CELL CULTURE MODEL.........................................................83 1. Time Course of Bioelectric Parameters...............................................................83 2. Effect of Culture Conditions on the Bioelectric Parameters ............................. 83 3. Morphology of Cultured Comeal Epithelial Cell Layers...................................86 3.1. Scanning Electron Microscopy (SEM) ..................................................... 86 3.2. Transmission Electron Microscopy (TEM )................................................ 87 4. Immunohistochemical Staining of Comeal Specific Keratin............................ 87 5. Cell Layer Integrity Assessment - Paracellular Permeability........................... 87 B. ACTIVE ION TRANSPORT PROCESSES ACROSS CULTURED RCrECL .................................................................................................................... 91 1. RCrECL Derived From Pigmented Rabbits....................................................... 91 1.1. Bioelectric Parameters of RCrECL Cultured at an Air-Interface (AIC). 91 1.2. Na+ and Cl' Contribution to Total Isc....................................................... 91 1.3. Effect of Pharmacological Agents on Isc..................................................94 1.4. 3 6 C 1 Flux .................................................................................................... 98 1.5. 2 2 Na Flux ................................................................................................... 98 2. Comparison of Bioelectric Data on RCrECL Derived From Albino Rabbits ..99 C. IN VITRO DRUG TRANSPORT PROCESSES........................................... 103 1. Passive Transcellular Transport (|3-Blockers).....................................................103 1.1. Effect of Lipophilicity on Drug Transport Across RCrECL.................... 103 1.2. In Vitro (Culture)-Ex Vivo (Tissue) Correlation for P-Adrenergics I l l 2. Active Transcellular Transport.............................................................................112 vii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.1. Permeability of Model Solutes in AIC and LCC Cultured RCrECL .......112 2.1.1. Mannitol Transport...........................................................................112 2.1.2. L-Arginine Transport....................................................................... 113 2.1.3. L-Camitine Transport...................................................................... 113 2.1.4. Cyclosporine A Transport ............................................................... 114 2.1.5. Guanidine Transport.........................................................................115 2.1.6. Lactic Acid Transport...................................................................... 116 2.1.7. Uridine Transport .............................................................................116 2.2. Evidence for Oligopeptide Transporter in AIC Cultured RCrECL..........117 2.2.1. Effect of pH on Gly-Sar Transport................................................. 117 2.2.2. Time Course of Gly-S ar Uptake..................................................... 121 2.2.3. Directionality....................................................................................121 2.2.4. Energy-Dependence on Gly-Sar Uptake ........................................121 2.2.5. Effect of Extracellular pH on Gly-Sar Uptake ...............................122 2.2.6. Concentration Dependence..............................................................122 2.2.7. Substrate Selectivity.........................................................................122 2.2.8. Differential Recognition of P-Lactam Antibiotics by RCrECL 122 2.3. Evidence for Nucleoside Transporter (s) in AIC Cultured RCrECL .......129 2.3.1. Temperature Dependent and Time Course of Uridine Uptake 129 2.3.2. Inhibition of Uridine Uptake ...........................................................129 2.3.3. Na+ -Dependent Kinetics................................................................... 130 2.3.4. Basolateral Na+ -Independent NBMPR Dose Response .................131 2.3.5. Molecular Identity ........................................................................... 132 2.3.6. Nucleoside Transporter in LCC Cultured RCrECL ...................... 132 2.4. Evidence for an Amino Acid Transporter..................................................141 2.4.1. Time Course of L-Camitine Uptake..................................................141 2.4.2. Concentration Dependency ............................................................... 142 2.4.3. Inhibition of L-Camitine U ptake.......................................................142 D. IN VITRO-IN VIVO CORRELATION..........................................................147 1. In Vitro-In Vivo Correlation ................................................................................147 1.1. Ocular Absorption and Tissue Distribution of P-Adrenergics .................147 1.2. Effect of Lipophilicity on In Vivo Drug Absorption .................................149 1.3. In Vitro-In Vivo Correlation for All P-Adrenergics ..................................152 viii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.4. Ocular Absorption and Tissue Distribution of Paracellular and Carrier- Mediated Substrates..................................................................................... 153 1.5. In Vitro-In Vivo Correlation of All Compounds ....................................... 154 1.6. In Vivo Inhibition of Actively Transported Compounds...........................157 2. Possible Reasons for the Lack of In Vitro-In Vivo Correlation........................ 158 2.1. Culture Age ..................................................................................................158 2.2. Effect of Antibiotics and Antifungal ..........................................................164 V. DISCUSSION.....................................................................................................168 A. DEVELOPMENT AND CHARACTERIZATION OF A CORNEAL EPITHELIAL CELL CULTURE MODEL.........................................................169 1. Primary Culture of Corneal Epithelial Cell Layers............................................. 169 2. Bioelectric Parameters of Corneal Epithelial C ells............................................ 170 3. Comparison of AIC vs. LCC Cultured RCrECL.................................................171 4. Effect of Culture Media Composition on the Bioelectric Parameters...............174 5. Morphology of Cultured Corneal Epithelial Cell Layers...................................176 6. Immunohistochemical Staining of Corneal Specific Keratin..............................177 7. Ion Transport Processes in AIC Cultured RCrECL .......................................... 177 B. IN VITRO DRUG TRANSPORT PROCESSES........................................... 183 1. Effect of Lipophilicity on Drug Transport ..........................................................183 2. Effect of Culture Conditions on Drug Transport.................................................187 2.1. Paracellular Transport................................................................................... 187 2.2. Active Transcellular Transporter................................................................. 191 2.2.1. Amino Acid Transporter..................................................................... 191 2.2.2. Organic Cationic Transporter.............................................................. 193 2.2.3. Monocarboxylic Acid Transporter.....................................................197 2.2.4. Nucleoside Transporter ...................................................................... 198 2.2.5. Oligopeptide Transporter....................................................................206 2.3. Active Transcellular Efflux ......................................................................... 210 C. IN VITRO-IN VIVO CORRELATION...........................................................211 1. In Vitro-In Vivo Correlation Outcome................................................................211 1.1. Passively Absorbed D rugs.......................................................................... 211 ix Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.2. Actively Absorbed D rugs........................................................................... 214 2. Possible Reasons for Lack of Correlation ......................................................... 219 2.1. Culture Age ................................................................................................. 220 2.2. Effect of Antibiotics and Antifungal ......................................................... 223 VI. CONCLUSIONS .............................................................................................. 226 A. SUMMARY OF FINDINGS............................................................................ 227 B. SIGNIFICANCE OF FINDINGS.....................................................................235 C. FUTURE CONSIDERATIONS ...................................................................... 239 1. Cell Culture M odel...................................................................................... 239 2. Corneal Epithelial Transporters.................................................................. 241 3. Normal vs. Disease Condition.................................................................... 243 4. Multiple/Selective Targeting.......................................................................245 VII. REFERENCES ............................................................................................... 247 X Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES Table Page 1-1 Summary of corneal epithelial cell culture models ..................................... 24 1-2 Summary of permeability across the cornea using ex vivo m odel............... 26 1-3 Amino acid transporter systems present in mammalian cells .................... 38 1-4 Monocarboxylic acid systems present in mammalian cells ........................41 1-5 Organic cation transport systems present in mammalian cells.................... 48 3-1 p-Blockers and its physiochemical properties and source ..........................60 3-2 p-Blockers, their log partition coefficient values and HPLC assay conditions ......................................................................................................73 4-1 Effect of culture media composition on the bioelectric parameters............85 4-2 Effect of seeding density on bioelectric parameters..................................... 85 4-3 Effect of AIC condition on bioelectric parameters....................................... 86 4-4 Permeability coefficient of paracellular marker solutes...............................86 4-5 3 6 C1-F1ux across AIC cultured RCrECL (± NPAA)..................................... 100 4-6 2 2 Na+ -Flux across AIC cultured RCrECL (± amiloride) ..............................101 4-7 Effect of Pharmacological agents on Isc across AIC culture RCrECL from albino rabbits .........................................................................102 4-8 Comparison of permeability coefficient for the permeation of P-adrenergic drugs across excised cornea vs. AIC cultured RCrECL ...... 105 4-9 Parameters for sigmoidal relationship between P-adrenergic Pa p p .............. 107 across in vitro cultured RCrECL with its corresponding lipophilicity 4-10 Parameters for sigmoidal relationship between P-adrenergic Pa p p across excised cornea with its corresponding lipophilicity........................ 109 4-11 Comparison of effective half-maximal Pappso for P-adrenergic drugs in various tissue and culture modules................................................ 110 4-12 Summary of Pa p p of various model solutes across AIC and LCC cultured RCrECL........................................................................................... 118 4-13 Effect of pH on transephithelial permeability of gly-sar across AIC cultured RCrECL................................................................................... 119 xi Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4-14 In vivo comeal epithelial absorption of p-adrenergics .................................150 4-15 Parameters for sigmoidal relationship between P-adrenergic in vivo comeal epithelial absorption with its corresponding lipophilicity .............151 4-16 Summary of in vitro Pa p p and in vivo absorption of various paracellular and carrier-mediated compounds............................................. 156 6-1 Components of normal tear and serum ..........................................................242 6-2 Corneal and conjunctival transporters for drug targeting............................ 246 xii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES Figure Page 1-1 Transverse section of the eye and the cornea................................................ 5 1-2 Schematic view of the comeal epithelium with the different mechanisms for transport of compounds across the membrane.................23 1-3 Summary of comeal permeability as a function of molecular weight and partition coefficient.................................................................... 29 1-4 Summary of the active transport processes in the rabbit comeal epithelium.......................................................................................................30 1-5 Summary of lactate production and transport across the cornea.................40 3-1 Comeal epithelial cell isolation methods ......................................................63 3-2 Ussing-type chamber used in the analysis of ion transport processes 68 4-1 Time course of biolelectric parameters .........................................................84 4-2 Scanning electron micrograph of AIC cultured RCrECL ............................88 4-3 Transmission electron micrograph of AIC cultured RCrECL..................... 89 4-4 Immunofluorescence staining for cornea-specific 64-kDa cytokeratin of AIC cultured RCrECL...........................................................90 4-5 Time-courses for Isc inhibition in AIC cultured RCrECL under sodium-free condition....................................................................................92 4-6 Time-courses for Isc inhibition in AIC cultured RCrECL under choloride-free condition ................................................................................93 4-7 Time-course of Isc inhibition of basolateral Na+ /K+ -ATPase activity with ouabain .....................................................................................95 4-8 Time-course of Isc inhibition of basolateral K+ channel activity with BaCL.......................................................................................................95 4-9 Time-course of Isc inhibition of basolateral Na+ (K+ )2C1' -cotransporter activity with bumetanide...................................................... 96 4-10 Time-course of Isc inhibition of apical Cl" channel activity with NPAA ..96 4-11 Time-course of Is c inhibition of apical Na+ channel activity with amiloride..........................................................................................................97 4-12 Relationship between Pa p p across AIC cultured RCrECL of P-adrenergic drugs as a function of log P ......................................................106 xiii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4-13 Relationship between excised tissue Pa p p of P-adrenergic drugs as a function of lo g P ...................................................................................... 108 4-14 In vitro (culture)- ex vivo (tissue) correlation for P-blockers ......................I l l 4-15 Relationship between drug Pa p p and transmembrane electrical conductance for LCC and AIC cultured RCrECL under 4°C and 37°C ...........................120 4-16 Time course of apical and basolateral uptake of gly-sar by AIC culture .. 123 4-17 Effect of pH of apical and basolateral gly-sar uptake by AIC culture........ 124 4-18 Effect of metabolic inhibitors on gly-sar uptake..........................................125 4-19 Concentration dependence of gly-sar uptake ............................................... 126 4-20 Effect of unlabeled amino acids, excess gly-sar, dipeptides and drugs structurally similar to di/tripeptides on gly-sar uptake................................127 4-21 Differential recognition of P-lactam antibiotics by basolateral RCrECL . 128 4-22 Time course of uridine uptake by AIC cultured RCrECL............................ 134 4-23 Inhibition of apical and basolateral uridine uptake by 4°C, unlabeled uridine, Na+ -free condition and NBMPR .....................................................135 4-24 Inhibition of uridine apical uptake by purines and pyrimidines..................136 4-25 Na+ -dependent uridine uptake by AIC culture ............................................. 137 4-26 Biphasic dose-response of NBMPR inhibition on Na+ -independent uridine uptake in basolateral AIC culture.....................................................138 4-27 Inhibition of basolateral Na+ -dependent and Na+ -independent uridine uptake by purines and pyrimidines in AIC cultured RCrECL ................... 139 4-28 RT-PCR analysis of concentrative nucleoside transporter and equilibrative nucleoside transporter ............................................................. 140 4-29 Evidence of Na+ -dependent high affinity nucleoside transporter on the apical membrane of LCC cultured RCrECL .................................... 141 4-30 Time course of apical and basolateral uptake of L-camitine by AIC cultured RCrECL................................................................................... 143 4-31 Concentration dependence of L-camitine uptake by AIC culture.............. 144 4-32 Inhibition of L-camitine uptake by AIC culture...........................................145 4-33 Inhibition of L-camitine uptake by LCC culture..........................................146 4-34 Ocular absorption and tissue distribution of P-adrenergic drugs ................148 4-35 In vivo CE absorption of P-adrenergic drugs as a function of logP ........... 151 xiv Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4-36 In vitro-in vivo correlation for all p-adrenergic drugs..................................152 4-37 Ocular absorption and tissue distribution of paracellular and carrier-mediated substrates.............................................................................155 4-38 Correlation between absorbed fraction in CE and permeability in AIC cultured RCrECL.................................................................................... 157 4-39 In vivo inhibition of mannitol, uridine, L-arginine and CsA absorption ... 159 4-40 The effect of culture age on L-arginine uptake in LCC and AIC cultured RCrECL............................................................................................160 4-41 The effect of culture age on CsA uptake in LCC and AIC cultures............161 4-42 The effect of culture age on L-camitine uptake in LCC and AIC Cultures........................................................................................................... 162 4-43 Effect of culture age on uridine uptake in LCC and AIC cultures ..............163 4-44 Effect of antibiotic and antifungal agents as a function of L-arginine uptake in LCC and AIC cultured RCrECL...................................................165 4-45 The effect of antibiotics and antifungal on CsA uptake in LCC and AIC cultured RCrECL................................................................................... 166 4-46 The effect of antibiotics and antifungal on uridine uptake in LCC and AIC cultured RCrECL................................................................................... 167 5-1 Summary of active transport processes across cultured rabbit comeal epithelial cells.................................................................................................182 5-2 Known regulatory features of nucleoside transporters.................................225 6-1 Summary of the active ion and drug transport processes in RCrECL.........234 6-2 The trilaminar components of the precorneal tear film ................................241 XV Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF SCHEMES Scheme Page 1-1 Precomeal factors that influence ocular bioavailability of topically administered eye drops...................................................................................... 11 2-1 Scheme of specific aim # 1 .............................................................................. 56 2-2 Scheme of specific aim # 2 .............................................................................. 57 3-1 Summary of comeal epithelial cell culture procedure.................................. 62 3-2 Summary of transport procedure across RCrECL.........................................67 3-3 Summary of RCrECL uptake procedure ....................................................... 74 3-4 Summary of in vivo topical dosing procedure...............................................78 xvi Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABBREVIATIONS AcN acetonitrile ACV acyclovir AH aqueous humor AIC air-interfaced condition AB apical-to-basolateral direction BA basolateral-to-apical direction BAC benzalkonium chloride BPE bovine pituitary extract BRS bicarbonate Ringer’s solution BSA bovine serum albumin L-camitine gamma-triethyl-beta-hydroxybutyrobetaine CE comeal epithelium CFR Chloride-free Ringer’s buffer Cj conjunctiva CS comeal-stroma-endothelium CsA cyclosporinA DIDS 4,4’-diisothiocyanotostilbene-2,2’-disulfonic acid DMEM/F12 Dulbecco’s modified eagle mediummutrient mixture F-12 (Ham) 1:1 dimethylsulfoxide 2,4-dinitrophenol ethylenediamine tetraacetic acid equilibrative NBMPR-insensitive nucleoside transporter equilibrative NBMPR-sensitive nucleoside transporter FITC-dextran 4,000 daltons fetal bovine serum fluorescein isothiocynate glycylsarcosine Hank’s balanced salt solution N-2-hydroxyethyl-piperazine-N’-2-ethanesulfonic acid high performance liquid chromatography cidofovir iris-ciliary body equivalent short-circuit current short-circuit current xvii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DMSO DNP EDTA ei es FD4 FBS FITC Gly-sar HBSS HEPES HPLC HPMPC IC IS internal standard K i inhibition constant K m Michaelis-Menten constant LCC liquid-cover condition LogP logarithm of n-octanol/water partition coefficient MCT monocarboxylic acid transporter MeOH methanol MES 2-[N-morpholino]ethanesulfonic acid MW molecular weight N1 sodium purine selective NT N2 sodium pyrimidine selective NT N3 sodium broad selective NT N4 sodium pyrimidine selective but also adenosine and guanosine N5 sodium-NBMPR sensitive-guanosine specific NT NBMPR nitrobenzylthioinosine NPAA N-phenylanthranilic acid NT nucleoside transporters Papp apparent permeability coefficient PC partition coefficient PD potential difference RCrECL rabbit comeal epithelial cell layers RT room temperature Rt retention time SD standard deviation SEM standard error mean SFR sodium-free Ringer’s buffer S-MEM Eagles minimum essential medium for cell suspension TEA triethylamine TEER transepithelial electrical resistance TEM transmission electron micrograph VACV valacyclovir V m ax maximal velocity xviii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT Carrier-mediated drug transport mechanism is relatively unexplored in the comeal epithelium due in part to the lack of a functional in vitro model. The therapeutic potential of targeting endogenous membrane proteins for delivery of hydrophilic drugs to the anterior region of the eye initiated the studies outlined in this dissertation. The overall goal is to characterize the dmg transport mechanisms in rabbit comeal epithelial cell layers (RCrECL) using a functional in vitro model and to determine how well these transport mechanisms correlate with that of in vivo. In particular, there are 2 specific aims outlined in this dissertation. For specific aim 1, an in vitro comeal model for dmg transport and uptake assessments was first established to exhibit in vivo characteristics of the rabbit comeal epithelium that are important for dmg transport evaluations (Pharm Res 17:6, 670-676, 2000). The optimized air-interface culture (AIC) condition led to the growth of RCrECL that closely exhibit the epithelium in vivo based on its bioelectric properties, morphology, permeability and ion transport properties. This is the first report of such tight primary comeal epithelial cell culture model to date. The second specific aim established the existence of apical carrier-mediated transport mechanisms in the rabbit comeal epithelium. Subsequent in vitro-in vivo correlation of 24 model substrates (12 p-blockers and 12 hydrophilic/lipophilic carrier-mediated substrates) revealed a good linear relationship for all compounds except for L- arginine, L-camitine, CsA, and uridine, which resulted in significantly higher in vivo comeal epithelial absorption than would have been predicted by its in vitro xix Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. counterpart. The functional activity of the apical corneal amino acid, nucleoside and the Pgp efflux transporter were affected to some extent by culture conditions, including culture age and the presence of gentamicin, streptomicin-penicillin and amphotericin B. In conclusion, a tight functional corneal epithelial culture model has been established for studying drug transport mechanisms. This is a very important step toward successful drug targeting of hydrophilic therapeutics and to understand the physiological role of these transporters in the corneal epithelium. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I. INTRODUCTION Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A significant challenge in ocular drug delivery is for the topically administered drug to penetrate across the rate-limiting tissue barrier to reach therapeutic target within the eye. Topical route of administration is preferred for many classes of drugs due to ease of administration and patient compliance. This route is mainly applied for treatment of precorneal region infections such as conjunctivitis (pink eye) and blepharitis, or to provide intraocular treatment via the cornea for diseases such as glaucoma, the second leading cause of blindness in the U.S. (Tielsch et al., 1990). Recent biotechnology advances in high throughput screening of active compounds have allowed for the development of new drugs within the field of ophthalmology. Unfortunately, the eye’s various defense mechanisms make it very difficult to achieve an effective concentration of these drugs within the eye and thus, continue to pose a great challenge in the treatment of ocular diseases. Drug permeability across the cornea is clinically important because it is the major factor determining the efficacy of topically applied preparations. Ocular bioavailability of topically administered ophthalmic drugs is less than 10% for lipophilic drugs (Tang-Liu and Weinkam, 1984) and less than 1% for hydrophilic drugs (Hui et al., 1984; Chiang and Schoenwald, 1986). This means that more than 90% of the drug is absorbed into peripheral tissues and into the systemic circulation. This leaves considerable room for improvement. Topical drug delivery is complicated by penetration barriers and effective removal mechanisms of precorneal area, which operate to keep the ocular surface free from foreign substances. The extent of ocular and systemic absorption following topical ocular administration 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. depends mainly on the penetration routes (Scheme 1), their importance relative to one another, and the rate at which they occur (Ahmed and Patton, 1987). The cornea is considered the main pathway for the permeation of drugs into the eye, whereas the noncomeal routes, by way of dmg absorption into the conjunctiva or via nasolacrimal duct, lead to direct systemic circulation, which can cause severe unwanted side effects. The absorption of drugs across the cornea is often very complex due to the many factors affecting the extent and rate of dmg absorption. These factors can be divided into three categories. The first category represents physicochemical factors including pKa, solubility, stability, diffusivity, lipophilicity, and hydrogen bonding capacity, salts, surface area, molecular size and shape. The second category contains dosage form factors such as solution, suspension, emulsion and gel. The third category comprises of physiological factors including metabolism, absorption or transport mechanisms. Much of the attention of ocular drag delivery has been devoted to the first, second and some aspects of third category since the early 1970s. Due in part to the lack of a tight functional in vitro comeal epithelial cell culture model for mechanistic studies, little is known about dmg absorption mechanisms; especially the active transport mechanisms across the comeal epithelium. Dmg transport research is a rapidly developing/expanding area in the pharmaceutical sciences. Today, 30% of clinically marketed drags are targeted to membrane transporters or channels. It is increasingly clear that membrane transporters play a critical role in dmg absorption and disposition. They play a 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. pivotal role in delivering nutrients and cellular detoxification through its capacity to efflux or transport compounds out of the cells. In the recent years, drug absorption across rate limiting physiological barriers such as the GI epithelium or the blood- brain barriers have enjoyed great success through the targeting of membrane transporters to deliver hydrophilic (polar) drugs across such impermeable barriers. Similarly, by targeting membrane transporters on the tight corneal epithelium, one can greatly increase absorption across the tissue barrier and thus, increase ocular bioavailability. Therefore, this thesis aims to develop a functionally tight corneal epithelial cell culture model for elucidating active transport mechanisms in the comeal epithelium. A. BARRIERS TO OCULAR DRUG ABSORPTION The eye is a spherical structure with a wall consisting of two major segments: anterior and posterior segments (Fig. 1-1). The anterior segment of the eye comprises the cornea, conjunctiva, sclera, iris, ciliary body, the anterior and posterior chambers, and the lens. The posterior segment of the eye refers to the parts of the eye that are situated behind the lens and includes the posterior sclera, vitreous, retina, choroids and optic nerve. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1. Limitations to Topically Applied Opthalmics 1.1. Precorneal Area and Drug Loss Drug eliminated from the precorneal area after topically administered eye drops is a net effect of precomeal factors, noncomeal absorption, and comeal absorption as shown in scheme 1-1 (modified from Ahmed and Patton, 1987). The efficiency with which drags are absorbed across the cornea into the anterior region after topical administration is an important determinant of ocular bioavailability. Before reaching the anatomical barrier of the cornea, any drag administered topically has to cross the precomeal barriers of tear factors. These barriers that can decrease ocular bioavailability include drag dilution, metabolism, increased tear turnover leading to an increased spillage and dilution of the drag and reflex blinking causing an accelerated clearance via tears. Furthermore, binding of the drag molecule to the Fig. 1-1: Transverse section of the eye and the cornea (modified from Chaffee EE and Greisheimer EM. Basic Physiology and Anatomy. Philadelphia JB Lippincott, p. 628,1980; Black JM and Matassarin-Jacobs E. Medical-Surgical Nursing 4th Ed. P.832, 1993). Conjunctiva Retina Choroid Bowman’s D escem ent’s Cornea Posterior Sclera Optic Nerve Vitreous body Anterior Chamber Stroma Posterior Cham ber Epithelium Endothelium Ciliary body Rectus muscle 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tear proteins reduces the effective concentration of drug in contact with the cornea and buffering action of the carbonic acid and weak organic acids present in tears affect the extent of ionized/unionized forms of the drug and hence its bioavailability. All these factors will determine the ocular bioavailability of the eye drop. The precorneal tear film is an approximately 10 pm thick fluid layer covering the anterior surface of the cornea and the bulbar conjunctiva. The functions of the tear film are optical, moisturizing, protective and nutritive. After topical administration of an ophthalmic solution, the drug is first mixed with the tear and thus diluted in the eye to approximately 25% of that in the drop (Maurice, 1987). The volume of a drop from a standard medication bottle is approximately 40-50 pL, whereas the volume of tears on the surface of the eye is about 7-9 pL (Mishima et al., 1966), most of which resides in the conjunctival sacs with approximately 1 pL covering the cornea. Though the human eye can hold up to 30 pL without overflowing if care is taken not to blink (Mishima et al., 1966), a large portion of the drug is lost through spillage following tear dilution. For example, the pharmacological activity of a 50 pL drop of 0.5% pilocarpine solution in humans is the same as that of a 20 pL drop due to the ineffectiveness of the excess instilled volume (File and Patton, 1980). In addition, the drug or formulation excipients can stimulate the protective physiological mechanisms to a varying degree depending on the extent of direct drug stimulation or the extent that formulation differs from the precorneal characteristics and thus, decreasing bioavailability of the drug. For example, the normal pH of tears varies between 7.0 and 7.4; however, when the eyes are open, the pH of tear film 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. increases through loss of carbon dioxide. This may explain why the eye can tolerate alkaline eye drops better (Raber and Breslin, 1978; Coles and Jaros, 1984). Irritation is caused primarily by a pH or osmolality far from that of the basal condition of the tears. Solutions instilled into the lower fornix with pH below 6.6 and above 9.0 are associated with irritation, reflex tears and rapid blinking which can result in accelerated tear clearance. Varying precorneal basal condition can also cause damage to the eye. Under basal conditions, the oxygen tension in the tear varies from 140 to 160 mmHg and minimal precorneal oxygen tension of -74 mmHg is needed to avoid corneal edema (Holden et al., 1984). Protein binding of drugs in the tear is another factor affecting ocular bioavailability (Mikkelson et al., 1973). The total protein content of human tear ranges from 0.6% to 2% (average 0.7% in human and 0.5% in rabbit) and is composed of albumin, globulin and lysozyme. Binding of these proteins to the drug decrease the ocular bioavailability of these drugs to the target site. In human and rabbits, topically administered eye drops are usually drained quickly from the nasolachrymal drainage system, aided by the blinking reflex, and the tendency of the precorneal region to return to its normal resident volume of 7-9 pL (Mishima et al., 1966; Sorensen and Jensen, 1979). Drainage of tears and instilled solutions away from the front of the eye is an extremely efficient process; i.e., removal of a 25-50 pL volume of instilled solution in the human is essentially completed at around 90 s (-30 s in rabbit). The nasal cavity, with its greater surface area and higher permeability of the nasal mucosal membrane compared to that of the 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cornea, contributes largely to the elimination of drug from the precorneal area. This drainage rate can be altered by eye irritating factors such as the instilled volume, viscosity, pH, tonicity and drugs as well as by the physiological state of the patient. It is known that increasing in viscosity from 1 to 100 centipoises (cps) causes a decrease in precorneal drainage and hence, increases corneal contact time for the drug. Over this viscosity range, the decrease in the drainage rate constant is ~ 10-fold in rabbits (Chrai and Robinson, 1974). Although mechanisms in the precorneal area decrease ocular bioavailability, there are also processes that can enhance ocular bioavailability. The evaporation process (up to 25% of tear) influences the tonicity of human tears when the eye is open. For the most part, normal eye accounts for this evaporation process by increase tear production or by blinking to rewet the ocular surface. Tear evaporation can also provide benefit to drug absorption by providing less tear for dilution as well as providing a more concentrated drug dose. 1.2. Systemic Absorption The portion of the drug that reaches the nasal mucosa following drainage is absorbed into the systemic circulation, which is contiguous with the conjunctival sac (Chang and Lee, 1987; Desai and Blanchard, 1994). Thus, noncomeal routes (transconjunctival and transnasal absorption) are undesirable for drugs that are targeted for the anterior region. For potent drugs, this can result in serious systemic toxicity. Traditionally, research has been focused on ocular drug absorption and 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. little attention has been placed on systemic absorption. Timolol, used since 1978 for treatment of glaucoma has been found to have adverse systemic side effects that resulted in 450 reported cases, of which 32 resulted in patient deaths (Nelson et al., 1986). Another example is the use of 10% phenylephrine, which though beneficial, has led to the development of cardiac arrhythmias, severe hypertension and myocardial infarction (Lai, 1989). Since then, more warning and emphasis has been placed on systemic exposure of topically applied drugs (Lee et al., 1988; Salminen and Huupponen, 1989; Lee, 1992; Urtti and Salminen, 1993; Urtti, 1994). Several methods of reducing systemic drug absorption have been employed. One of these methods is the nasolacrimal occlusion, which is effective when fingers are pressed over the medial canthal areas for up to 5 min, with (Zimmerman et al., 1984) or without eyelid closure (Ellis et al., 1992). Other methods include increasing viscosity of drug formulation to prolonged ocular contact (Kyyronen and Urtti, 1990), alteration in solution pH and tonicity, adjustment of preservative concentration and coadministration of low doses of vasoconstrictors such as phenylephrine and epinephrine (Lee et al., 1991). Further methods include design of poor systemically absorbed drugs or prodrugs that are rapidly inactivated in the systemic circulation (Chien and Schoenwald, 1986; Chang et al., 1988), chemical modification of the drug to obtain higher comeal penetration and decrease systemic exposure (Chang et al., 1987), and select dosing time that minimizes systemic absorption while maximizing ocular drug absorption (Gregory, 1990; Ohdo et al., 1991). 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2. Significance of the corneal route The comeal route is considered the major pathway by which most drugs enter the eye, despite its tight epithelial barrier (Ahmed and Patton, 1987). However, a minor route, contiguous with the cornea and involving the conjunctiva/sclera (noncomeal route), has been reported (Doane et al., 1978; Ahmed and Patton, 1987). Comeal absorption is a much slower process than elimination. A simplified ocular pharmacokinetic model describing the movement of a topically applied drug to the eye showed that the rate of drug loss is approximately 0.5-0.7/min and rate of comeal absorption is about 0.001/min. The sum of these two rate constants controls the fraction of the applied dose absorbed into the eye (Lee and Robinson, 1986). The ocular bioavailability can thus be increased significantly by decreasing the rate of drug loss or by increasing the rate of comeal absorption. Another way to look at this is to increase the ocular/systemic ratio of the drug (Urtti, 1994). One method to achieve this is to target the rate-limiting barrier, the comeal epithelium, for a facilitated or an enhanced delivery across this barrier. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Increase Ocular Bioavailability Decrease Ocular Bioavailability Tear Evaporation Drug-Protein Binding (reversible) Metabolism PRECORNEAL AREA EYE DROPS CORNEAL ROUTE I Cornea Sclera Drug Dilution in Tear Reflex Blinking (accelerate tear clearance) Tear Turnover pH Difference or Irritation Buffering Action NONCORNEAL ROUTES Conjunctiva Nasolacrimal Duct (Drainage) Anterior Chamber Intraocular Tissues t Ocular Circulation ♦ Contralateral Eye / N Systemic Circulation Precorneal Characteristics in the Human Normal tear volume 7-9 |iL Maximum dosing volume 50 pL T ear turnover 1 pL/min 16%/min Blinking rate 15-20/min Drainage rate constant 1,40/min 50 |iL solutions Evaporation rate 0.1 nL/min Protein content of tear 0.7% Surface tension @ 33°C 44-50 mN/m T ear osmolarity (open eye) 311 -350 mOsmol (closed eye) 288-293 mOsmol T ear viscosity 1.3-5.9 m Pa.s Basal tear oxygen tension 140-160 mmHg T ear pH 7.0-7.4 Scheme 1 -1: Precorneal factors that influence ocular bioavailability of topically administered eye drops (modified from Ahmed and Patton, 1987) and precorneal characteristics in human. 2.1. Structure and Function of the cornea The cornea is the first line of defense on the ocular surface. It is an optically transparent tissue that conveys images to the back of the eye and covers about one- sixth of the total surface area of the eyeball. This avascular tissue obtains nutrients (glucose, vitamins and amino acids) and oxygen from tear fluid and aqueous humor as well as from blood vessels that line the junction between the cornea and sclera. The cornea is approximately 0.5 mm thick in the central region and increasing to approximately 0.7 mm at the periphery. It is composed of the following five layers: (1) the epithelium, providing an impermeable barrier, (2) the Bowman’s membrane, forming a boundary between epithelium and stroma, (3) the stroma, key player in maintaining corneal hydration and transparency, (4) the Descement’s membrane, forming a boundary between stroma and endothelium, and (5) the corneal endothelium, also a key player in corneal hydration (Fig.l). The chief function of the corneal epithelium is to provide barriers between the eye and its environment and between ocular compartments. These barriers maintain the composition of the eye and the differences in composition between ocular compartments. The epithelium consist of 5-6 layers of cells (increasing to 8- 10 layers at the periphery), has a total thickness of around 50-100 pm and a turnover of about one cell layer per day. The tight junctions and hydrophobic domains in this layer make it the most important barrier to drug delivery of hydrophilic drugs. The Bowman’s membrane is an acellular homogenous sheet of about 8-14 pm thick. Its position between the basement membrane of the epithelium and the 12 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. stroma serves as a boundary between epithelium and stroma. The corneal endothelium consists of a single layer of flattened hexagonal cells 5 pm high and 20 pm wide and is responsible for maintaining normal corneal hydration. It is in direct contact with the anterior chamber and is subject to a passive influx of water from the aqueous humor towards the stroma. It was demonstrated in the 1950s that active metabolic processes are necessary for maintenance of normal stromal hydration. It is now well established that both the epithelium and the endothelium act as barriers to the movement of water and ions into the stroma. The greatest resistance to electrolyte diffusion lies primarily on the surface layers of the epithelium. The epithelial cell outer membranes are relatively impermeable to the passage of ions, and the epithelial cells are connected to surrounding cells by tight junctions, which, although not as effective as the cell membranes, also significantly impede ion flow. Comparatively, the endothelium is 200 times more permeable to electrolytes than the epithelium, but it is still 10 times more resistant than the stroma. The resistance to intercellular passage of ions is created by both the elaborate interdigitation of the cell borders, increasing the distance substances must travel, and by localized occlusive cell junctions. The stroma, or substantia propria, accounts for around 90% of the corneal thickness. It contains approximately 85% water, and about 200-250 collagenous lamellae that are superimposed onto one another and run parallel to the surface. The lamellae provide physical strength while permitting optical transparency. The stroma has a relatively open structure and will normally allow the diffusion of 13 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. hydrophilic solutes. The Descemet’s membrane, which is secreted by the endothelium, lies between the stroma and the endothelium. 2.2. Resistance to Corneal Drug Penetration For a drug to penetrate the cornea effectively, it has to have both hydrophilic and lipophilic properties, and be sufficiently small enough to pass through the tight junctions (Grass and Robinson, 1988). This is because the epithelium is cellular and composed largely of lipid membranes. Nonpolar substances penetrate readily but polar or hydrophilic substances penetrate poorly. The stroma, on the other hand, is composed primarily of water, and polar groups pass through it more easily. Many ophthalmic medications are weak bases, which tend to penetrate the cornea well because they exist in equilibrium between a neutral and an ionic form. The neutral form penetrates the epithelium well; however, once in the stroma the equilibrium shifts to favor the ionic form. The ions are able to pass through the stroma to the endothelium, where the reverse process occurs. Most ocular drugs seem to penetrate the cornea by diffusion. The paracellular (through the intercellular space) and the transcellular pathways (through intracellular space) are the most explored mechanisms for drug transport across the cornea. Its barrier function depends favorably on the lipophilicity of molecules and excludes macromolecules (r>10A). The epithelial layers consist of tightly packed desquamatory cells that consist of intercellular tight junctions forming a strong barrier to non-lipophilic/or polar drugs and the differential penetration of non- 14 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ionized drug forms. For lipophilic molecules, the stroma plays a significant role. The comeal stroma is composed of collagen, organized in parallel lamellae. Its highly hydrophilic, porous, and open-knit organization allows hydrophilic dmgs to pass easily and hinders penetration of lipophilic molecules. For macromolecules, the stroma provides a greater barrier than the endothelium (Prausnitz and Noonan, 1998). The comeal endothelium is a single-cell layer, lining the posterior surface of the cornea, in direct contact with the aqueous humor. This cell layer is rich in phospholipids and thus is permeable to lipid-soluble materials and almost impermeable to ions. This layer regulates the passage of substances from the aqueous humor to the stroma and high metabolic activity is associated with the cytoplasm of these cells. B. MEANS TO IMPROVE CORNEAL DRUG ABSORPTION A great deal of preclinical ocular studies has been accomplished by ex vivo and in vivo animal testing. There continues to be a great interest in the development and validation of in vitro methodologies to better understand the physiology of the comeal epithelium so as to give up animal testing or to markedly decrease the number of animals used for some early ADME studies. Pharmacology and especially toxicology have traditionally been the areas where most of the animal testing has been used principally to fulfil the regulatory requirements. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1. Models for Corneal Drug Delivery The absorption of topically administered drugs across the cornea can be studied in different models: in vivo model, in isolated tissue (ex vivo) perfusion/diffusion model and in in vitro cell culture models. Intact corneal tissue models most closely resemble the in vivo condition, but due to their complexity, they do not allow one to distinguish between effects caused by corneal epithelium, stroma or endothelium tissue of the cornea. 1.1. In Vivo Animal Models Much of the earlier ophthalmic testing has been carried out in in vivo animal models with the rabbit being the most popular animal used. This is mainly because rabbit eyes are more sensitive to chemical irritation than that of higher primates or man (Carpenter and Smyth, 1946). In addition, maintenance cost, availability, parasitic disease, lack of genetic uniformity and problems of control and handling make primates and pound animals far less desirable than rabbits as experimental subjects. Moreover, in comparison with the rats and mice, rabbits have a much larger ocular surface. Besides the need to reduce animal suffering, the limitations of in vivo method for providing information at the mechanistic level and the necessity for studying specified tissues without the interference of nearby tissues or vascular response is an additional driving force for exploring ex vivo methods. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.2. Ex Vivo Tissue Models Ex vivo models use native tissues taken from animals or if possible, from human biopsies. Tissues are mounted into suitable diffusion chambers and can subsequently be used for drug transport studies at least for a period of several hours. Much of the earlier ex vivo corneal studies were focused on understanding the ion transport properties or the barrier integrity of the tissue. Transepithelial measurements (electrophysiology) of the cornea were possible using Ussing-type chambers equipped with a voltage-clamp unit, two pairs of calomel and Ag/AgCl electrodes to measure voltage and current, respectively, across the epithelium (Smulders and Wright, 1971). Much of our current understanding of the active ion transport properties of the rabbit cornea using this technique was established by Donn et al. (1959), Green (1965; 1968), Zadunaisky (1966), Maurice (1967) and Klyce (1972). Using this setup, it is possible to obtain continual measurements of bioelectric parameters (transepithelial potential difference (PD) and transepithelial electrical resistance (TEER)) within a few seconds. In addition, clamping the PD of the epithelium at 0 mV (i.e., under short-circuit conditions) allows for measurements of short-circuit current (Isc), an index of active ion transport measured as the ratio of PD to TEER. This method had since been used in determining the effect of drugs on bioelectric parameters and drug permeability across ocular tissues. Another type of ex vivo model for assessing drug permeability is the diffusion chamber (static chamber), a similar device to the Ussing-type chamber. The only difference is that this model uses hydration as a measure of tissue integrity as 17 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. opposed to electrical properties. A water content of 76-80% (hydration, 3.2-4 g water/g dry wt) is considered normal (Schoenwald and Huang, 1983). One disadvantage of these conventional ex vivo chambers is that the drug (esp. with potent or high concentration of drug) of interest is in constant contact with the tissue, which can cause toxic damage to the tissue even before the termination of the study. This led to the development of a perfusion chamber (Richman and Tang-Liu, 1990) that allows for the assessment of drug penetration across the cornea in a way that mimics in vivo condition of a continuous wash out of acute doses. Thus, allowing for determination of comeal contact time, tear turnover rate, formulation contents, comeal metabolic first-pass, and ocular penetration in a more precise and controlled manner. These ex vivo tissue chambers continue to go through modifications for comeal penetration and toxicity assessments (Camber et al., 1987; Thiel et al., 2001). Although these ex vivo models have added great benefits over in vivo model, they still lack the capability of high throughput screening of drug compounds. More importantly, these models impose some difficulties in studying mechanisms at a specific cell layer, particularly the comeal epithelium. 1.3. In Vitro Cell Culture Models Cell culture systems of isolated comeal epithelial cells provide a means to perform mechanistic binding and transport studies. Since the comeal epithelium constitutes the major barrier to ocular bioavailability of polar drugs, the comeal epithelial cell Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. culture model is an important tool for understanding absorption mechanisms across this barrier. There are three major culture types of comeal epithelial cells including the organ/tissue explant, transformed cells (cell line) and the primary cultures. The earliest culture models and the simplest were the organ/tissue explant cultures, which started in the late-1950s. These cultures designed for short-term studies (e.g., cell differentiation, cell vitality assessment, continual growth capability) depend upon cellular proliferation from either whole cornea (McPherson et al., 1956) or stripped comeal epithelium (Stocker et al., 1958; Gilbard et al., 1984). In this method, the cornea (or a stripped off comeal epithelium) is placed into cell growing conditions to allow the epithelium to proliferate and migrate onto a supporting substrate. This method provides information on factors affecting cell growth and wound healing but is limited by contamination with other cell types (e.g. stroma fibroblast, comeal endothelium, etc.), the lack of a uniform multilayer cell growth, and it takes a long time to develop. In 1980s, with improved culture techniques (e.g. culture media, permeable support, etc.), cultured cells were found to grow and differentiate like the in vivo condition. Since then, cell cultures have been applied for in vitro transport/absorption studies involving a multitude of compounds. Cells cultured on a permeable membrane permit the study of both the intracellular (apical-to-basolateral) and extracellular (basolateral-to-apical) transport of dmgs. As it is the case for other cell culture models (e.g. intestine, pulmonary, BBB, etc.), comeal epithelial cell culture models can principally be either cell lines (immortal, passagable cells, 19 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. immortalized by transfection) or primary cell cultures (normal cells, freshly isolated from cornea). Cell lines may lose certain features during passaging, e.g. enzyme activity, expression of membrane proteins or carrier systems. One commercially available rabbit corneal cell line from American Type Culture Collection (Rockville, MD) is the Statens Seruminstitut Rabbit Cornea, SIRC (ATCC CCL 60). In earlier years, SIRC rabbit corneal cell line has been used in a monolayer form as a model for a number of in vitro studies, including alternative assays to the Draize ocular irritation test (North-Root et al., 1982; Jacaruso et al., 1985). This cell line has been criticized because it lacks markers and structures found in epithelial cells, as well as exhibiting fibroblast morphology (Niederkom et al., 1990). Recently, various investigators reported formation of multiple layers of SIRC cells in culture and continued to use them to evaluate ocular irritation (Dicbold et al., 1998), growth pattern (Hutak et al., 1997), permeability (Goskonda et al., 1999; Goskonda et al., 2000; Tak et al., 2001) and carrier-mediated transport mechanism (Balakrishnan et al., 2002). What the investigators fail to mention is that although the cells exhibit multilayers, it does not consist of the three different cell types (superficial, wing and basal cells) that are uniquely arranged to make up the whole corneal epithelium. More importantly, permeability characteristics have not been well established in this cell line and thus, would not be an appropriate model for use in studying transport mechanisms. Primary corneal epithelial cells on the other hand have the ability to exhibit relatively high similarity to the corneal barrier in vivo, as they differentiate into cell 20 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. multilayers of high transepithelial resistance, and display morphological and metabolic properties alike those of the in vivo tissue. As observed from Table 1-2, primary corneal epithelial cell cultures have been evaluated for over 3 decades for various purposes. As cell culture techniques (e.g., culture media, supplements, substrata, etc.) continue to improve, so do the cell culture models. Disadvantages of the primary cell culture system are the laborious isolation procedures, and that the cells do not allow passaging. Primary cell cultures, isolated from native tissues, are often contaminated by other “unwanted” cell types. The corneal epithelial cell culture systems have been used (1) to determine the optimal physicochemical characteristics (e.g., size, charge, lipophilicity, hydrogen bonding potential, conformation) of a drug for passive diffusion; (2) to determine how components of the medium can influence cell growth and wound healing; (3) to elucidate drug transport pathways (mainly passive diffusion). Typically, these corneal epithelial cells are cultured on plastic support submerged in growth medium containing fetal bovine serum (2-20%), fibroblast feeder layers, or growth supplements such as epidermal growth factor (EGF) (10 ng/ml), insulin (5 pg/ml) and cholera toxin (0.1 pg/ml) (Jumblatt and Neufeld, 1983). Aside from the fact that plastic petri dish, by their impermeable nature, is not suitable for transepithelial drug transport studies, epithelial cells grown on a plastic support may not fully express their in vivo ion transport function that may be integral to secondary active drug transport (Steele et al., 1986). 21 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Comeal epithelial cell layers cultured on a permeable support have been attempted on both cell lines (Kahn et al., 1993) and primary cells (Kawazu et al., 1998). The SIRC cell line culture has been morphologically evaluated and has been found to achieve an average maximum cell layer growth of 4.54 cell layers at day 10 (Blein-Sella and Adolphe, 1995). This cell line has also been used for cytotoxicity screening (Blein-Sella and Adolphe, 1995), assessment of cell life span (Kahn et al., 1993), evaluation of basic plasma membrane transport processes (Araie and Maurice, 1987) and testing of antiviral drug efficacy (Rong et al., 1988). Despite the claim of epithelial origin, the SIRC cells exhibit a fibroblast phenotype (Niederkom et al., 1990). Kawazu et al. (Kawazu et al., 1998), on the other hand, used preserved rabbit comeal epithelial cells. The resulting primary cell culture cell layers were, however, extremely leaky, as indicated by a TEER of 144 fi.cm2. This is far below the value of 3,000-8,000 Q.cm2 (Klyce, 1973; Klyce and Wong, 1977) for the excised rabbit comeal tissue. 2. Solute Transport Across the Corneal Epithelial Cell Layers The cornea is the main route of absorption for clinically used ocular drugs (Doane et al., 1978; Maurice and Mishima, 1984; Lee and Robinson, 1986). The concentration of drag in the precomeal area provides the driving force for its transport across the cornea via passive diffusion or active absorption. Thus, efficient ocular drug absorption requires good comeal penetration. Transport across the comeal epithelial membrane is a complex process that includes passive and carrier-mediated 22 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. components for influx and efflux (Fig. 1-2). These processes are not mutually exclusive. While molecular size is the major determinant of whether a solute may traverse the paracellular pathway (pathway 1), lipophilicity may provide indirect information on how likely that pathway is to play a significant role. Drug lipophilicity directly influences the degree of membrane partitioning and hence, passive membrane transport of solutes (pathway 2). Both molecular size and lipophilicity may give insight into substrate-transporter interactions (pathway 3, 4) for determining the predominant pathway. Fig. 1-2: Schematic view of the corneal epithelium with the different mechanisms for transport of compounds across the membrane: (1) paracellular route is a passive, difffusional transport pathway determined by the molecular weight of hydrophilic molecules (e.g. mannitol) which can pass through the tight junctions between adjacent epithelial cells. The transcellular routes include: (2) passive diffusion of lipophilic molecules across cell membranes, (3) carrier-mediated active transport that requires energy expenditure and (4) efflux transporter that also requires energy expenditure. These pathways are not mutually exclusive. Apical nnr > / > Basolateral Active Efflux Paracellular Transcellular Active Carrier-mediated 23 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 1-1: Summary of Corneal Epithelial Cell Culture Models SpeciesT reatment Substratum Media Composition Study Objective Reference Bovine 0.5% trypsin G lass HBSS,0.5% LAH,15% calf serum , pen-strep Virological Investigation Wilcox, 1969 Rabbit n-Heptanol Plastic Method of cell isolation Cintron etal., 1979 Chick embryos 0.1% papain G lass Ham s F10,10% FC S,pen-strep Contact-induce spreading Brown and Middleton, 1981 Rabbit Trypsin Multiwell-plastic Medium 199, 20% FCS, G ent Model for cell attachm ent Friend etal., 1982 Rat Dispase II Plastic plates DM EM -F12,10% FBS, insulin, HC, G ent Morphology a sse ssm e n t Forbes etal., 1984 Chick em bryos 0.1% papain G lass H am s F10,10% FCS, Pen-Strep Morphology of locomotion Brown and Middleton, 1985 Rabbits Dispase II Plastic DMEM-F12,5% CS, 1% AA, pen-strep, ITS Substratum on cell growth Trinkaus-Randall etal., 1988 Bovine Dispase I I Plastic MEM/10% FCS,glutamine, pen-strep, kanamycin, AmB Growth characterization Eggli etal., 1989 Procine & Rabbit Dispase II Plastic MCDB 153, EGF, insulin, HC, BPE Effect of serum -free Hackworth etal., 1990 Rabbit Dispase II Plastic MCDB 151 or 153, ITS, HC, PhE and E, EGF Serum -free clonal growth Kruse and Tseng, 1991 Rabbit Dispase I I Plastic DMEM, 5% FCS, ITS, progesterone, HC, pen-strep, AmB Surfactant cytotoxicity G rant etal., 1992 Human Trypsin/EDTA Plastic DM EM /F12,10% FCS, HC, EGF, CT, insulin, adenine For transplantation Lindberg etal., 1993 Rabbit Dispase/trypsin Plastic DMEM, 5% FCS Cytotoxicity of surfactants Yang and Acosta, 1994 Bovine Dispase II Plastic DMEM/F12, 10% FCS, HC, pen-strep W ound healing T ao etal., 1995 Rabbit Dispase II Plastic DMEM/F12, BSA, insulin, transferrin, HC, CT, adenine Serum free culture Castro-M unozledo et al., 1997 Human D ispase II Plastic K-SFM, pen-strep, EGF, BPE, calcium chloride For toxicity screen Geerfing et al., 2001 DMEM/F12 = Dulbecco's modified Eagles m edium 1:1 with H am s F-12 Gent = gentamicin, AmB = amphotericin B, pen-strep = penicillin and streptom ycin, FC S = fetal calf serum , OF = calf serum , BPE = bovine pituitary extract ITS = insulin-transferrin-selenium mixture, HC = hydrocortisone, CT = cholera toxin, AA = nonessential amino acids Permeability studies of corneal drugs are usually performed in vitro using isolated rabbit corneas mounted in modified Ussing chambers (Schoenwald and Ward, 1978; Huang et al., 1983; Morimoto et al., 1987; Wang et al., 1991). Table 1-3 is a comprehensive list of literature data on permeability across rabbit comeal tissue of compounds with varying molecular weight and lipophilicity. This list of data is shown graphically in Figure 1-3 A and 1-3B. Figure 1-3 A does not appear to indicate that permeability is dependent on molecular weight (or radius). Although there is some scatter in the data, there seems to be no general trend in the effect of molecular weight on permeability. This may be deceiving because no data on macromolecules are included in the graph. That’s probably because cornea provides such a great barrier that compound larger than about 10 A (or molecular weight <500) generally cannot cross at measurable rates (Hamalainen et al., 1997). In contrast, Figure 1-3B shows more dependence of comeal permeability on the octanol-water partition coefficient of the transported compound. Although there is considerable scatter in the data, there is a general trend indicating that permeability increases with increasing partition coefficient. This trend has been seen both from previous experimental studies (Huang et al., 1983; Ahmed et al., 1987; Wang et al., 1991) and from theoretical predictions (Yoshida and Topliss, 1996; Fu and Liang, 2001; Edwards and Prausnitz, 2001; Fu and Liang, 2002) using molecular radius or log distribution coefficient as parameters. Combined, this yields the assertion that comeal permeability is a function of both partition coefficient and molecular size. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 1-2: Summary of Permeability Across the Cornea Using Ex Vivo Model Compound MW Log P Permeability (x 1 C T 6 cm/s)a Reference Acebutolol 336 1.63 1.1 W ang etal., 1991 Acebutolol 336 1.63 0.85 Huang etal., 1983 Acetazolamide 222 -0.26 0.95 Tang-Liu etal., 1994 Acetazolamide 222 -0.26 1.3 Duffel ef al., 1986 Acetazolamide 222 -0.26 0.51 Maren etal., 1983 Alprenolol 249 2.65 29 W ang etal., 1991 Aniline 93 0.92 36 Kishida and Otori, 1980 Atenolol 266 -0.11 1.1 W ang etal., 1991 Atenolol 266 -0.11 0.68 Huang etal., 1983 Benzolamide 320 0.32 0.33 Conroy and Buck, 1992 Benzolamide 320 0.32 0.14 Edelhauser and Maren, 1988 Benzolamide 320 0.32 1.4 (human) Edelhauser and Maren, 1988 Benzolamide 320 0.32 0.14 Maren etal., 1983 Betaxolol 307 2.17 27 W ang etal., 1991 Bevantolol 345 2.65 54 Huang etal., 1983 Bromacetazolam ide 301 -0.02 0.42 Edelhauser and Maren, 1988 Bromacetazolam ide 301 -0.02 3.6 (human) Edelhauser and Maren, 1988 Bromacetazolam ide 301 -0.02 0.36 Jankow ska etal., 1986 Bromacetazolam ide 301 -0.02 0.67 Jankow ska etal., 1986 Bromacetazolam ide 301 -0.02 0.4 Maren etal., 1983 Bromacetazolamide 301 -0.02 2.1 (F) Maren etal., 1983 Bufuralol 261 3.40 72 Huang etal., 1983 Butanol 74 0.82 76 G rass and Robinson, 1988 6-Carxocyfluorescein 376 -3.1 0.46 Toropainen etal., 2001 Chloramphenicol 323 6.8 Schoenwald, 1985 Chlorzolamide 276 1.53 18 Maren etal., 1983 Cimetidine 252 0.35 0.7 Tang-Liu etal., 1994 Clonidine 230 1.37 44 Chiang and Schoenwald, 1986 Clonidine 230 1.37 31 Chiang etal., 1986 Cocaine 303 2.72 6.1 Igarashi etal., 1984 Cortexolone 346 2.62 30 Schoenwald and W ard, 1978 Corynanthine 354 3.01 11 Chiang etal., 1986 Cromolyn 468 1.85 1.1 G rass and Robinson, 1988 Cyclosphosphamide 261 0.80 11 Schoenwald and Housem an, 1982 Cyclosporin 1201 11 Schoenwald, 1990 Deoxycorticosterone 330 3.25 40 Schoenwald and Ward, 1978 2-deoxyglucose 165 -3.12 7.4 Schoenwald, 1990 D exam ethasone 392 1.49 5.0 Schoenwald and W ard, 1978 D exam ethazone acetate 434 2.02 37 Schoenwald and Ward, 1978 2,5-dimethoxyaniline 153 1.25 33 Kishida and Otori, 1980 Edetic acid (EDTA) 292 -4.69 2.1 G rass and Robinson, 1988 Ethanolamine 61 -1.30 0.5 Kishida and Otori, 1980 Ethoxzolamine 258 2.02 44 Eller etal., 1985 Ethoxzolamine 258 2.02 36 (human) Edelhauser and Maren, 1988 Ethoxzolamine 258 2.02 56 Edelhauser and Maren, 1988 Ethoxzolamine 258 2.02 56 Maren etal., 1983 Fluorometholone 376 2.01 17 Schoenwald and W ard, 1978 Flurbiprofen 244 3.75 21 Tang-Liu etal., 1994 Flurbiprofen amide 243 2.79 22 Tang-Liu etal., 1994 Glycerol 92 -2.19 4.5 G rass and Robinson, 1988 Hydrocortisone 362 0.54 8.5 Schoenwald and W ard, 1978 26 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 1-2. Continued. C o m p o u n d MW Log P P erm eability (x 10~6 c m /s)a R e fe re n c e Hydrocortisone 362 0.54 3.5 G rass and Robinson, 1988 Ibufenac 0.303 21.2 Rao etal., 1992 Ibuprofen 0.806 22.4 Rao etal., 1992 Indomethacin 358 4.18 69 Muchtar etal., 1997 Inulin 5000 0.55 Ahmed etal., 1987 Labetalol 328 2.5 14 W ang etal., 1991 Levobunolol 291 2.26 29 Tang-Liu etal., 1994 Levobunolol 291 2.26 23 W ang etal., 1991 Levobunolol 291 2.26 17 Huang etal., 1983 Mannitol 182 -4.67 2.4 G rass and Robinson, 1988 Mannitol 182 -4.67 91 G rass and Robinson, 1988 Mannitol 182 -2.2 0.38 Toropainen etal., 2001 Methazolamide 236 2.6 E delhauser and Maren, 1988 Methazolamide 236 4.9 (human) Edelhauser and Maren, 1988 Methazolamide 236 4.2 Schoenwald, 1990 Methazolamide 236 1.9 Jankow ska etal., 1986 Methazolamide 236 7.3 Jankow ska etal., 1986 Methazolamide 236 2.6 Maren etal., 1983 Methylenedianiline 198 1.61 2.5 Kishida and Otori, 1980 Metoprolol 267 1.20 28 W ang etal., 1991 Metoprolol 267 1.20 24 Huang et al., 1983 Nadolol 309 0.23 1.4 W ang etal., 1991 Nadolol 309 0.23 6.9 Ahmed etal., 1987 Nadolol 309 0.23 1.6 Huang etal., 1983 Nadolol diacetate 393 2.02 4.8 Schoenwald, 1990 Oxprenolol 265 1.69 32 W ang etal., 1991 Oxprenolol 265 1.69 28 Huang etal., 1983 Penbutolol 291 4.04 22 Ahmed etal., 1987 Penbutolol 291 4.04 60 Huang etal., 1983 Phenylephrine 167 -0.72 0.94 Chiang etal., 1986 Pilocarpine 208 0.74 17 Schoenwald, 1985 Pilocarpine 208 0.74 2.8 Suhonen etal., 1991 Pilocarpine 208 0.74 1.2 Siefert and Keipert, 1997 Pindolol 248 1.67 10 W ang et al., 1991 Prednisolone 360 0.72 3.7 Schoenwald and W ard, 1978 Prednisolone 360 0.72 4.5 Tang-Liu etal., 1994 Prednisolone 360 0.72 2.7 Musson etal., 1992 Prednisolone acetate 402 1.26 33 Schoenwald and W ard, 1978 Procaine 236 2.38 4.2 Igarashi etal., 1984 Progesterone 314 3.78 20 Schoenwald and Ward, 1978 Progesterone 314 3.78 18 Schoenwald and W ard, 1978 Propranolol 259 2.75 31 G rass and Robinson, 1988 Propranolol 259 2.75 34 W ang etal., 1991 Propranolol 259 2.75 46 Ahmed etal., 1987 Propranolol 259 2.75 58 Huang etal., 1983 Rauwolfine 314 2.22 9.2 Chiang etal., 1986 Rhodamine 479 2.3 18.1 Toropainen etal., 2001 Sotalol 272 0.23 0.7 W ang etal., 1991 Sotalol 272 0.23 1.0 Huang etal., 1983 Sucrose 342 -3.70 4.3 Ahmed etal., 1987 Sulfacetam ide 214 -0.93 1.9 Schoenwald, 1985 Sulfacetam ide 214 -0.93 1.0 Tang-Liu etal., 1994 Sulfanilamide 172 -0.50 0.5 Kishida and Otori, 1980 27 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 1-2. Continued. C o m p o u n d MW Log P P erm eability (x 10“6 c m /s)a R e fe re n c e Tetracaine 264 3.65 1.5 Igarashi etal., 1984 Testosterone 288 3.22 42 Schoenwald and W ard, 1978 Timolol 316 1.61 12 W ang etal., 1991 Timolol 316 1.61 18 Chang etal., 1987 Timolol 316 1.61 8.0 Ahmed etal., 1987 Timolol 316 1.61 12 Huang etal., 1983 Tobramycin 467 -7.32 0.52 Schoenwald, 1985 Triamcinolone acetonide 4.35 1.60 12 Schoenwald and Ward, 1978 Triamcinolone acetonide 435 1.60 16 G rass and Robinson, 1988 T richlormethazolamide 339 6.5 Conroy and Buck, 1992 T richlormethazolamide 339 10 Jankow ska etal., 1986 T richlormethazolamide 339 24 Jankow ska etal., 1986 T richlormethazolamide 339 11 Maren etal., 1983 T rifluormethazolamide 339 3.9 Maren etal., 1983 Vidarabine 1.7 O’Brien and Edelhauser, 1977 W ater 18 -1.38 150 G rass and Robinson, 1988 Yohimbine 354 2.87 18 Chiang etal., 1986 Alpha-Yohimbine 354 2.92 23 Chiang etal., 1986 n.d. = not determined. aPermeability w as a sse sse d in rabbit corneal tissue unless otherwise noted. Table modified from Prausnitz and Noonan, 1998. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. SI < 0 a > E 1 o a > Q _ f _ O < 0 t < U X c w 1 - o a Si (0 0 ) E V a r o o > c o o 180 140 100 60 20 -20 3600 5600 1600 -400 Molecular Weight 180 n 140 - u E 100 - o ID o X 60 - 20 - -20 -10 5 0 5 Log P Fig. 1 -3: Summary of corneal permeability as a function of (A) molecular weight and (B) octanol-water partition coefficient. Data in these plots are taken from Table 1-3. Corneal permeability appears to be a function of the partition coefficient. 29 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.1. Ion Transport Processes Much of what is known of active solute transport was from ion transport studies since the late 1950s due to its importance in understanding transparency and maintenance of the cornea. Majority of data were obtained in rabbit cornea, active contribution was found to come from a net inward sodium transport from tear to the stroma side (Donn et al., 1959; Green, 1965) and a net outward chloride transport in the reverse direction (Zadunaisky, 1966; Klyce et a l, 1973) as shown in Fig. 1-4. Apical (Tears) Basolateral (Strom a) c i- Na* Lysine C l- c i- Fig. 1-4: Summary of the active transport processes in the rabbit corneal epithelium. (Note: the Na+ channel present on the apical membrane of the corneal epithelium is controversial) Corneal transparency is tied to an inherent tissue hydration (swelling) that is counteracted by active ion transport processes present in the two surface cell layers, the corneal epithelium and endothelium (Zadunaisky, 1966; Zadunaisky and Lande, 1971; Fischer et al., 1978; Hamann, 2002). The corneal epithelium, making up less than 10% of the whole corneal thickness (500-600 pm), consists of several cell layers that function as a tight protective barrier to prevent invasion of foreign substances. In addition, these cell layers, located between the external environment 30 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and comeal stroma, also function as a refractive surface that is essential for visual acuity. Maintenance of functional comeal epithelial cell layers is a dynamic process that involves cell division of the mitotically active basal cells moving toward terminal differentiation, stratification into superficial cells, and senescence, yielding a complete epithelial turnover around 7 days. Active ion transport has been suggested to modulate these cell biological processes. For example, it has been shown in several cell types that K+ channel expression is required for cell division to occur (Deutsch, 1990). In addition, maintenance of intracellular Ca2 + and pH has been reported to be important in rabbit comeal epithelial cell’s ability to respond to growth factors, to initiate differentiation and proliferation and to maintain normal metabolic pathways (Grant and Acosta, 1996). Comeal pathology leads to the loss of barrier function and diminution of the optical properties of the cornea. Thus, characterization of active ion transport mechanisms of comeal epithelial cells is important for a better understanding of comeal homeostasis. Active Na+ absorption in the tear to aqueous direction and Cl' secretion in the opposite direction were suggested in corneas of both albino rabbit (Donn et al., 1959) and frog (Zadunaisky, 1966). Active Na+ and Cl' transport across excised albino rabbit cornea, was confirmed where Is c consists of 50% Na+ absorption (Klyce and Zadunaisky, 1972) and 50% Cl' secretion (Klyce, 1959; Klyce et al., 1972). Frog comeal tissue, on the other hand, exhibits -80-90% Cl' secretion and -10-20% Na+ absorption (Yorio and Bentley, 1978). Transepithelial electrical resistance (TEER), potential difference (PD) (apical negative) and short-circuit current (Isc) 31 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. have been reported in the range of 2-12 kQ.cm2, 5-42 mV and 2.8-14.4 pA/cm2, respectively (Donn et al., 1959; Friedman and Kuper, 1960; Green, 1965; Ehlers and Ehlers, 1968; Klyce et al., 1973; Fischer et al., 1974; Van Der Heyden et al., 1975), for excised rabbit cornea. In addition, net Na flux in the apical-to-basolateral (a->b) direction and net Cl flux in the basolateral-to-apical (b->a) direction have been reported to be in the range of 0.023-0.54 and 0.042-0.057 pEq/hr/cm2, respectively (Donn et al., 1959; Friedman and Kuper, 1960; Green, 1965; Ehlers and Ehlers, 1968; Klyce et al., 1973; Fischer et al., 1974; Van Der Heyden et al., 1975), in excised corneas of many species. These varied values may exemplify the difficulty associated with handling corneal tissues. Factors contributing to this variability may include, inter-animal and/or -species differences, maintenance of tissue viability and integrity, experimental conditions and different analytical techniques used. Rabbits have been utilized for corneal studies mostly due to its established methods of handling and wide availability. Physiological properties of rabbit cornea are similar to those (e.g., the swelling (i.e., hydration) characteristics) of human cornea (Hodson et al., 1981). Ion transport properties of the corneal epithelium have been examined primarily by Ussing-type chamber and patch-clamp techniques. To date, corneal ion transport properties have mainly been assessed in excised whole tissues comprised of the corneal epithelium, stroma (constituting 90% of the tissue thickness) and the endothelial cell layer. The attempt to study corneal epithelial cell layers alone without the interference from the stroma and endothelial cells using excised tissues 32 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. has not been very successful. In addition, other problems inherent in the study of the excised cornea include tissue damage due in part to the tissue curvature which at extreme has been thought to activate and/or deactivate (e.g., due to stretch) various cellular functions (Rae et al., 1990). These problems have been eliminated by the development of primary cultured corneal epithelial cell layers on permeable support (Snapwells™) that can be mounted in Ussing-type chambers for the study of active ion transport properties of corneal epithelial layers alone under well a controlled experimental environment. 2.2. Paracellular Transport The epithelial cells join together by tight junctions which divide the cells into apical and basolateral surfaces, and the structure and composition of these two surfaces are generally quite different in terms of lipid composition and transport proteins. The presence and properties of the tight junctions determine the ability of ions and charged molecules to cross the corneal epithelium between the cells; the paracellular (extracellular) pathway (Fig. 1-2). The ability of solutes to move through this space is limited by the presence of the zonula occludens or tight junction, which is reflected in the electrical resistance of the epithelium. This barrier is known to be very tight to small electrolytes such as Na, K, and Cl with the transepithelial resistance in the 3-8 kO.cm2 range. This barrier is categorized as tight epithelium compared to intestinal epithelium, which is considered to be leaky epithelium. 33 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Passive transport across this barrier depends primarily on the size of the solute as well as the concentration gradients across the epithelium. In the cornea, the paracellular route is preferred by hydrophilic drug for which no specialized transport mechanisms exist. Such markers as mannitol, fluorescein and PEG are commonly used for the assessment of the paracellular route. This pathway involves passive transport of compounds between the epithelial cells and through the tight junctions. It is mainly affected by the size and charge of the compound as well as the tightness of the intercellular space that narrows as the cells undergo differentiation from a columnar to a flattened arrangement. The flattened superficial cells are characterized by numerous tight junctions forming a morphological as well as functional barrier between the apical and the basolateral cell surface domains that result in high transepithelial electrical resistance (or reciprocal of conductivity). It has been reported that this tight junction serves as a barrier not only to the paracellular diffusion of solutes but also to the lateral diffusion of lipid soluble solutes from the apical to the basolateral compartment of the membrane as reported in MDCK cells (Van Meer and Simons, 1986). 2.3. Passive Transcellular Similar to the paracellular, the passive transcellular pathway does not require specialized membrane protein for the transport. The passive transcellular route across the comeal epithelium is governed mainly by the lipophilicity of the drug molecule. Lipophilicity, a measure of drug-membrane partitioning, is inarguably one 34 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of the most important physicochemical parameters in predicting and interpreting membrane permeability. Transcellular pathway is the translocation of compounds across either the apical or basolateral membrane through the cytoplasm to the other side of the membrane. This route includes passive diffusion, endocytosis/ transcytosis and carrier-mediated transport that can be governed by drug properties such as its lipophilicity, its pKa, structural conformation and molecular size. Passive transcellular diffusion is driven by a chemical gradient between both sides of the membrane and is therefore dependent on the concentration gradient of a compound as well as the physicochemical properties determining the passage of the lipid membrane. Internalization of solutes by the endocytic pathway has been extensively studied by evaluating for the transferrin-transferrin receptor (Tf-TfR) complex (Fuller and Simons, 1986; Shah and Shen, 1994). Transferrin receptor has been found to be present on human comeal epithelial cells using immunohistochemical techniques (Lauweryns et al., 1993). Carrier-mediated pathway is an energy dependent process facilitated by a specialized transport membrane protein. Many hydrophilic nutrients as well as toxins can be actively absorbed or secreted via this mechanism. 2.4. Carrier-mediated active transport The attempt to deliver enough drugs into a particular compartment in the body has typically been accomplished by increasing the concentration gradients (dose) and the 35 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lipid solubility of the drug. The obvious disadvantage is that this will also increase the delivery of the drug to all cells and tissues, with the risk of undesirable side effects. Membrane-bound protein transporters facilitate the movement of nutrients (e.g. amino acids, di-and tripeptides, sugars, nucleosides, bile acids, vitamins, etc.) into cells and prevent toxic cellular accumulation of endogenous and especially exogenous agents. These nutrients are either Na+ - or H+ - coupled transporters that may utilize endogenous electrochemical gradient driving forces such as the activity of the Na+ /H+ exchanger or the Na+ -HC0 3~ symport. While passive permeability of the cornea had been extensively studied, very little information is known of the carrier-mediated influx and efflux transport. Therefore, it is important that further studies be aimed at identifying membrane transporters that influence comeal absorption and ways to regulate these transporters for optimized delivery of polar therapeutics across the comeal epithelium. 2.4.1. Amino Acid Transporters Amino acids, especially essential amino acids are required for protein synthesis and as energy sources in all living cells. Since most amino acids are hydrophilic, special membrane proteins are necessary for their transmembrane transport. Amino acid transporters accomplish in parallel with peptide transporters the uptake of amino acids from food in the small intestine, the release into blood, and subsequent uptake of amino acids from the blood into tissues such as liver or skeletal muscle or the reabsorption of amino acids from the urine along the kidney nephron. Building on 36 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the pioneer work of Halvor N. Christensen’s group in the early 1960s, numerous amino acid transporter systems of different types with broad substrate specificity have been identified (Table 1-4). In the central nervous system, amino acid transporters regulate the transport of amino acids across the blood-brain barrier or are involved in the reuptake of neurotransmitter amino acids such as glycine, aspartate, or glutamate from the synaptic cleft and are important for the metabolic coupling of astrocytes and neurons. In the placenta, amino acid transporters supply the fetal blood with nutrients from the maternal side and help detoxify the fetal blood. Other amino acid transporters are involved in basic cellular functions such as cell volume regulation, the synthesis of glutathione (GSH), the provision of amino acids for protein synthesis, and energy metabolism. According to their specialized functions, amino acid transporters fall into different families that are distinguished by the functional properties (specificity of amino acids transported, transport mechanism, coupling to ions) and their molecular similarity or dissimilarity. Many of these amino acid transporters have been first described as transport systems in tissues or cell cultures and have only in recent years been identified on the molecular level. In the eye, amino acids have been implemented as a source of nutrient for the corneal epithelium supplied from the aqueous humor or tear. The first evidence of an amino acid transporter activity on the apical (tear) membrane of the corneal epithelium was the report of a Na+ -dependent L-lysine transport mechanism (Liaw et al., 1991). The functional or molecular identity of this transporter was never 37 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 1-3: Amino acid transporter systems present in mammalian cells. T ra n sp o rte r S u b strate s/In h ib ito rs T issu e E x p re ssio n Cationic A A B0 ,+ Na+ -dependent; broad specificity for zwitterionic and dibasic AA; accepts BCH but not with MeAlB; in blastocytes, Xenopus oocytes and probably also in brush-border m em branes b+ Na+ -independent; cation AA preferring; does not intereact with hom oserine even in presence of sodium; variants known; in m ouse conceptuses y+ Na+ -independent; cationic and zwitterionic AA with sodium; variants known; sensitivity to A/-ethylmaleimide; ubiquitous y+L Na+ -independent; cationic and zwitterionic AA with high affinity only with sodium; insensitive to /V-ethylmaleimide; in erythrocytes and placenta b°.+ Na+ -independent; like B0,+ but limited by positions of branching; not inhibited by BCH; in blastocysts and in brush-border m em branes Anionic AA X 'ag K+ -dependent; reactive with L-Glu and D- and L-Asp; ubiquitous x'c electroneutral; cystine com petes and exchanges with Glu; hepatocytes and fibroblasts Zwitterionic A A A Na+ -dependent; mainly small AA; highly regulated; tolerates an W-methyl group; sensitive to pH changes; frans-inhibition associated; ubiquitous ASC Na+ -dependent; excludes A/-methylated amino acids; frans-stimulation associated; ubiquitous N Na+ -dependent; for Gin, Asn, and His; sensitive to pH changes; so far restricted to hepatocytes; variant Nm in m uscle BETA Na+ and Cf - dependent; p-Ala, taurine and GABA; variants known; ubiquitous GLY Na+ and Cf - dependent; Gly and sarcosine; variants known; present in several tissues IMINO Na+ -dependent; proline, hydroxyprolines, and N-methylated glycines; interacts with MeAlB; in intestinal brush-border m em branes PHE Na+ -dependent; primarily Phe and Met; in brush-border m em branes B° Na+ -dependent; broad specificity for most zwitterionic AA, including branched aromatic ones; accepts BCH but not MeAlB; in brush-border m em branes L Na+ -independent; for bulky side chain AA; frans-stimulated; bicyclic AA a s model substrates; variants described; ubiquitous AA = amino acids determined. In the recent year, functional evidence supported the finding of a B0 ,+ system present on the mucosal membrane of the conjunctiva (Hosoya et al., 1998). It is possible that the originally detected L-lysine transport may be that of the B0 ,+ system in the comeal epithelium but has yet to be determined. 38 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.4.2. Monocarboxylic Acid Transporters Instead of being a waste product, lactate is now regarded as a valuable substrate that significantly contributes to the energy production of heart, non-contracting muscles and even the brain. The "lactate shuttle hypothesis" holds that lactate plays a key role in the distribution of carbohydrate potential energy that occurs among various tissue and cellular compartments such as those between: cytosol and mitochondria, muscle and blood, blood and muscle, active and inactive muscles, white and red muscles, blood and heart, arterial blood and liver, liver and other tissues such as exercising muscle, intestine and portal blood, portal blood and liver, zones of the liver, and skin and blood. Studies on resting and exercising humans indicate that most lactate (75- 80%) is disposed of through oxidation, with much of the remainder converted to glucose and glycogen. Lactate traverses the cell membranes of many tissues, including the heart and skeletal muscle via a facilitated monocarboxylate transport system that functions as a proton symport and is stereoselective for L-lactate. In the recent years, eight monocarboxylate transporters have been cloned (Table 1-5). Monocarboxylate transporters are ubiquitously distributed among many tissues, and the transcripts of several monocarboxylate transporters are present within many of the same tissues. This complicates the identification of their metabolic function. There is also evidence that there is some species specificity, with differences in MCT tissue distributions in hamsters, rats, and humans. MCT1 and MCT3-M/MCT4 are present in rat and human muscles, and MCT1 expression is highly correlated with the 39 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. oxidative capacity of skeletal muscles and also with their capacity to take up lactate from the circulation. MCT1 is also present in the heart and is located on the plasma membrane (in subdomains), T-tubules, and in caveolae. Current evidence is that muscle and other cell membrane lactate transporters are abundant with characteristics of high Km and Vm ax . Lactate is known as the end product of anaerobic glycolysis, a pathway that is of key importance in the comeal epithelium due to relatively few mitochondria (Kinoshita, 1962; Thoft and Friend, 1971; Maurice, 1994) present in the epithelium to limit light scatter and thus enhance transparency. Therefore, the extent of ATP production via oxidative metabolism is limited and the glycolytic pathway fulfills up to 85% of its energetic need (Maurice, 1984; Riley, 1969a) resulting in accumulation Tears Stroma Aqueous Humor Epithelium Endothelium AAA Die Y ' -4 -( g l u t i) - > - Glucose c Anaerobic Pyruvate < 0 Lactate Aerobic Fig. 1-5: Summary of lactate production and transport across the cornea. 40 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of lactate in the comeal epithelium. In 1990, Bonanno proposed the presence of a lactate-proton cotransporter on the stromal or basolateral membrane of the comeal epithelium for the transport of lactate out the epithelium (Bonanno, 1990). Later, Giasson and Bonanno reported findings of apical and basolateral facilitated transport of lactate by rabbit comeal endothelium (1994). Anaerobic glycolysis produces pyruvate and lactate, which, under aerobic conditions, can then be converted to carbon dioxide by the TCA cycle. Since lactate cannot pass into the tear film via the tight comeal epithelium (Riley, 1969a), it diffuses through the comeal stroma (Riley, Table 1-4: Monocarboxylic acid systems present in mammalian cells. T ra n sp o rte r T is s u e e x p re ssio n S u b strate s/in h ib ito rs MCT1 Intestine, ovary, testis, prostate, Benzoic acid, pyruvate, lactate, oxam ate, proprionate, thymus, spleen, kidney, muscle, salicylate, mersalyl acid, formate, quercitin, oxobutyrate, liver, placenta, brain, heart bicarbonate, glyoxylate, glycolate, phloretin, niflumic acid, a-cyanocinnam ate, and derivatives, acetate, 3-isobutyl-1- methylxanthine, 4,4’-substituted stilbene-2,2’-disulfonates MCT2 Testis, liver, brain, heart, muscle, Kidney, skin Lactate, pyruvate MCT3 Retinal pigm ented epithelium Lactate MCT4 Leucocyte, intestine, ovary, testis, Prostate, thymus, spleen, skeletal Muscle, lung, placenta, heart Lactate MCT5 Intestine, ovary, testis, prostate, Kidney, muscle, liver, placenta, Heart Not known; assum ed to be monocarboxylic acids MCT6 Leucocyte, intestine, testis, lung, Kidney, placenta, heart, muscle Not known; assum ed to be monocarboxylic acids MCT7 Leucocyte, intestine, ovary, lung, Placenta, pancreas, muscle, heart, Brain, testis Not known; assum ed to be monocarboxylic acids MCT8 Intestine, ovary, testis, prostate, thymus, pancreas, liver, kidney, muscle, placenta, brain, heart Not known; assum ed to be monocarboxylic acids MCT = monocarboxylic acid transporter 41 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1969b; Klyce, 1981) and crosses the endothelium into the aqueous humor (Giasson and Bonanno, 1994) (Fig. 1-5). As a result of hypoxia due to contact lens wear, oxygen diffusion through the tears from the atmosphere is limited and thus glucose demand and lactate production are increased. It is conceivable that the corneal epithelium may accommodate this increase in lactate by increasing the MCT transporter on the corneal epithelium. 2.4.3. Nucleoside Transporters Concentrative nucleoside transporters (CNT) are limited to specialized cells such as intestine (Belt et al., 1993; Chandrasena et al., 1997), epididymis (Leung et al., 2001), liver (Mercader et al., 1996; Felipe et al., 1998) and kidney (Plagemann and Aran, 1990) epithelia, and several mammalian cell lines (Belt et al., 1993; Crawford et al., 1998), whereas the housekeeping equilibrative nucleoside transporters (ENT), specifically the es system, are ubiquitous and function bidirectionally in the transmembrane flux of nucleosides in accordance with the concentration gradient. Nucleosides and nucleotides play important roles in the metabolism and regulation of physiological responses (e.g. neuromodulation, vasodilation, cardiac contractility, lipolysis, etc.). Since these molecules are polar in nature, their cellular homeostasis depends critically on specific transport proteins in the plasma membrane. The transport of nucleosides across membranes may be important for cells that lack or have low capacity for de novo nucleotide biosynthesis, to undergo nucleotide biosynthesis by the salvage pathways. There are two major classes of 42 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. nucleoside transport systems in mammalian cells, the Na+ -dependent and Na+ - independent. The concentrative Na+ -dependent nucleoside transporters (CNT) are able to transport/concentrate nucleosides against a concentration gradient. There are five subtypes of CNTs based on substrate selectivity. N1 (SPNT or CNT2) is purine nucleoside selective but also accepts uridine whereas N2 (CNT1) is pyrimidine nucleoside selective but also accepts adenosine. These CNT subtypes have been cloned from rat, pig and human. Recently, the only known rabbit CNT, termed rbSPNTl, has been cloned from rabbit intestine (Gerstin et al., 2000). N3 (CNT3) has broad selectivity for both purine and pyrimidine nucleosides which has a characteristic Na+:nucleoside coupling ratio of 2:1. N4 (CNT4) is pyrimidine nucleoside selective, but it also transports adenosine and guanosine. N5 (CNT5) is NBMPR sensitive and appears to be guanosine specific. Both CNT2 and CNT1 systems have been cloned in rat, mouse, and human. The equilibrative nucleoside transport (ENT) systems mediate the bi directional/downhill flux of nucleosides and are characterized by their sensitivity to nitrobenzylthioinosine (NBMPR). There are two subtypes of ENTs. The equilibrative-sensitive transporter (es) termed ENT1 is potently inhibited by a nanomolar concentration of NBMPR. In contrast, the equilibrative-insensitive transporter (ei) termed ENT2 is resistant to NBMPR up to 1 pM. Both the ENT1 and ENT2 transporters have been cloned from rat, mouse, and human. Only rbENT2 has recently been cloned from rabbit tracheal epithelium (Accession # AF323951). 43 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Although these carriers are found in most cells, their transport characteristics vary in different cell types, especially in cells, which constitute epithelial barriers such as the corneal epithelium; these differences may be the consequence of different metabolic specifities of the tissues behind these barriers. The knowledge of nucleoside transport and metabolism in ocular barrier is now made even more important with the increasing use of nucleoside analogues in the treatment of ocular malignancies. It is currently unknown if nucleoside transporter(s) are present in the comeal epithelium. 2.4.4. Oligopeptide Transporters Oligopeptide transport in the mammalian small intestine plays an important role in the absorption of peptide from dietary proteins. The kidney plays an important role in the metabolism of circulating small peptides, which are hydrolyzed in the lumen of the proximal tubule and are subsequently reabsorbed in the form of their constituent amino acids. These exogenous, as well as endogenous di- and tripeptides play important roles in the overall homeostasis of the tissue or organ. Endogenous dipeptides such as camosine are found in muscle, nerve tissues (Margolis, 1980), brain (Marchis et al., 1997; Kohen et al., 1988) and retina (Margolis and Grillo, 1984; Panzanelli et al., 1997). Such peptides have been shown to protect cell membrane from oxidative damage (Babizhayev, 1996; Boldyrev et al., 1995; Babizhayev et al., 1994; Chan et al., 1994). A 5% camosine eye solution have been approved for medical use in Russia (Maichuk et al., 1997) which have been found to 44 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. exert a good therapeutic effect in comeal erosion, trophic keratitis, postherpetic epitheliopathy, primary and secondary cornea dystrophy and bullous keratopathy (Maichuk et al., 1997; Egorova et al., 1989). Used in a combined treatment, the eye drops accelerated healing of comeal ulcers in herpesvirus and bacterial infection or dry keratoconjunctivitis. The mechanisms in which this and other similar dipeptides induce these therapeutic effects are unclear. Thus far, at least four carrier-mediated, proton-coupled peptide transporters have been cloned. The first, PepTl, was cloned from the rabbit small intestine in 1994 (Fei et al., 1994). This transporter is also found in the kidney, the liver, and the brain. Its human homolog, hPepTl, shows an 81% amino acid sequence identity and a 92% conserved sequence with rabbit PepTl (Liang et al., 1995). In addition to the tissues mentioned hPepTl mRNA is found in the placenta and the pancreas. Based on similarity with the intestinal peptide transporters, the renal clones were designated as the PepT2 series. Hydropathy analysis predicted the gene products to contain 12 membrane-spanning domains and a large extracellular loop. PepT2 in the rabbit was also found in the lung (Meredith and Boyd, 1995; Yamashita et al., 1996), the heart, the brain, and the mammary glands but not found in the intestine (Liu et al., 1995). Compared with the intestinal isoforms of the same species, -50% identity and 70% homology in amino acid sequences were found with highest levels of identity in the transmembrane-spanning regions and lowest homology in the large extracellullar loop (Boll et al., 1996). Thus far, PepTl and PepT2 (scheme 1-3) have received far more 45 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. attention than HPT-1, which has been found in the brain and the retina (cloned from Caco-2 cells) (Liang et al., 1995; Dantzig et al., 1994). 2.4.5. Organic Cation Transporters Two major families of organic cation transporters have been identified: organic cation transporters (OCT) and organic cation/carnitine transporters (OCTN). Multiple members of each family have been identified as listed in Table 1-6. Each family was originally described on the basis of its function in the kidney and liver, the major sites for organic cation elimination. Recently, there is accumulating evidence for a role of each in the brain (Koepsell, 1998). Organic molecules with a transient or permanent positive net charge are categorized as organic cations. They include cationic drugs such as anticholinergics, adrenergic, antineoplastic, sympathomimetic, antihistamine, xenobiotics, some vitamins, and a variety of endogenous amines such as choline, epinephrine, dopamine and guanidine. The homeostasis of such organic cations is determined by their endogenous production, their absorption into the cell. Organic cations have many important biological functions. They are transported into cells where they act as essential nutrients or have other influences on cell metabolism. Transport of organic cations is mediated by substrate-specific, sodium-dependent transporters and by less specific sodium-independent transporters. Sodium-independent cation transport systems have been described in liver, kidney, small intestine (Iseki et al., 1993; Ruifrok, 1981), and brain. Transporters in small intestine (Tumheim and 46 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Lauterbach, 1977; Tumheim and Lauterbach, 1980), liver (Elferink et al., 1995; Meijer et al., 1997) and kidney (Roch-Ramel et al., 1992; Ullrich, 1994) mediate the absorption and excretion of organic cations. These functions are mediated by two transport processes in series: an electrogenic facilitated diffusion step via organic cation transporters (OCTs) at the basolateral membrane and an organic cation- proteon exchange mechanism at the apical membrane that may be mediated by OCTN2 (Pritchard and Miller, 1993; Zhang et al., 1998a; Dresser et al., 2001). In recent years, significant progress has been made in understanding the molecular and cellular characteristics of the OCT and OCTN groups. Presently, three members of the OCT group, OCT1, OCT2, and OCT3 (Zhang et al., 1998a; Koepsell et al., 1999; Burckhardt and Wolff, 2000; Dresser et al., 2001), as well as three members of the OCTN group, OCTN1, OCTN2 and OCTN3 (Tamai et al., 1997; Wu et al., 1998) have been cloned from various species and functionally characterized in heterologous expression systems. The existence of organic cation transporter present on the comeal epithelium is currently unknown. L-camitine has been reported to be absorbed into the cornea epithelium, traverse across the endothelium to reach aqueous humor after topical administration (Pessotto et al., 1994). In addition, a carrier-mediated organic cation drug transport activity was recently reported to be present in the rabbit conjunctival epithelium (Ueda et al., 2000). An inside-negative membrane potential-dependent type of organic cation transporter was identified on the mucosal membrane of the conjunctival epithelium using guanidine as a model substrate but the molecular 47 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. identity has yet to be determined. Although, OCT and OCTN are the two main types of organic cation transporters known to date, it is also conceivable that there may be other classes of organic cation transporters that are more tissue specific and have yet to be identified. For example, a novel cationic drug transporter in human retinal pigmented epithelial cells was determined to be functionally distinct from the known members of the OCT or OCTN family of organic cation transporters (Han et al., 2001). Table 1-5: Organic cation transport systems present in mammalian cells. Transporter Tissue expression Substrates/inhibitors OCT1 Liver, kidney, intestine, heart, Vecuronium, 2-chloroadenosine, 2 ’-deoxytubercidin, skeletal muscle, brain decynium-22, dopamine, norepinephrine, am antadine, taurocholate, choline, TEA, 1-methyl-4-phenyl pyridum ion N-methylnicotinamide, zidovudine, serotonin (5- hydroxytryptamine) OCT2 Kidney, brain Nicotine, procainamide, quinine, quinidine, dopamine, norepinephrine, am antadine, TEA, MPP, cimetidine, A/-methylnicotinamide, serotonin OCT3 Placenta, intestine, heart, lung, Choline, dimethylamiloride, desipram ine, clonidine, brain, kidney 1 -methyl-4-phenyl-1,2,3,6-tetrahydropyridine, TEA, MPP, cimetidine, /V-methylnicotinamide, guanidine, serotonin OCTN1 Placenta, brain, heart, uterus, fetal mepyramine (pyrilamine), ofloxacin, levofloxacin, TEA, liver, kidney, trachea, spleen, lung, cefaloridine, L-carnitine, cimetidine, procainamide, bone marrow, skeletal muscle, spinal quinidine, quinine, verapamil, nicotine, clonidine cord, pancreas OCTN2 Kidney, trachea, spleen, placenta, Glycine betaine, butyrobetaine, acetylcarnitine, TEA, bone marrow, skeletal muscle, L-carnitine, guanidine heart, brain OCTN3 T estes, kidney L-carnitine OCT = organic cation transporter OCTN = organic cation/carnitine transporter TEA = tetraethylammonium 2.4.6. Efflux Transporters P-glycoprotein typically pumps from the apical membrane, whereas MRP efflux is from the basolateral membrane. To date, at least six MRP homologues have been 48 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. identified in human with MRP1 and MRP5 being ubiquitous with MRP4 in prostate, lung, muscle, pancreas, testis, ovary, bladder and gallbladder (Borst et al., 2000). MRP2, MRP3 and MRP 6 (mainly liver and kidney) can be found mainly in liver, kidney and intestine (Borst et al., 2000). P-gp substrates include the anticancer drugs (vinca alkaloids, anthracyclines, epipodophyllotoxins, paclitaxel, actinomycin D, topotecan, mithramycin, mitomycin C, etc.), other cytotoxic agents (colchicine, emetine, ethidium bromide, puromycin, etc.), cyclic and linear peptides (gramicidin D, valinomycin, etc.), HIV protease inhibitors (ritonavir, indinavir, saquinavir, etc.) and other compounds such as Hoechst 33342, Rhodamine 123 and Calcein-AM. Similarly, MRP substrates include anticancer drugs such as daunorubicin, doxorubicin, etoposide and vincristine. It appears that P-gp prefers neutral and/or positive charged lipophilic molecules (Zamora et al., 1988), whereas MRP has preference for organic anions (Broxterman et al., 1995) such as leukotriene C4 (LTC4) (Leier et al., 1994), dinitrophenyl glutathione (Paul et al., 1996), and aflatoxin B1 exo-8,9-epoxide-GSH (Cole and Deeley, 1998). The physiological role of P-gp and MRP have received much attention in recent years and plays important roles such as: 1) a gate keeper of preventing entry of toxic compounds into the body, 2) to remove toxic compounds from the circulation once they have entered, 3) to keep toxic compound from leaving the circulation into tissues that are especially sensitive to their toxicities, and 4) at a cellular level to protect cells if toxic compounds have left the circulation to enter the interstitial space. 49 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The role of efflux transporters in the eye is unclear, but one can safely speculate that the physiological function of these transporters may act to prevent and/or detoxify toxic compounds. P-gp has been found in the iris (Holash and Stewart, 1993; Schlingemann et al., 1998), retina (Holash and Stewart, 1993; Esser et al., 1998; Kennedy and Mangini, 2002), capillary endothelium (blood-aqueous barrier) (Kajikawa et al., 1999), ciliary epithelium (Wu et al., 1996; Wang et al., 1998), conjunctiva (Saha et al., 1998; Yang et al., 2000), cornea (Kawazu et al., 1999) and lens (Merriman-Smith et al., 2002). In addition, based on immunohistochemistry results, P-gp has also been found in 12 ocular melanomas intrinsically present before exposure of drugs (McNamara et al., 1996). To date, MRP activity/expression in the eye has been relatively unexplored. The broad spectrum and the ubiquitousness of P-gp pump indicate the possible importance of MDR in modulating chemoresistance in ocular melanoma or act as modulators of chloride channels that regulate cell volume (Wu et al., 1996; Idriss et al., 2000). Modem pharamcology aims to design drugs that can easily partition across the epithelial membrane, usually because of hydrophobic properties; many of these newly created drugs, if positively charged or neutral, will be substrates for detection and extrusion by P-gp or MRP (anionic drugs with similar properties). Thus, knowledge of whether a newly designed drug is a substrate for P-gp will be important in determining the likelihood of drug absorption, the pharmacokinetics of uptake and excretion, and the penetrance of the drug into the eye. 50 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. II. STATEMENT OF PROBLEM Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1. Goals and Objectives The cornea is a major pathway for the penetration of topically applied ocular drugs in treating anterior chamber related diseases such as glaucoma and interstitial keratitis. The corneal epithelial cells play a rate-limiting role in the overall corneal transport of hydrophilic to moderately lipophilic drugs. The relatively unexplored active transporters in corneal epithelial cells can offer a great opportunity as an endogenous drug carrier system for the delivery of hydrophilic agents to the anterior segment of the eye. This research is based on the central hypothesis that there are specialized carrier-mediated transport processes on the apical and basolateral corneal epithelium that mediate the influx/efflux of extracellular physiological substrates. These carrier- mediated transport processes can be targets for delivering polar drugs across the tight corneal epithelium. The long-term goal of this project was to develop a functionally tight rabbit corneal epithelial cell culture model to study corneal epithelial drug transport processes and to determine how well these processes correlate with in vivo absorption mechanisms. The following two specific aims were carried out to test our hypothesis and to achieve our goals: 2. Specific aims: 2.1. Specific aim # 1: To develop and characterize a rabbit corneal epithelial cell culture model for drug transport studies. Corneal drug transport studies are typically performed in isolated tissues mounted on 52 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. modified Ussing chambers (Ashton et al., 1991). Probably due to the low number of in vitro cell culture systems that successfully mimic this barrier as well as technical difficulties with isolating and maintaining an intact epithelial sheet, the role of the corneal epithelia in overall drug transport is rarely studied directly, if at all. Rather, this is inferred, by comparing drug flux in tissues stripped of its epithelial layers against that afforded by the intact tissue (Ashton et al., 1991). A cell-based model offers the potential to account for transcellular and paracellular drug diffusional processes, metabolism and active transport processes. The technical difficulties with isolating and maintaining an intact epithelial tissue have greatly limited drug transport studies in the cornea epithelium. Thus, the advantages of a cell culture model system include: (1) high sample throughput; (2) ability to access both the apical and basolateral cell surface directly in mechanistic transport studies; and (3) lack of possible confounding interference from non- epithelial cells. To date, there exists no functionally tight comeal epithelial cell culture model. There has been several immortalized epithelial cell line models proposed for use in ocular toxicity studies. These models include the SIRC cell lines of rabbit origin, which have a fibroblast morphology (4) and the MDCK line of canine origin, which is derived from the kidney. However, neither of these cell lines provide specie and tissue specificity. In addition, immortalized rabbit and human comeal epithelial cultures have been extensively developed using the SV40 vims or a recombinant SV40-adenovirus vector. All aforementioned cell lines have been used for assessing ocular irritation (Blein-Sella and Adolphe, 1995; Kruszewski et 53 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. al, 1997), assessment of cell life expansion (Kahn et al., 1993), and cell differentiation (Castro-Munozledo, 1994; Beebe and Masters, 1996). Whether or not these immortalized epithelial cells retained bioelectric properties or transport properties consistent with in vivo comeal epithelia has not been critically evaluated. These cell cultures may or may not fully express their in vivo ion transport function that are integral to secondary active drag transport (e.g. Na+ /H+ exchanger and Na+ /K+ ATPase). Recently, Kawazu et al. (1998) reported on a primary rabbit corneal epithelial cell culture system. However, their culture system suffered from the drawback of leakiness with a transepithelial electrical resistance of 144 ohm.cm2 and reported no potential difference (PD). Culturing cells under the air-interface condition (AIC) has been attempted for conjunctiva (Yang et al., 2000), airway (Robison and Kim, 1994), and epidermis (Kennedy et al., 1996). In the case of airway epithelial cells, culturing under the AIC condition resulted in enhanced morphological features, polarity, and ion transport characteristics comparable to those of the native tissue (Johnson et al., 1993). This was attributed to improved oxidative metabolism made possible by improved access of oxygen to the epithelial cells when cultured in the air-interface configuration (Kondo et al., 1997). Thus, the purpose of specific aim # 1 (scheme 2- 1) was to identify the growth conditions that would yield a functional primary culture of pigmented rabbit comeal epithelial cells that exhibits similar morphological characteristics, bioelectric parameters, paracellular permeability and ion transport properties to those observed in the intact tissue. 54 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.2. Specific aim # 2: To investigate in vitro drug transport processes and to determine how well these processes correlate with in vivo transport mechanisms. The second objective of this dissertation aims at elucidating the presence of dipeptide, monocarboxylic acid, nucleoside, organic cation transporters and Pgp efflux pump in the corneal epithelium by functional characteristic and molecular identity (if known in rabbit). These transporters have been demonstrated to be involved in the transport of substrates, which are specific to each transporter, into the epithelial cells of tissues such as intestine and kidney. This provides a particularly useful targeting approach for delivering hydrophilic drugs across the restrictive epithelial barrier, the corneal epithelium. The existence of these transporters in the corneal epithelium is relatively unknown. Therefore, the aim of this study was to establish functional and molecular evidence for active carrier-mediated transport processes in the corneal epithelial cell culture model and to determine how well this correlate with in vivo absorption (Scheme 2-2). Gly-sar, lactic acid, uridine, guanidine and cyclosporin A were used as model substrates for dipeptide, monocarboxylic acid, nucleoside, organic cation transporters and Pgp efflux pump, respectively. These representative substrates have commonly been used due to their broad spectrum specificity and they are metabolized to a lesser extent compared to other physiological substrates in its class. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f the copyright ow ner. Further reproduction prohibited without p erm ission. Specific Aim # 1 Corneal Epithelial Cell Culture Development and Characterization Cellular Integrity Culture Conditions I AIC vs. LCC I Culture Media Components Morphology Ion Transport Properties 1 Bioelectric Properties 1 1 Paracellular Permeability Immunohistochemistry 1 Scanning Electron Microscopy Transmission Electron Microscopy Na+ and Cl- Contribution I Na+ -channel Cl' -channel Na+ /K+ - ATPase K+ -channel Na+ (K+ )2C1' cotransporter Scheme 2-1: Scheme of specific aim # 1 U\ ON Reproduced with permission o f the copyright ow ner. Further reproduction prohibited without perm issio n . Specific Aim # 2 Elucidation of Drug Transport Mechanisms: In Vitro - In Vivo Correlation / In Vitro Transport AIC vs. LCC 1 Bidirectional Transport In Vitro Transport In Vivo Absorption 1 1 Passive Paracellular Passive Transcellular Active Transcellular Passive Paracellular Passive Transcellular Active Transcellular Amino Acid Transporter Dipeptide Transporter Monocarboxylic Acid Transporter Nucleoside Transporter Pgp Efflux In Vitro-In Vivo Correlation \ Scheme 2-2: Scheme of specific aim # 2 1. Correlation 2. Lack of Correlation Culture Age Effect of Culture Conditions III. MATERIALS AND METHODS Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A. MATERIALS 1. Animal Model Male, Dutch-belted pigmented and New Zealand albino rabbits, weighing -2.5 kg, were purchased from Irish Farms (Los Angeles, CA). The investigations utilizing rabbits described in this report conformed to the Guiding Principles in the Care and Use of Animals (DHEW Publication, NIH 80-23) and the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 2. Cell Culture Protease, epidermal growth factor (EGF), hydrocortisone, and fluorescein isothiocyanate-labeled dextran (FD-4) were purchased from Sigma Chemical (St. Louis, MO). Sodium fluorescein and fluorescein isothiocyanate (FITC) were from ICN Pharmaceuticals, Inc. (Costa Mesa, CA). S-MEM, fetal bovine serum (FBS), gentamicin, Fungizone™ and DMEM/F12 media were from GIBCO BRL/Life Technologies (Gaithersburg, MD), while PC-1, penicillin, and streptomycin were from BioWittaker, Inc. (Walkersville, MD). Insulin-transferrin-selenium (ITS+ ) premix, collagen, fibronectin, laminin, and bovine pituitary extract (BPE) were from Collaborative Biomedical (Bedford, MA). [14C]-Mannitol was from Moravek Biochemicals, Inc. (Brea, CA). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 3-1: p-Blockers and its physiochemical properties and source. Drug Solubility (mM) MWa (g/mol) Log PCb pKa Molecular Charge Source (±) Sotalol-HCI 64.8 308.8 -0.62 8.15, 9.65 +1 Sigma lot # 99H4042 (RT, desiccated) Atenolol 1.13 266.3 0.16 9.32 +1 Sigma lot #99H1335 (RT) Nadolol 162 (EtOH) 309 0.93 9.39 +1 E.R. Squibb & Sons, lot # 02-806-20096 (RT) Pindolol 80.5 (0.1 N HCI) 248.3 1.75 8.8 +1 Sigma lot# 68H1314 Acebutolol-HCI 134 372.9 1.77 9.2 +1 Sigma lot # 17F04511 (4°C, desiccated) (±) Metoprolol-tartrate salt 73 684.8 1.88 9.24 +1 Sigma lot # 68H0845 (RT) Timolol-maleat salt 115 432.5 1.91 9.21 +1 Sigma lot#128H0686 Levobunolol-HCI 291 2.40 9.32 +1 G enerous gift from Allergan (RT, light sensitive) Labetaloi-HCI 137 364.9 2.55 -1 Sigma lot # 85F0253 (RT) Alprenolol-tartrate salt 25 399.4 2.61 9.65 +1 Sigma lot # 115F-0451 (RT) DL-Propranolol-HCI 169 (heat) 295.8 3.21 9.23 +1 Sigma lot # 97H0311 (4°C, desiccated) Betaxolol 347 3.44 +1 Generous gift from Alcon (4°C) a Molecular weight. b Log octanol/ pH 7.4 buffer partition coefficient. Ref: Schoenwald and Huang (1983) J. Pharm. Sci. 72(11), 1266-1272. RT = room temperature. Note: solubility was evaluated in water unless indicated otherwise. ON O 3. Chemicals All solutions used were prepared fresh and all chemical used were of high purity unless otherwise indicated. Please see Table 3-1 for information on P-adrenergics. B. METHODS 1. Preparation of Bicarbonate Ringer’s Solution Unless otherwise indicated, all experiments were conducted in the HEPES-buffered bicarbonated Ringer’s solution (BRS) maintained at 37°C and pH 7.4 equilibrated with 95% air/5% CO2. The HEPES-buffered Ringer solution contained (in mM) 116.4 NaCl, 5.4 KC1, 25 NaHC03, 0.78 NaH2P 0 4, 1.8 CaCl2-2H20 , 0.81 MgCl2-6H20 , 5.55 D-glucose and 15 HEPES (pH 7.4). Na+ -free Ringer’s solution was prepared by equimolar replacement of NaCl, NaH2P 04j and NaHC03 with choline chloride, KH2P 04 ) and choline bicarbonate, respectively. The Cl' -free Ringer’s solution was prepared by equimolar replacement of NaCl, KC1, and CaCl2-2H2 0 with Na-isethionate, D-gluconic acid (K salt) and CaS04-2H20 , respectively. The osmolarity of these Ringer’s solutions was adjusted to 300 mOsm/1 H2 0 with D-mannitol. 2. Primary Culture of Corneal Epithelial Cell Layers (RCrECL) 2.1. Primary Corneal Epithelial Cell Culture Procedure A simple method of primary RCrECL culture is outlined in Scheme 3-1. Briefly, rabbits were sacrificed by a rapid intravenous overdose of sodium pentobarbital 61 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pigmented Wash 2x Centrifuge Filter Reconstitute Transwell™ or Snapwell™ Day 6-8 Scheme 3-1: Summary of comeal epithelial cell culture procedure. (120 mg/kg body weight). Upon excision, the whole eyeball was placed in an eyeball holder (Fig. 3-1) that only exposed the epithelial layer to 0.2% protease in S- MEM for 60 min at 37°C. The suspension of dissociated epithelial cells was centrifuged at 200x g for 10 min. The resulting pellet was suspended in S-MEM containing 10% FBS, filtered through a 40 pm cell strainer, centrifuged and the pellet was resuspended in PC-1 or DMEM/F12 media supplemented with antibiotics (1 pg/mL Fungizone "(amphotericin B), 50 pg/mL gentamicin, 100 U/mL penicillin, and 100 pg/mL streptomycin), Insulin-transferrin-selenium (ITS+ ) premix, 30 pg/mL bovine pituitary extract (BPE), 1 ng/mL EGF and 0.36 pg/mL hydrocortisone. Cell count and viability were assessed by 0.1% trypan blue exclusion. Unless otherwise indicated, cells were plated at 0.9 x 106 cells/cm2 onto Costar 6.5-mm Clear Transwells pre-treated with 3:1:1 proportion of collagen (3 mg/mL)/ fibronectin 62 Excised eyeball Protease ^ Treatment DNase/ FBS ------------ Treatment Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (1 mg/mL)/ laminin (1 mg/mL). Air-interface condition was created by removing media from the apical compartment and keeping 0.3 mL of media in the basolateral compartment on day 1. Cells were kept in a humidified incubator at 37°C in 95% air/5 % CO2, and media was changed every day. Stripping Method Holder Method Corneal Endothelium Eyeball Stroma Protease Solution Corneal Epithelium Corneal Epithelium Fig. 3-1: Corneal epithelial cell isolation methods. 2.2. Bioelectric Parameter Measurements Transepithelial electrical resistance (TEER) and potential difference (PD) were measured with an EVOM epithelial voltohmmeter (World Precision Instruments, Sarasota, FL) and corrected for the background contributed by the filter coated with only the extracellular matrix materials. Ieq (pA/cm2) was estimated from the ratio of PD over TEER. 2.3. Transmission and Scanning Electron Microscopy of RCrECL Primary cultured cells on days 7-8 were fixed in half-strength Kamovsky’s 2% paraformaldehyde-2.5% glutaraldehyde fixative in phosphate buffer, postfixed in 1% Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. osmium tetroxide (OSO4), and stained with 1% uranyl acetate, each for 1 hr at 4°C before being dehydrated in a graded series of ethanol. For transmission electron microscopy, cultures were infiltrated and embedded in Epon, thin-sectioned (60-70 nm), placed on Formvar-coated slot grids, and stained with uranyl acetate and lead citrate for examination in a Hitachi H-7000 electron microscope. Cells prepared for scanning electron microscope were dried in a tousimis autosamdri 810 critical point dryer, coated with gold-palladium in Technics Humer VI Sputtering system and examined. 2.4. Immunohistochemical Staining of Cytokeratin Primary cells cultured to days 7-8 were fixed with 4% paraformaldehyde/phosphate buffer saline, pH 7.4 (PBS) buffer for 1 hr and incubated overnight at 4°C in 30% sucrose/PBS (pH 7.4) solution. Cell layers were embedded in OCT compound (Miles, Elkhart, IN) and quick-frozen in liquid nitrogen. Cryostat sections of 8 pm thick were cut on a Tissue-Tek cryostat from each frozen specimen, placed on gelatin-coated slides, and air-dried overnight at 4°C. Non-specific binding was blocked by incubating the samples in 0.1 M cacodylate/1% BSA/0.1% Triton X-100 in PBS for 10 min at room temperature. The samples were incubated with the primary antibody (IgG), monoclonal mouse anti-epithelial keratin-AE5 (1:50 dilution) (ICN Pharmaceuticals, Inc., Costa Mesa, CA) for 30 min at 37°C, gently washed three times with PBS, and then incubated with the secondary antibody coupled rhodamine (CY3-rabbit anti-mouse IgG , Chemicon, Temecula, CA) at 64 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1:100 dilution for 30 min at 37°C. Those samples not treated with the primary antibody served as the negative control. 2.5. Paracellular Permeability Assessment All permeability experiments were performed in Bicarbonated Ringer’s Solution (BRS) at an osmolality of 300 mOsm/kg under a humidified atmosphere of 5% CO2/ 95% air at 37°C (Scheme 3-2). Transport was initiated by dosing the apical compartment with BRS containing a paracellular marker. The paracellular markers were: (a) [14C]-Mannitol (MW=182.2; 5 pCi/mL (100 pM)); specific radioactivity, 50 mCi/mmol; (b) Na fluorescein (MW=376; 0.2 mg/mL); (c) FITC, MW=389; 0.25 mg/mL); and (d) FD-4 (MW= 4,400; 5 mg/mL). Fluorescein, FITC and FD-4 were dosed on the mucosal side. Samples of 200 pi were collected from the basolateral side every 30 or 60 minutes for up to 4 hours. Radiolabeled samples were mixed with liquid scintillation cocktail (Econo-safe, Research International, Mount Prospect, IL) for assay using a liquid scintillation counter (LS 1801, Beckman, Fullerton, CA) and the fluorinated samples were analyzed in a Hitachi fluorescence spectrophotometer. The 200 pi aliquot removed was immediately replenished with an equal volume of fresh buffer. The cumulative appearance of marker in the receiver compartment was plotted as a function of time. The steady-state flux (J) and the apparent permeability coefficient (Pap p ) were estimated by linear regression using the equation: Pa p p = J/(AC0) = (dQ/dt)/(ACo) (1) 65 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 where J = dQ/dt is the solute flux, A is the diffusional surface area (0.33 cm ), and Co is the initial drug concentration. 2.6. Active Ion Transport (Ussing-Type Chamber) Processes in RCrECL 2.6.1. Ussing-Type Chamber Set Up All experiments were performed under short-circuit conditions in an Ussing-type chamber using automatic voltage clamp device (558C-5, Bioengineering Department, University of Iowa, Iowa City, IA) (Fig. 3-2). The whole chamber set up is thermostatically maintained at 37°C utilizing a circulating water bath. Comeal epithelial cell layers grown on Snapwells-Clear™ to confluency of days 7-8 in culture were mounted in Ussing chamber and incubated with BRS buffer pre equilibrated at 37°C with 5% CO2/ 95% air. Six milliliters of BRS buffer on each side of RCrECL were agitated with bubbling of 5% CO2/ 95% O2 into the reservoir fluid during the experiment to help maintain the pH of Ringer’s solution and to yield adequate mixing. The nominal surface area of the exposed RCrECL is 1.0 cm2. Direct current flowing across the cell layers was applied with a pair of matched Ag/AgCl electrodes, completing an electric circuit with two conducting polyethylene bridges filled with 4% agar in 3M KC1, whose tips were positioned away from cell layer surfaces at far ends of two reservoirs. The electrical output of the voltage-sensing, calomel electrodes (also connected via agar bridges whose tips were positioned near RCrECL surfaces) was amplified by the voltage-clamp unit. Short-circuit current (Isc ) flowing in the bath-RCrECL-bath circuit was monitored 66 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.2 ml BRS 240 m in transport / 200 ml 1 Time ( ' Incubation Dosing > 37°C Day 7-8 Cell culture 0.8 ml BRS Replace with fresh BRS buffer P app dQ 1 dt C0 *A x 0 50 100 150 200 250 300 Tim e (m in) I P = apparent permeability coefficient (cm/sec) dO/dt = flux (amount/time, mole/sec) C = initial dosing drug concentration (mole/cm) A = normalized area of the epithelial barrier (0.33 cm2 ) Mix samples with scintillation cocktail for assay in a scintillation counter or analysis by HPLC Scheme 3-2: Summary of transport procedure across RCrECL with a strip chart recorder (Kipp and Zonen, Delft, Netherlands). At 60-sec intervals, 2 mV voltage pulse (AV) was imposed for 3 sec across the short-circuited RCrECL. The observed deflection in the resultant current pulse (Al) flowing on top of Isc was used to estimate the transepithelial electrical resistance (TEER) according to the relationship: TEER = (AV/AI)A, where A is the nominal surface area (1 cm2 ) of the Ussing chamber opening. Before each experiment, the background resistance (<100 Q.cm^) contributed by bathing fluids, agar bridges, and a blank filter was compensated for by the automatic voltage clamp unit. 67 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Voltage Clamp Calomel Electrode Ag/Ag/CI / I Agar Bridges Electrode Corneal Epithelial Cell Layers BRS Fig. 3-2: Ussing-type chamber used in the analysis of ion transport processes of RCrECL (diagram courtesy of Dr. Michael Shiue). 2.6.2. Pharmacological Assessment of RCrECL Ion Transport Properties Several compounds were used to assess the ion transport properties of RCrECL. Ouabain, a Na+ /K+ -ATPase inhibitor, was added to the basolateral fluid at a final concentration of 10 pM. N-phenylanthranilic acid (NPAA; 1 mM), a Cl'-channel blocker, was added to the apical fluid. BaCL (2 mM), a specific K+ channel inhibitor, was added to the basolateral fluid. Amiloride and benzamil, Na+ -channel inhibitors, were added to the apical fluid at a concentration ranging from 0.001 to 200pM. The presence of transport processes for Na+ -absorption and Cl'-secretion across RCrECL was also evaluated by studying bioelectric properties of cultured RCrECL under either Na+ -free or Cl'-free conditions by employing the superfusion 68 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. technique. The bathing fluids in the Ussing chamber were replaced with Na+ or C l- free BRS buffer on both sides by gravity feeding at a 5 mL/min rate to the bottom of the bathing fluid, while simultaneously suctioning off the excess volume at the surface of the bathing fluid (i.e., keeping the bathing fluid volume at a constant level). A replacement of bathing fluids was carried out with 100 mL volume of the new buffer for each side of the RCrECL. Isc and TEER were continuously monitored during the whole superfusion process. This method of buffer exchange reduces the adverse effects induced by sudden changes in hydrostatic pressure. 2.6.3. Sodium and Chloride Flux Studies Unidirectional sodium and chloride fluxes were determined in the apical-to- basolateral and basolateral-to-apical directions by separate experiments using 2 2 Na (lpCi/mL) and 3 6 C 1 (0.5pCi/mL). Either radionuclide was added to the upstream (i.e., donor) fluid, where 3 H-mannitol (10 pCi/mL) as a paracellular marker was concurrently added. Samples at 500 pi size were collected from the downstream (i.e., receiver) fluid every 15 or 30 min for up to 3 hours, mixed with liquid scintillation cocktail (Econo-safe, Research International, Mount Prospect, IL) and assessed for their radioactivity using a liquid scintillation counter (LS 1801, Beckman, Fullerton, CA). The cumulative appearance of radioactivity in the receiver fluid was plotted as a function of time and the steady-state flux (J) was estimated by linear regression. Apparent permeability coefficient (Pap p ) was calculated from the relation, 69 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pa p p = J/(AC0 ) = (dQ/dt)/(AC0 ) (1) where J = dQ/dt is the radionuclide flux, A is the RCrECL surface area (1 cm2 ), and Co is the initial sodium or chloride concentration. 3. In Vitro Drug Transport Processes through the RCrECL 3.1. Radiometry Analysis All permeability experiments were performed in a humidified atmosphere of 5% CO2/ 95% air at 37°C as outlined in Scheme 3-1. Prior to each experiment, the cell layers were washed and incubated with bicarbonated Ringer’s solution (BRS) at an osmolality of 300 mOsm/kg for at least an hour. The study was initiated by dosing the apical compartment with BRS containing radiolabeled marker. [1 4 C]-Mannitol was dosed at 5 pCi/mL (100 pM) on the apical side. [1 4 C]-Guanidine, [5,6-3 H]- Uridine and L-[l-1 4 C]Lactate were dosed at 1 pCi/mL either in the apical or basolateral side. [3 H]L-Camosine was dosed at 2 pCi/mL and 10 pM cold camosine. [3 H]-Betaxolol was dosed at 1 pCi/mL and 1 mM cold betaxolol. Samples of 200 pi were collected from the basolateral side every 30 or 60 minutes for up to 7 hours. Radiolabeled samples were mixed with liquid scintillation cocktail (Econo-safe, Research International, Mount Prospect, IL) for assay using a liquid scintillation counter (LS 1801, Beckman, Fullerton, CA) and the fluorinated samples were analyzed in a Hitachi fluorescence spectrophotometer. The 200 pi aliquot removed was immediately replenished with an equal volume of fresh buffer. 70 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3.2. HPLC Analysis The (3 -adrenergic drug concentration used in the in vitro permeability studies was 1 mM except propranolol (0.5 mM). Propranolol was found to be toxic to the cell layers as assessed by bioelectric parameters before and after 4 hr permeability experiment. The HPLC conditions used in the analysis of P-adrenergics are listed in Table 3-2. 4. In Vitro Drug Uptake Studies 4.1. Gly-Sar Uptake [3 H]Glycylsarcosine (4 Ci/mmol) was purchased from Moravek Biochemicals (Brea, CA). Unlabeled gly-sar, L-camosine, L-histidine, P-alanine, gly-L-phe, gly-D-phe, L-pyroglutamyl-L-histidyl-L-prolinamide (TRH), 2,4-dinitrophenol (DNP), and sodium azide (NaN3 ) were purchased from Sigma-Aldrich (St. Louis, MO). Cyclacillin and cefadroxil were generously provided by Takeda Chemical Industries (Osaka, Japan) and Bristol-Myers Squibb Pharmaceutical Research Institute (Princeton, NJ), All uptake experiments were performed in bicarbonated Ringer solution (BRS) at an osmolality of 300 mOsm/kg under a humidified atmosphere of 5% CO2/ 95% air at 37°C as outlined in Scheme 3-3. The buffer contained (in mM) 116.4 NaCl, 5.4 KC1, 25 NaHC03, 0.78 NaH2 P 0 4, 1.8 CaCl2-2H2 0 , 0.81 MgCl2-6H20 , 5.55 D-glucose and 15 HEPES (pH 7.4). Prior to each experiment, the cell layers 71 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. were washed and equilibrated in pre-warmed BRS buffer for 30-60 min. Gly-sar uptake studies were carried out in the presence of an inwardly directed proton gradient (apical pH 6.0, basolateral pH 7.4) from the apical side. Uptake was initiated by dosing the apical (0.2 mL) or basolateral (0.8 mL) fluid with 5 pCi/mL (1.25 pM) [3 H]gly-sar and an appropriate amount of unlabeled inhibitors. After a predetermined period of time, uptake was terminated by suctioning off the dosing solution and cell layers were immersed in three washes of ice-cold pH 7.4 BRS buffer (20 mL each wash). The cell layers were then solubilized in 0.5 mL of 0.5% Triton X-100 solution. Ten microliters of the solution were taken for protein assay by the method of Bradford (Bradford, 1976), using a Bio-Rad protein assay kit (Hercules, CA) with bovine serum albumin as a standard. The rest of the sample was mixed with 5 mL of EconoSafe scintillation cocktail (Research Products International Corp., Mount Prospect, IL) for measurement of radioactivity in a liquid scintillation counter (Beckman, Fullerton, CA). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. R eproduced with permission o f the copyright ow ner. Further reproduction prohibited without p erm ission. Table 3-2: p-Biockers, their log partition coefficient (PC) values and HPLC assay conditions. Drug MW 9 Log PC' % AcN/MeOH ISa Wavelength Drug Rtb (min) IS Sotalol 308.8 -0.62 3.5% AcNc Atenolol 225 nm 10 12 Atenolol 266.3 0.16 3.5% AcN Sotalol 225 nm 12 10 Nadolol 309 0.93 15% AcN Pindolol (10 pM) 220 nm 5.7 8.0 Pindolol 248.3 1.75 15% AcN Nadolol (10 pM) 220 nm 8.0 5.7 Acebutolol 372.9 1.77 18% AcN Nadolol (10 pM) 220 nm 8.7 4.3 Metoprolol 684.8 1.88 15% AcN Timolol 280 nm 8.6 Timolol 432.5 1.91 53% MeOHd Propranolol 290 nm 4.2 8.2 28% AcN Propranolol 290 nm 4.2 11.5 Levobunolol 291 2.40 28% AcN Betaxolol 225 nm 4.5 12 Dihydrolevobunolol (metabolite) EX 265; EM 310 nm Labetalol 364.9 2.55 28% AcN Alprenolol (10 pM) 272 nm 7.9 11.8 Alprenolol 399.4 2.61 28% AcN Labetalol (10 pM) 272 nm 11.8 7.9 Propranolol 259 3.21 53% MeOH Timolol 290 nm 4.2 8.2 Betaxolol 347 3.44 28% AcN Levobunolol 224 nm 12 4.5 alnternal standard (IS), bRetention time (Rt), c Acetonitrile (AcN), dMethyl alcohol (MeOH). Mobile phase: 0.2% TEA, pH 3.0 (adjust with phosphoric acid) Flow rate=1 mL/min for all com pounds except satolol and atenolol (1.5 mL/min). e Molecular W eight. f Log octanol/ pH 7.4 buffer partition coefficient. '- a u > 0.1 ml BRS 100 ml 20 min uptake Dosing Time Incubation > Day 6-8 o.6 ml BRS Cell culture Cut filter containing cell layers 2 1 3 37°C Aspirate apical and basal media A ---------------- J Ice-cold BRS buffer ” P i Incubate in TritonX I A I Add scintillation cocktail ► Incubate in dark overnight Scheme 3-3: Summary of RCrECL uptake procedure. Count radioactivity in a scintillation counter 4.2. Functional and Molecular Evidence of Nucleoside Transporters 4.2.1. Uridine Uptake Na+ -free Ringer’s solution was prepared by equimolar replacement of NaCl, NafUPOzt, and NaHCC> 3 with choline chloride, KEyPO^ and choline bicarbonate, respectively. Confluent multilayers of cells were washed three times and equilibrated in BRS buffer for -60 min. One hundred microliters of BRS containing 5 pCi/mL (0.12 pM) [3 H]uridine was then added to the apical compartment of each well (or 500 pL to basolateral compartment). After 1 min uptake, both apical and basolateral fluid were aspirated and each well was washed three times rapidly with ice-cold BRS buffer. In the case of Na+ -free conditions, cells were washed in ice- cold Na+ -free BRS buffer. Cell layers were cut off and solubilized in 0.5 mL of 74 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.5% Triton X-100 overnight. Next day, 20 pL of solubilized protein was taken for protein assay and 5 mL of scintillation cocktail was added to the sample. The radioactivity was measured by radiometric analysis. The protein content was determined spectrophotometrically using a commercial protein assay kit (Bio-Rad, Hercules, CA). NBMPR (100 mM) was initially dissolved in DMSO and was diluted to working concentration (up to 0.2 mM) in [3 H]uridine-containing uptake solution. As a control, 0.5% DMSO was also included in all uptake solution wherever necessary. Control experiments with 0.5% DMSO had no effect on total [3 H]uridine uptake by cultured RCrECL. 4.2.2. Concentration Dependency of Uridine Uptake Both apical and basolateral 1 min [3 H]uridine uptake was determined in the presence of various unlabeled uridine concentrations (apical = 0, 5,10, 20, 40, and 80 pM; and basolateral = 0, 0.05, 0.1, 0.2, 0.5, 1.0, and 1.5 mM) in BRS, Na+ -free BRS or Na+ -free BRS containing 0.2 mM NBMPR. RCrECL were incubated in various buffer conditions (e.g. Na+ -free BRS or Na+ -free containing 0.2 mM NBMPR) for 30 min before dosing from the apical or basolateral compartment. The amount of radiolabeled drug taken up by cells was normalized for the protein content of the cell layers. The initial uptake rate was estimated from the slope of a plot of amount of uridine transported vs. time. The Michaelis-Menten constant (Km ) and the maximal saturable uptake rate (Vm a x ) were estimated by nonlinear regression against uptake rate-concentration profile using the following equation, 75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. v=vmax[C]/(Km + [C]) + Kd[C] where V, [C], and Kd are total uptake rate, applied concentration, and nonspecific uptake rate, respectively. 4.2.3. RNA Isolation and RT-PCR Total RNA was isolated from freshly isolated corneal epithelial cells (in the same manner as cell isolation for cell culture) using TRIzol reagent (GIBCO BRL). Four micrograms of total RNA was used for first-strand cDNA synthesis using random hexamer primers and Superscript II Rnase H- Reverse Transcriptase (SuperScriptPreamplification System, GIBCO BRL). The resulting first-strand cDNA was directly used for PCR amplification. Different sets of primers were designed and synthesized for PCR analysis. The two primers used for amplifying rbSPNTl (rbNl or rbCNT2) were CTTCATATCCTTTGGG (sense primer corresponding to nucleotides 1105-1120) and GCAAATCCACAGAG (antisense primer corresponding to nucleotides 1649-1662), which generated a 557-bp rbCNT2 PCR product. The two primers used for amplifying rbENT2 were CCTGGCCGGGATCTTTGCT (sense primer corresponding to nucleotides 518-536) and GCAGATGGGGTTGAAGAACT (antisense primer corresponding to nucleotides 994-1013), which generated a 495-bp rbENT2 PCR product. The primers were used in PCR amplification with corneal epithelium cDNA as the template under the following conditions: 94°C for 1 minute; 45°C for 1 minute; 72°C for 1 minute for 30 cycles for the rbCNT2 primers. The PCR condition for rbENT2 76 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. primers were the same except with the second temperature change at 55°C instead of 45 °C. 5. In Vivo Ocular Absorption 5.1. In Vivo Topical Dosing All in vivo studies were carried out as outlined in Scheme 3-4. All radiolabeled drugs were dosed at 1 pCi/ eye and all non-radiolabeled compounds ((3-adrenergics) were dosed at 15 mM prepared in 10 mM Tris buffer pH 7.4 (adjusted with 10 N NaOH) at 25 pL. Thirty- minute post-dosing, approximately 10 mL blood was taken from the rabbit ear artery and sacrificed by an over dose of sodium pentobarbital through the marginal ear vein. Aqueous humor (-0.2 mL/ eye) was first extracted from the eye using a 1 mL syringe and 26-gauge needle. Each tissue will carefully be removed and rinsed in saline to remove excess (unabsorbed) radioactive drug and blotted on Wattman filter paper. The entire procedure took less than 5 min per eye. Wet weight was determined using an analytical balance. 77 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ✓ Topical 25 (il drug 30 min absorption Extract eyeball and aspirate aqueous humor Take blood Blood Centrifuge 1500g fo r 10 min I Remove ocular tissues, rinse in ice-cold saline and weigh each tissue Corneal epithelium Conjunctiva Stroma-endothelium Iris-ciliary body Lethal injection of Na Pentabarbital (120 mg/kg) Obtain plasma for drug extraction HPLC analysis of non radiolabeled com pounds Overnight at Room temperature Ocular tissues in tissue solubilizer, Protosol™ A ssay in a scintillation spectrom eter Scheme 3-4: Summary of in vivo topical dosing procedure 5.2. Radiometry Analysis Tissues were oxidized in glass scintillation vials containing 1 mL (all ocular tissues) or 2 mL (conjunctiva) of tissue solubilizer, TS-2™ (Research Products International Corp.; Mount Prospect, Illinois) overnight. After all tissues have been solubilized, 10% glacial acetic acid was added to the mixture before adding 10 mL biodegradable scintillation cocktail, Bio-Safe II™. The samples were then counted in a scintillation counter. After background correction, the final count for each sample was converted to pg of total drug compound based on total radioactivity and was normalized for the tissue weight as indicated. 78 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5.3. p-Blockers (HPLC Analysis) Drugs extracted from plasma and ocular tissues were processed as follow: Blood samples were collected from the ear marginal vein 30 min after administration of the drug. After centrifugation (10 min at 12000 rpm), the plasma samples were separated and stored at -20°C before analysis. Plasma of 1.5 mL was mixed with 100 pL of internal standard (20 pM propranolol or betaxolol), 800 pL of 10 mM Tris buffer pH 7.4 (260 mosmol kg-1 in tonicity), 500 pL of 1 M ammonium acetate buffer (pH 9.3) and 8 mL of diethyl ether in a 15 mL screw-capped conical centrifuge tube. After vortexing for 3 min, the mixture was centrifuged at 1500 g for 10 min. The supernatant was transferred to a 10 mL screw-capped conical centrifuge tube containing 200 pL of 0.2 N HC1, vortexed for 3 min, and centrifuged at 1500 g for 10 min. The supernatant was discarded, while 100 pL of the aqueous phase containing drug and internal standard was injected into the HPLC. Due to the fast metabolism of levobunolol to dihydrolevobunolol, extraction of levobunolol and dihydrolevobunolol were carried out according to Tang-Liu et al. (1988). Briefly, three milliliters of blood was immediately mixed with 1 mL of 1 M ammonium acetate buffer (pH 9.3) and 100 pL of internal standard (20 pM betaxolol) in an ice bath and centrifuged within 20 min at 1500 g for 10 min at 4°C to prevent in vitro metabolism of levobunolol to dihydrolevobunolol. The supernatant will be transferred to a 15 mL screw-capped tube, and extracted with 8 mL of diethyl ether by vortexing for 3 min and centrifuging at 1500 g for 10 min. 79 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The supernatant was extracted of levobunolol and dihydrolevobunolol using the same procedure described for timolol and betaxolol. Aqueous humor extracted (150-200 pL) was injected directly (with addition of internal standard) into the HPLC without any processing. The eyes were enucleated and rinsed with normal cold saline solution, and then the ocular tissues and aqueous humor were removed within 10 min of the animals’ death. Pre-weighed excised tissues were soaked in 1 mL of 0.6% HCIO4 at 8°C for 12 hr (only for drugs that are sensitive to basic conditions). Thereafter, the samples were mixed with 0.1 mL of 10 pg/mL of internal standard solution and 0.5 mL of 1 M ammonium acetate buffer (pH 9), extract with 8 mL of diethyl ether by vortexing for 3 min, and then centrifuge at 1500 g for 10 min. The upper organic layer was transferred to a 15-mL screw-capped conical centrifuge tube containing 200 pL of 0.2 N HC1, vortexed for 3 min, and centrifuge at 1500 g for 10 min. The organic phase was discarded and 100 pL of the aqueous phase, containing the drug and internal standard, was injected into the HPLC. The same procedure was used to extract drug from 2 mL of plasma. Similar to plasma extraction of levobunolol and dihydrolevobunolol, ocular tissues were also treated as described by Tang-Liu et al. (1988). Tissues were separated and placed in a prelabeled tube containing 1 mL of the internal standard solution in acetonitrile. Ocular dissection was completed within one minute after animal euthanasia. Tissue samples were stored in acetonitrile immediately in order to cease any enzymatic reactions. On the day of analysis, samples were brought to room temperature and vortexed before centrifugation. The organic phase was 80 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. collected and evaporated to dryness under N2 at about 30°C. The dried residue in each tube was vortexed with 500 pL of distilled water and 30 pL of 3N NaOH for 30 sec. Five mLs of ethyl ether was then added to each tube and vortexed for 1 min. All tubes were centrifuged for better phase separation. The upper ether layer was then transferred and evaporated to dryness under N2 at about 30°C. The residue to each tube was dissolved in 300 pL of the mobile phase before injection. Ten milligrams of levobunolol was dissolved in 400 pL of methanol containing 10 mg of sodium borohydride and mechanically agitated for 30 sec. The resultant solution was left in sealed tubes for 30 min in darkness. After the reduction, the sample was dried under a nitrogen stream and then reconstituted in 250 pL of eluent. Because of the presence of a conjugated carbonyl group, levobunolol has different UV spectral characteristics from dihydrobunolol. The latter absorbs UV poorly. Levobunolol fluoresces poorly, while dihydrobunolol has two excitation maxima at around 225 nm and 275 nm. C. CALCULATIONS AND STATISTICAL ANALYSIS All calculations (linear and nonlinear regression analysis) were performed by using Prism 2.01 (Graph PAD, Los Angeles, CA). Data are presented as means ± standard error of the mean. Statistical significance (p<0.05) was determined by One-Way Analysis of Variance (ANOVA) followed by Student-Newman-Keuls multiple comparison tests for determining statistical difference among multiple (n >3) group means. 81 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IV. RESULTS Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A. DEVELOPMENT AND CHARACTERIZATION OF A PRIMARY CORNEAL EPITHELIAL CELL CULTURE MODEL 1. Time Course of Bioelectric Parameters Our cell isolation method yielded 7-8 x 106 cells per animal, equivalent to 30 Transwell™ (0.33 cm3 ) samples, with viability greater than 98%. Both AIC and LCC cell multilayers began to exhibit a measurable TEER and PD from day 3 onwards, reaching confluency on days 7-8. Compared with LCC, AIC developed a 5- fold higher TEER and a 3-fold higher PD (Fig. 4-1). 2. Effect of Culture Media Composition on the Bioelectric Parameters Of the two culture media we tested, cells grown in PC-1 media achieved a PD and a TEER value about two-thirds of that afforded by supplemented DMEM:F12 + 1 ng/ml EGF media, whereas DMEM:F12 non-supplemented resulted in ~60- and -45- fold lower in TEER and PD, respectively (Table 4-1). Therefore, we used a supplemented DMEM:F12 medium in subsequent cultures. In addition, FBS was eliminated from the medium due to its adverse effect on the TEER and PD of the cell layers when compared to DMEM:F12 + 0 ng/ml EGF (Table 4-1). The EGF concentration was held at 1 ng/ml because cells grown in DMEM/F12 at 5 and 10 ng/ml concentrations resulted in a lower TEER (Table 4-1). Cells grown under no EGF resulted in significantly lower in PD and Ieq . In terms of seeding cell density, <r r y 0.9 x 10 cells/cm was found to be optimal (Table 4-2). At this seeding density, 83 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. bioelectric parameters were found to be optimal when cells were placed under AIC condition on day 1 rather than on days 2 or 3 (Table 4-3). 30. 25 IN E o d cc L it L U H 10 12 14 4 8 0 2 6 Time (day) Time (day) Time (day) Fig. 4-1: Time course of bioelectric parameters in AIC (■) and LCC (□) RCrECL. Peak TEER, PD, and le q for AIC are 5.08 ± 0.17 kQ.cm2, 26.3 ± 2.8 mV, and 5.1 ± 1.0 |aA.cm2 (n = 10), respectively. The corresponding values for LCC are 0.21 ± 0.01 kQ.cm2, 1.1 ± 0.15 mV, and 5.9 ± 1.0 pA.cm2 (n = 29). Error bars represent s.e.m. Asterisk denotes significant difference (p < 0.001) between AIC and LCC on a given day by Student’s t-test. Where not shown, the error bar is smaller than the size of the symbol. 84 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4-1: Effect of Culture Media Composition on the Transepithelial Electrical Resistance (TEER), Potential Difference (PD), and Short-Circuit Current (le q ) of AIC Cultured RCrECL on Day 7-8. Media Composition TEER (kQ.cm2 ) PD (mV) leq (pA/cm2 ) DMEM/F12 non-supplemented 0.054 ± 0.02 * 0.49 ±0.12* 33 ± 9.9 * DMEM/F12 + 1% FBS 1.63 ±0.21 * 8.3 ± 0.7 * 5.5 ± 0.3 DMEM/F12 + 0 ng/ml EGF 3.11 ±0.20 10.1 ±0.4* 3.3 ± 0.2 * DMEM/F12 -i-1 ng/ml EGF 3.21 ± 0.03 22.0 ± 2.0 6.8 ± 0.6 DMEM/F12 + 5 ng/ml EGF 1.52 ±0.13* 14.9 ±1.3 10.1 ±0.9 DMEM/F12 + 10 ng/ml EGF 1.18 ±0.06* 11.9 ±0.8 10.1 ±0.4 PC-1 2.26 ±0.10* 6.3 ± 0.5 * 2.8 ± 0.2 * Media composition for all cultures w as supplem ented (see Materials and Methods), unless otherwise indicated. Optimal conditions are indicated in bold. Values are m ean ± s.e.m ., n = 5-23. Statistical significance w as tested by one-way ANOVA and the Fisher’s PLSD post hoc test. Statistical significance for media composition analysis w as set at p < 0.05 when com pared with results from the DMEM/F12 + 1 ng/ml EGF culture condition. For all media composition analysis, cells were cultured at a plating density of 1.5 x 10® cells/cm2. Table 4-2: Effect of Seeding Density on the Transepithelial Electrical Resistance (TEER), Potential Difference (PD), and Short-Circuit Current (le q ) of AIC Cultured RCrECL on Day 7-8 of Culture. Cells/ cm2 TEER (kQ.cm2 ) PD (mV) leq (pA/cm2 ) (A) 1.5 x106 3.21 ± 0.03* 22.0 ± 2.0 6.9 ± 0.6 (B) 0.9x10® 5.08 ±0.17 26.3 ± 2.8 5.1 ±0.5 (C) 0.6x10® 3.40 ± 0.08* 20.4 ±1.0 6.0 ± 0.3 (D) 0.3x10® 2.34 ±0.16* 19.7 ± 1.3 8.6 ±0.8 Media composition for all cultures w as supplem ented (see Materials and Methods), unless otherwise indicated. Optimal conditions are indicated in bold. Values are m ean ± s.e.m ., n = 5-23. Statistical significance w as tested by one way ANOVA and the Fisher’ s PLSD post hoc test. In this analysis, cells w ere placed under AIC condition on day 1. The TEER show ed statistical significance betw een groups A and B, A and D, B and C, B and D, C and D; PD and leq w ere not statistically significant (p > 0.05). 85 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4-3: Effect of Air-Interfaced Condition (AIC) on the Transepithelial Electrical Resistance (TEER), Potential Difference (PD), and Short-Circuit Current (le q ) of AIC Cultured RCrECL on Day 7-8 of Culture. Day T E E R (kQ.cm2 ) PD (mV) leq (pA/cm2 ) 1 5.08 ±0.17 26.3 ± 2.8 5.1 ± 0.50 2 4.20 ± 0.23 10.8 ±1.4* 2.5 ±0.31* 3 3.61 ± 0.20* 7.7 ± 1.4* 2.0 ±0.31* Media composition for all cultures w as supplem ented (see Materials and Methods), unless otherwise indicated. Optimal conditions are indicated in bold. Values are m ean ± s.e.m ., n = 5-23. 'Statistical significance w as tested by one way ANOVA and the Fisher’ s PLSD post hoc test. For the effect of AIC condition, cells were seed ed at an optimal seeding density of 0.9 x 106 cells/cm2 a s cells w ere placed under AIC condition at day 1, day 2 and day 3. The TEER w as statistically significant betw een day 1 and day 3; PD and leq w ere statistically significant between day 1 and day 2 a s well a s betw een day 1 and day 3 (p<0.05). Table 4-4: Permeability Coefficient (Pa p p ) of Paracellular Marker Solutes in AIC and LCC Cultured RCrECL Compared With that in the Excised Cornea. Papp (10‘6 cm/s) Agent MW AIC LCC Tissue [1 4 C]-Mannitol 182 0.044 ±0.0036 1.93 ±0.12 0.1a Fluorescein 376 0.32 ± 0.01 1.13 ±0.15 0.43 ± 0.09b FITC 389 0.20 + 0.017 1.07 ±0.14 nd FD-4 4,400 0.018 ±0.003 0.66 ±0.09 0.056 ±0.016° Mean + SEM (n = 4-6); nd = not determined aFrom ref. (Klyce and Wong, 1977), b from ref. (Araie and Maurice, 1987) and cfrom ref. (Kruszewski etal., 1997) 3. Morphology of Cultured Corneal Epithelial Cell Layers 3.1. Scanning electron microscopy (SEM) At day 1 after seeding, the corneal epithelial cells are well attached and the cells are beginning to proliferate (Fig. 4-2). By day 7, the cells have grown to cover all surface area of the permeable support. 86 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3.2. Transmission electron microscopy (TEM) The corneal epithelial cells gradually flattened as they approached the surface (Fig. 4-3b) and can be distinguished into basal, wing and flattened superficial cell layers. The superficial cells showed microvilli and the formation of tight junctions and desmosomes (Fig. 4-3a). Numerous fine cytoplasmic filaments were also present (Fig. 4-3c). 4. Immunohistochemical Staining of Corneal Specific Keratin The cultured corneal epithelial cells stained positively for a cornea-specific basic K3 (64 kDa) keratin (Schermer et al., 1986) with a commercially available monoclonal antibody (AE5) (Fig. 4-4b), as did the freshly excised tissue (Fig. 4-4a). The omission of primary antibody as negative control did not stain (Fig. 4-4c). 5. Cell Layer Integrity Assessment - Paracellular Permeability Table 4-4 shows that the paracellular permeability of AIC was comparable to that of the isolated tissue. Compared with the AIC, the LCC was 44 times more permeable to mannitol, 4 times to fluorescein, 5 times to FITC, and 37 times to FD-4. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f the copyright ow ner. Further reproduction prohibited without p erm ission. Fig. 4-3: Transmission electron micrograph of day 7 AIC cultured RCrECL illustrating: (a; x 21,800) tight junction (TJ), (b; x 8,000) the major cell layers of the corneal epithelium (SCL = superficial cell layers; WCL = wing cell layers; BCL = basal cell layer), and (c; x 47,700) numerous cytokeratin. O O v O Fig. 4-4: Im m u n o flu o resc en c e stain in g for c o rn e a -sp e c ific 64 -k D a cytokeratin of AIC cu ltu red RCrECL: (a) positive sta in in g of rabbit tis s u e c o rn e a l epithelium with A E5 m o n o clo n al antibody (SC L = superficial cell lay ers; W C L = w ing cell layers; BCL = b a s a l cell layer), (b) positive stain in g of d a y 7 cu ltu red c o rn e a l epithelial ce lls with A E5 a n d (c) n e g a tiv e control of d a y 7 cu ltu red c o rn e a l epithelial cells. T h e c o rre sp o n d in g p h a s e -c o n tra s t m ic ro g ra p h s a r e in th e left-hand colum n. 90 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. B. ACTIVE ION TRANSPORT PROCESSES ACROSS RCrECL 1. RCrECL Derived From Pigmented Rabbits 1.1. Bioelectric Parameters of RCrECL Cultured at an Air-Interface (AIC) Air-interface cultured RCrECL grown on Snapwells™ reached confluency on days 7- 8 as monitored by it bioelectric properties (TEER, PD and Isc ). The corneal epithelial cell layers grown in an air-interface developed a PD of >15 mV, an Isc >3.5 pA/cm2 and has a TEER >4.5 kQ.cm2 on days 7-8. All the data below were obtained using RCrECL grown for 7-8 days, unless noted otherwise. 1.2. Na+ and Cl' Contribution to Total Isc Bioelectric parameters attained steady-state within 30-60 min from mounting the RCrECL in the Ussing-type chamber. Replacing Ringer’s solution in either the apical or basolateral fluid with Na+ -free Ringer’s solution caused Is c across RCrECL to decrease by 64 ± 1.5% (n=4) or 15 ± 2.9% (n=4), respectively (Fig. 4-5). Returning to Na+ containing Ringer’s solution led to a near complete 98 + 1.1% of the baseline value return to the Is c observed with BRS. When both apical and basolateral fluids were rendered to Na+ -free condition, Is c dropped to near zero, exhibiting 94 ± 8% (n=4) inhibition, in a nearly completely reversible manner. Placing Cl'-free Ringer’s solution on both apical and basolateral sides resulted in a decrease of Is c by 44 ± 4.4% (n=6), in a reversible fashion (90.3 ± 2.9% of baseline 91 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Isc ) (Fig. 4-6). Placing Cl-free Ringer’s solution on basolateral side alone resulted in a decrease of Is c by 80 ± 1.9%, also in a reversible mode (97 ± 1.2% (n=6)). Na+ - Free (A/B) E 94% Inhibition 6 4 2 0 0 20 40 60 80 100 120 Time (min) B C M E =JL Na+ - Free (B) 15% Inhibition Na+ - Free (A ) 64% Inhibition 6 4 Recovery 2 Recovery 0 0 30 60 90 120 150 180 Time (min) Fig. 4-5: T im e -c o u rs e s for lsc inhibition in prim ary AIC cu ltu red R C rEC L from p ig m e n te d rabbits u n d e r so d iu m -fre e condition. Effect of rem oving N a+ from both ap ical a n d b a s o la te ra l fluids on lsc is sh o w n in p a n e ls A. T h e effec t of rem oving N a+ from apical o r b a s o la te ra l fluid on lsc with its rec o v ery is sh o w n in p an e l B. E ac h d a ta point re p re s e n ts th e m e a n ± s .e .m . (n=4). W h e re no erro r b a r is sh o w n , it falls within th e sym bol. N ote: T h e re is no lo ss of T E E R during th e tim e period of an a ly sis. 92 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. E ■ 3 - C l - Free (A/B) % Inhibition 6 4 2 Recovery 0 120 60 80 100 20 40 0 Time (min) B 5 = L o ( 0 C l - Free (B) 80% Inhibition 6 4 Recovery 0 80 100 120 0 20 40 60 Time (min) Fig. 4-6: Time-courses for ls c inhibition in primary AIC cultured RCrECL from pigmented rabbits under chloride-free condition. Effect of removing Cl' from basolateral fluids on ls c is shown in panels A. The effect of removing Cl' from apical or basolateral fluid on ls c with its recovery is shown in panel B. Each data point represents the mean ± s.e.m. (n=6). Where no error bar is shown, it falls within the symbol. Note: There is no lost of TEER during the time period of analysis. 93 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.3. Effects of Pharmacological Agents on Is c Within 90 min from the application of 10 pM ouabain to the basolateral fluid (Fig. 4- 7), 86 ± 1.5% (n=5) reduction in Isc with a Tyz of 25 min was observed. Exposure of the basolateral compartment to 2 mM BaCl2 resulted in 18 ± 1.8% (n=4) decrease in Isc with a T1/2 of 54 sec (Fig. 4-8). Exposure of the basolateral compartment to 10 pM bumetanide resulted in 13 ± 1.9% (n=6) decrease in Isc with a T1/2 of 10 min (Fig. 4-9) without significantly affecting TEER. With 100 pM NPAA added to the apical fluid, a slow and gradual Isc decrease to 53 ± 3% (n=5) of baseline Is c (Fig. 4- 10) with a T1/2 of 20 min was observed. A typical time course for amiloride inhibition of Isc is illustrated in Fig. 4-11 A. A maximum inhibition by amiloride was about 50% of baseline Isc. The dose-dependent inhibition of Isc by apically applied benzamil and amiloride exhibited a half-maximal inhibition (IC50) at 0.1 and 1 pM, respectively (Fig. 4-1 IB). Benzamil also showed a maximal Isc inhibition of 50%. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. E § i O 0 ) 6 10 pM Ouabain (Basolateral) 4 86% Inhibition at 100 min T Vi s 25 sec 0 80 100 120 20 40 60 0 Time (min) Fig. 4-7: Time-course of ls c inhibition of basolateral Na+ /K+ -ATPase activity with ouabain in primary AIC cultured RCrECL from pigmented rabbits. Cell layers were mounted in modified Ussing-type chamber. Following stabilization of baseline bioelectric properties, 10 pM ouabain was added to the basolateral compartment. Data is reported as mean ± s.e.m. (n=5). The gradual decrease in ls c is accompanied by a gradual decrease in TEER to -15% of baseline TEER after 100 min treatment with ouabain. E o 6 2 mM BaCI2 (Basolateral) 4 2 18% Inhibition at 70 min T Vi = 54 sec 0 15 30 0 45 60 75 90 Time (min) Fig. 4-8: T im e -c o u rse of ls c inhibition of b a so la te ra l K+ c h a n n e l activity with BaCI2 in prim ary AIC cultured RCrECL from pigmented rabbits. Cell layers were mounted in modified Ussing- type chamber. Following stabilization of baseline bioelectric properties, 2 mM BaCI2 was added to the basolateral compartment. Data is reported as mean ± s.e.m. (n=4). The gradual decrease in Is c is accompanied by a gradual decrease in TEER to -15% of baseline TEER after 70 min treatment with BaCl2. 95 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5 10 p M Bumetanide (Basolateral) 13% Inhibition at 70 min T 1 /2 = 10 min — i----------------------1 ----------------------1 ----------------------1 ----------------------1 --------------------- 1 15 30 45 60 75 90 Time (min) Fig. 4-9: Time-course of ls c inhibition of basolateral Na+ (K+ )2Cr-cotransporter activity with bumetanide in primary AIC cultured RCrECL from pigmented rabbits. Cell layers were mounted in modified Ussing-type chamber. Following stabilization of baseline bioelectric properties, 10 pM bumetanide was added to the basolateral compartment. Data is reported as mean ± s.e.m. (n=6). Bumetanide had no significant effect on the RCrECL TEER during the treatment period. 6 53% Inhibition at 100 min T V 2 s 20 min 4 2 100 pM NPAA (Apical) 0 0 20 40 60 80 100 120 Time (min) Fig. 4-10: Time-course of ls c inhibition of apical Cl' channel activity with NPAA in primary AIC cultured RCrECL from pigmented rabbits. Cell layers were mounted in modified Ussing-type chamber. Following stabilization of baseline bioelectric properties, 100 pM NPAA was added to the apical compartment. Data is reported as mean ± s.e.m. (n=5). The gradual decrease in ls c is accompanied by a gradual increase in TEER to -17% of baseline TEER after 100 min treatment with NPAA. 96 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. < N E s d L 6 10 n M Amiloride (A ) 4 2 45% Inhibition at 70 min T 1 /2 « 1.5 min 0 90 45 60 75 0 15 30 B Time (min) Benzamil (A ) IC S 0 = 0.1 n M 100 O ( 0 -9 8 7 6 5 4 -3 Log [ Inhibitor (M) ] Fig. 4-11: Time-course of ls c inhibition of apical Na+ channel activity with 10 pM amiloride in primary AIC cultured RCrECL from pigmented rabbits (panel A). Cell layers were mounted in modified Ussing-type chamber. Following stabilization of baseline bioelectric properties, either amiloride or benzamil was added to the apical compartment (panel B). Log dose- response curve of benzamil (A) and amiloride (•) for inhibiting short circuit current (ls c ) across RCrECL from pigmented rabbits. Data are expressed as % of maximal ls c inhibition vs. amiloride concentration. Either compound was added to the apical fluid in increasing concentrations in a cumulative manner. Each data point represents the mean ± s.e.m. (n=4- 8) determinations from two or more culture preparations. Where no error bar is shown, it falls within the symbol. Amiloride increased RCrECL TEER by -6% during the treatment period. (Note: 100% in Alsc indicates a maximal inhibition of l^ («50% decrease) by amiloride or benzamil when added to the apical fluid.) 97 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.4. 3 6C 1 Flux Net Cl'-flux was determined by assessing bi-directional fluxes of 3 6 C1 simultaneously ■ y with measuring apparent permeability coefficient of H-mannitol across short- circuited RCrECL in the presence and absence of apically-applied 1 mM NPAA (Table 4-5). NPAA inhibited Isc and PD by 46.2±3.5% and 38±4% (n=4-6), respectively. The reduction in Isc by NPAA in this set of data is not significantly different from that shown in Figure 4-10. In contrast, transepithelial resistance (TEER) in the presence of NPAA increased by 17±9%, indicating NPAA blockage 'yzr of ion conductive pathway. At baseline conditions, Cl flux in the basolateral-to- apical (b->a) direction was 37% higher than that in the apical-to-basolateral (a->b) direction, yielding net CP secretion in the b->a direction of 0.046±0.009 pEq/cm2 /hr (n=10) (p<0.05, as compared to zero). NPAA significantly decreased b->a Cl'flux by 46.5+5% without affecting the a->b Cl'flux (n=4-6), yielding zero, net Cl' transport. Integrity of paracellular routes was assessed by mannitol Pa p p during the flux study. There was no significant difference in mannitol fluxes in the a->b and b- >a directions in the presence of NPAA. 1.5. 2 2 Na Flux , - y - y Net Na -absorption was determined by assessing bi-directional fluxes of Na (with simultaneous measurements of apparent permeability coefficient of 3 H-mannitol) across RCrECL in the presence and absence of 10 pM amiloride (Table 4-6). No significant difference in mannitol Pa p p before and after the addition of amiloride in Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the apical fluid was observed. Amiloride inhibited Isc by 46±2.8% (n=4-8), while TEER increased by -6.4%. At baseline conditions, 2 2 Na flux in the a->b direction was 44±6.8% (n=4-8) higher than that in the b->a direction, yielding net Na+ absorption in the a->b direction of 0.062 + 0.006 pEq/cm2 /hr (n=4-8) (p<0.05 as compared to zero). No significant difference in mannitol fluxes in the a->b and b->a 00 directions was found. Amiloride significantly decreased the a->b Naflux by 43±8% (n=4-8) without affecting the b->a flux, yielding zero net Na+ transport. There was no significant change in mannitol fluxes in either direction in the presence of amiloride. 2. Comparison of Bioelectric Data on RCrECL Derived from Albino Rabbits Bioelectric parameters (TEER, PD and Isc) as well as the day reaching confluency (day 7-8) for RCrECL obtained from albino rabbits were similar to the respective parameters measured for those from pigmented rabbits. Some key ion transport properties were also evaluated in RCrECL obtained from albino rabbits for comparison (Table 4-7). The relative inhibition by various pharmacological agents was similar in both cultures except for basolateral K+ channel and apical Cl‘-channel activities. The K+ channel response was significantly greater (38±1.8%; n=3) with a T1 /2 of 20 sec for comeal cultures of albino rabbits compared to those of pigmented rabbits (22+1.7%; n=4) with a T 1/2 of 54 sec. On the other hand, the Cl'-channel response was significantly smaller (36±1.6%) with a T1 /2 of 25 min for comeal 99 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cultures of albino rabbits compared to those of pigmented rabbits (53±3.0%; n=5) with a T1/2 of 20 min. Table 4-5: Cl'-Flux and Apparent Permeability (Pa p p) of Mannitol Across AIC Cultured RCrECL From Pigmented Rabbits in the Absence and Presence of 1 mM NPAA in the Apical Fluid. Control NPAA PD (mV) TEER (kQ.cm2 ) ls c (pA/cm2 ) 16.4 + 1.25 4.71 +0.15 3.46 ± 0.27 10.1 ± 1.15a 5.39 ± 0.39a 1.86 ± 0.12a Mannitol Pa p p (x 10'7 cm/sec) AB BA 0.83 + 0.10 1.10 ± 0.12 0.72 + 0.10 0.96 + 0.08 p Net r app 0.27 ±0.12 0.24 ±0.10 Cf fluxes (pEq hr/cm2 ) j A B j B A 0.077 ± 0.008 0.123+ 0.007b 0.094 ± 0.004 0.067 ± 0.005a j N e t 0.046 ± 0.008° -0.027 ±0.010° Transport of 36CI (0.5 pCi/ml) and 3H-D-mannitol (10 pCi/ml) w as studied in the ab sen ce and presence of apical 1 mM NPAA under short-circuit conditions. AB = apical-to-basolateral direction. BA = basolateral-to-apical direction, jnet _ j A _ _ j A B > o; net secretion. Values represent m ean ± s.e.m . (n = 4-6) from two or more culture preparations. a Significantly different from control (p<0.001). b Significantly greater than that in the AB direction (p<0.001). 0 Significantly different from zero (p<0.001). 100 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4-6: Na+ -Flux and Apparent Permeability (Pa p p ) of Mannitol Across AIC Cultured RCrECL From Pigmented Rabbits in the Absence and Presence of 10 pM Amiloride in the Apical Fluid. Control Amiloride PD (mV) TEER (kQ.cm2 ) ls c (pA/cm2 ) 17.1 ±1.58 4.40 ± 0.23 3.82 ± 0.25 8.81 ±1.13a 4.12 ±0.33 2.06 ± 0.17a Mannitol Pa p p (x 10'7 cm/sec) AB BA 0.72 ± 0.09 0.89 ± 0.05 0.78 ± 0.06 0.85 ± 0.05 p Net lapp 0.17 ±0.09 0.07 ± 0.06 Na+ fluxes (pEq/hr/cm2 ) j A B jBA 0.140 ±0.006 0.078 ± 0.006b 0.086 ± 0.007 a 0.083 ± 0.007 JN e t -0.062 ± 0.0060 -0.003 ± 0.007 Transport of 22Na (1 pCi/ml) and 3H-D-mannitol (10 pCi/ml) w as studied in the absence and presence of apical 10 pM amiloride under short-circuit conditions. AB = apical-to-basolateral direction. BA = basolateral-to-apical direction. jn e t _ j A _ jAB < q . n e { a [jS o r p (jo n Values represent m ean ± s.e.m . (n = 4-8) from two or more culture preparations. a Significantly different from control (p<0.001). b Significantly greater than that in the AB direction (p<0.001). 0 Significantly different from zero (p<0.005). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4-7: Effect of Pharmacological Agents on ls c Across Primary AIC Cultured RCrECL From Albino Rabbits. % Inhibition Function Excised Tissue Albino RCrECL T 1 /2 Na+ /K+ -ATPase (B) 90** (rabbits, frogs) 85 ±0.70 3 37 min Na+ - channel (A) Present3 (human, bovine, rat) 40 ± 0.93 2 min Cl‘ -channel (A) ~ 50b (rabbits) 36 ± 1.60* 25 min Na+ (K+ )CI'-cotransporter (B) 15±5C (rabbits) 15 ±3.50 5 min K+ -channel (B) Presentb (rabbits) 38 ±1.80* 20 sec Values are m ean + s.e.m ., n = 4-6. Ti/2 = drug inhibition of lS c at half-time. A=drug added to the apical fluid; B=drug added to the basolateral fluid. * Significantly different from that corresponding to RCrECL from pigmented rabbits (p<0.01). " Candia (1972) Am. J. Physiol. 223:1053-1057; Klyce and W ong (1977) J. Physiol. 266:777-799. 3 Midelfart (1987) Pflugers Arch. 408:243-248; Mirshahi etal. (1999) Exp. Eye Res. 69:21-32; Mirshahi etal. (2001) Ophthalmic. Res. 33:7-19. b Klyce (1985) Curr. Eye Res. 4(4): 323-331; R ae etal. (1990) Invest. Ophthalmol. Vis. Sci. 31:1799- 1809. c Bonanno etal. (1989) Am. J. Physiol. 257: C290-296. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C. IN VITRO DRUG TRANSPORT PROCESSES 1. Passive Transcellular Transport 1.1. Effect of Lipophilicity on Drug Transport Across RCrECL The effect of drug lipophilicity was examined by evaluating the in vitro transport characteristics of a series of P-adrenergic drugs that vary in the log octanol/pH 7.4 buffer partition coefficient (logP), by about 4 orders of magnitude with similar molecular weights (-300) and pKa’s (-9.2) (Table 3-1). The permeability coefficient of each P-blocking agent was obtained by linear regression of the steady- state flux. HPLC analysis of the transported samples revealed no evidence of metabolite formation for all evaluated P-adrenergic drugs except levobunolol, which resulted in a fluorescence detectable metabolite, dihydrolevobunolol. Levobunolol and dihydrolevobunolol were assayed by a dual UV/Vis and fluorescence detection set up with the HPLC unit. All evaluated p-adrenergic drug permeability across RCrECL was evaluated in both AB and BA directions as a function of lipophilicity as a measure of establishing the passive transcellular mechanism of transport (Table 4-8). The AB Papp’s ranged from 0.041 x 10'6 cm/sec for atenolol to 12.9 x 10'6 cm/sec for betaxolol, about a 315-fold difference. Similarly, the BA Papp’s ranged from 0.037 x 10’6 cm/sec for atenolol to 12.8 x 10"6 cm/sec for betaxolol, about a 346-fold difference. The trend in the influence of increasing solute logP on the corneal epithelial Papp in both AB and BA direction, are best described by a sigmoidal relationship (Fig. 4-13)(r2 = 0.98, p<0.01 for AB and r2 = 0.98, p<0.01 for BA 103 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. transport). From the nonlinear least square curve fitting, the parameter estimates of the minimum Papp was 0.30 ± 1.44 (x 10'6 cm/sec), the maximum Papp was 13.65 ± 3.0 (x 10~ 6 cm/sec), the logP at the half-maximal Papp was 2.58 ± 0.21, and the transitional slope representing the gradual increase in Papp between Pappmin and Pappmax was 0.27 ± 0.22. Previous studies using isolated tissue cornea from two investigators (Table 4- 8) were also well described by a sigmoidal relationship (Fig. 4-13) for all or most of the (3-adrenergic evaluated in the cultured RCrECL. The nonlinear least square curve fitting values represent predicted permeability’s ± standard error of the parameter estimates are listed in Table 4-10 for permeability values reported in corneal tissues. The logP at Papp5 0 value for cultured RCrECL is comparable to the reported literature values for various primary culture models (Table 4-11). Although most of the compounds tested were absorbed by passive diffusion (as seen by similar AB and BA permeability values), sotalol, acebutolol, labetalol and propranolol resulted in a net transport in the BA direction, whereas, timolol and levobunolol resulted in a net AB transport. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 4-8: Comparison of Permeability Coefficient for the Permeation of p-Adrenergic Drugs Across Excised Cornea vs. AIC Cultured RCrECL. Compound LogPa AIC Cultured RCrECL P a p p (x1 O '6 cm/sec) Excised Tissue Pa p p b (x 10'6 cm/sec) AB BA Tear Side -> Aqueous Humor side Sotalol -0.62 0.087 ± 0.0096 0.48 ± 0.09* 1.60 (±0.4)°; 0.7d Atenolol 0.16 0.041 ± 0.002 0.037 ± 0.01 0.67 (± 0.10)c ; 1.1d Nadolol 0.93 0.15 ±0.044 0.13 ±0.025 1.03 (±0.12)c; 1.4d Pindolol 1.75 1.17 ± 0.14 0.80 ±0.019 10.4 (±0.40)d Acebutolol 1.77 0.047 ± 0.004 0.165 ±0.009* 0.85 (±0.06)c; 1.1 (±0.06)d Metoprolol 1.88 2.57 ± 0.03 2.46 ±0.10 22 (± 1.6)°; 27.9 (±1.3)d Timolol 1.91 1.85 ± 0.075 1.07 ± 0.034* 11.7 (±1.3)c; 12.3 (± 0.7)d Levobunolol 2.40 4.93 ± 0.43 3.99 ± 0.25* 16.4 (± 1.4)°; 22.8 (± 0.65)d; 24.3 (± 0.07)e Labetalol 2.55 0.16 ±0.03 0.44 ± 0.03* 14.3 (±1.0)d Alprenolol 2.61 11.7 ± 0.40 9.79 ± 0.70 28.6 (±1.1)d Propranolol 3.21 12.8 ±0.33 14.7 ±0.40* 47.6 (±1.7)c ; 34d Betaxolol 3.44 12.9 ±0.08 12.8 ±0.05 27 d a log octanol/ pH 7.4 buffer partition coefficient. b m ean Papp ± s.e.m ., n=4-12. 0 Ref. From Huang etal., 1983 (n=4-8, m ean ± s.d .) d Ref. From W ang etal., 1991 (n=6-12, m ean ± s.e.m .). e Ref. From Ashton etal., 1991. * Significantly different from AB direction (p<0.05). AB = apical-to-basolateral direction. BA = basolateral-to-apical direction. Each (3-blocker w as dosed with 1 mM nonradiolabeled drug except metoprolol (1 mM + 1 pCi/ml) and propranolol (0.5 mM + 0.5 pCi/ml). 0 1 1 10 - o X a a. ( 0 a. 4 -1 0 1 5 2 3 LogP 15-| o 0 < / > 1 10- < 0 o T — X a a ra 0. 5 2 -1 1 0 3 4 LogP Fig. 4-12: Relationship between (A) apical and (B) basolateral permeability across AIC cultured RCrECL of p-adrenergic drugs as a function of LogP. The in vitro permeability was carried out in both AB (■ ) and BA (□) direction. Key in increasing order lipophilicity: 1- sotalol, 2-atenolol, 3-nadolol, 4-pindolol, 5-acebutolol, 6-metoprolol, 7-timolol, 8-levobunolol, 9-labetalol, 10-alprenolol, 11-propranolol, 12-betaxolol. The effective half-maximal LogP for apical and basolateral were 2.58 ± 0.21 (^=0.97) and 2.61 ± 0.073 (^=0.98), respectively. The effective half-maximal LogP for excised tissue permeability was found to be 2.90 ± 0.27 (^=0.97). Where no error is shown, it falls within the symbol. 106 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4-9: Parameters for Sigmoidal Relationship Between p-Adrenergic Apparent Permeability (Pa p p ) Across In Vitro Cultured RCrECL with its Corresponding Lipophilicity. Parameters Cultured RCrECL Pa p p (x 10‘6 cm/s) CE (AB) CE (BA) Pappm in 0.30 ± 1.44 0.80 ± 0.75 Pappm ax 13.65 + 3.02 13.34 ±1.51 LogP at Papp5 0 2.58 ± 0.21 2.61 ± 0.07 Transitional slope 0.27 ± 0.22 0.12 ±0.08 r2 0.9711 0.9817 Papp = Pappm in + Pappm ax - Pappm in 1 + exp [(LogP at Papp5 0 - LogP)/Slope)] Pappm in represent the minimum predicted permeability that estimates the paracellular transport of hydrophilic molecules, Pappm a x is an estimate of the maximal transepithelial permeability that represents the transcellular permeability of lipophilic molecules, LogP at Papp5 0 is the logP corresponding to the half maximal permeability, and the slope is the transition between Pappm jn and Pappm a x , respectively. Note: The transitional slope represents the gradual increase in Pa p p . Parameters for the sigmoidal relationship between Papp and LogP were performed using GraphPad Prism version 3.00 for Windows, GraphPad Software, San Diego California USA, www.graphpad.com''. CE = corneal epithelium. AB = apical-to-basolateral BA = basolateral-to-apical Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 40-. (A ^ 30- o o 20 - t — X C l Q . ^ 10- 4 1 2 3 0 LogP 75-, 50- £ 2 5 - 1 1 2 3 0 4 5 LogP Fig. 4-13: Relationship between excised corneal tissue permeability of p-adrenergic blockers as a function of LogP. Graph A represent reported values obtained from Wang et a!., 1991, whereas graph B reported values were obtained from Huang et a!., 1983. Key in increasing order lipophilicity: 1-sotalol, 2-atenolol, 3-nadolol, 4-pindolol, 5-acebutolol, 6-metoprolol, 7- timolol, 8-oxprenolol, 9-levobunolol, 10-labetalol, 11-alprenolol, 12-bevantolol, 13- propranolol, 14-betaxolol, 15-bufuralol, 16-penbutolol. The effective half-maximal LogP for graph A and graph B were 2.02 ± 0.38 (r^O.73) and 2.45 + 0.16 (^=0.91), respectively. Where no error is shown, it falls within the symbol. Note: Graph A and B data values are represented in mean ± s.e.m. and mean ± s.d., respectively. 108 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4-10: Parameters for Sigmoidal Relationship Between p-Adrenergic Apparent Permeability (Pa p p ) Across Excised Cornea with its Corresponding Lipophilicity. Corneal Tissue P a p p (x 10'6 cm/s) Parameters Wang etal., 1991 Huang etal., 1983 Pappm in 0.07 ± 5.55 2.29 ± 4.35 Pappm a x 30.9 ±9.12 52.9 ± 5.68 LogP at Pappso 2.02 ± 0.38 2.45 ±0.16 Transitional slope 0.44 ± 0.39 0.30 ±0.14 r2 0.729 0.914 Papp = Pappm in + Pappmax- P a p p m in 1 + exp [(LogP at Papp5 0 - LogP)/Slope)] Pappm in represent the minimum predicted permeability that estimates the paracellular transport of hydrophilic molecules, Pappm a x is an estimate of the maximal transepithelial permeability that represents the transcellular permeability of lipophilic molecules, LogP at Papp5 0 is the logP corresponding to the half maximal permeability, and the slope is the transition between Pappm in and Pappm a x , respectively. Note: The transitional slope represents the gradual increase in Pa p p . Parameters for the sigmoidal relationship between Papp and LogP were performed using GraphPad Prism version 3.00 for Windows, GraphPad Software, San Diego California USA, www.graphpad.com''. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 4-11: Comparison of Effective Half-Maximal Papp5 0 for p-Adrenergic Drugs in Various Tissue and Culture Models. Species Tissue/ Culture TEER (ohm.cm2 ) LogP at Papp5 0 References Pigmented rabbit Corneal tissue n.d. 2.45 Huang etal., 1983 Pigmented rabbit Corneal tissue n.d. 2.02 Wang etal., 1991 Albino rabbit Primary corneal epithelial cells (LCC) -144 1.93 Kawazu et at., 1998 Pigmented rabbit Corneal epithelial culture (AIC) > 3,000 2.58 Chang-Lin, Thesis Pigmented rabbit Conjunctiva tissue n.d. 1.63 Wang etal., 1991 Pigmented rabbit Conjunctiva tissue > 1,000 1.00 Yang etal., 2000 Pigmented rabbit Conjunctival epithelial culture (LCC) 1090 + 0.07 1.20 Saha etal., 1996 Pigmented rabbit Conjunctival epithelial culture (AIC) > 1,000 1.20 Yang etal., 2000 Pigmented rabbit Conjunctival epithelial culture (LCC) > 1,000 1.30 Yang etal., 2000 Rat Alveolar epithelial culture -2,000 1.95 Saha etal., 1994 Pigmented rabbit Tracheal epithelial culture -1,200 2.08 Mathias etal., 1996 o 1.2. In Vitro (Culture) -Ex Vivo (Tissue) Correlation for p-Adrenergic Drugs Permeability of P-blockers across tissue cornea (ex vivo) has been reported by Huang et al. (1983) and Wang et al. (1991). Plotting tissue permeability values from either investigator as a function of permeability values from AIC cultured RCrECL resulted in good linear correlation as seen in Fig. 4-14 (r2 = 0.701 and r2 = 0.925). 5 0-| F T = 0.701 E 4 0 - < 0 20 - 10 - 1,2, 3,5 6 9 12 15 0 3 Cell Culture Pa p p (x10'6 cnVsec) 50 n R 2 = 0.925 40- to 3 0 - " 20 - 10 - 1 ,2 ,3 , 5 3 6 9 12 15 0 Cell Culture Pa p p (x106 cnVsec) Fig. 4-14: In vitro (culture)- ex vivo (tissue) correlation for (3-blockers. Tissue Pa p p values were obtained from (A) Wang etal., 1991 and (B) Huang etal., 1983. Key in increasing order lipophilicity: 1-sotalol, 2-atenolol, 3-nadolol, 4-pindolol, 5-acebutolol, 6-metoprolol, 7- timolol, 8-levobunolol, 9-labetalol, 10-alprenolol, 11-propranolol, 12-betaxolol. I l l Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2. Active TranscelluJar Transport 2.1. Permeability of Model Solutes Across AIC and LCC Cultured RCrECL The permeability characteristics of the AIC and LCC cultured corneal epithelial cell layers was evaluated using various model substrates to determine their predominate transport process. Bidirectional permeability of each model substrate was determined in both LCC and AIC cultured RCrECL under 37°C and 4°C. At peak bioelectric parameters (days 7-8), LCC cultures (< 1 kQ.cm2 ) were at least 3-fold lower in TEER compared to AIC cultures (>3.5 kQ.cm2). In addition, placing either cell cultures at 4°C increases TEER by > half the original value before the start of the experiment (-2.22 and 5.88 kQ.cm for LCC and AIC cultures, respectively). 2.1.1. Mannitol Transport The paracellular permeability across cultured RCrECL was determined using a widely employed reference marker, mannitol. The cumulative appearance of mannitol in the receiver fluid was linear for the 4 hr duration of the experiment for both LCC and AIC cultured RCrECL. Permeability in the AB direction was not significantly different from that in the BA direction for both culture systems. Permeability in the AB direction at 37°C was 2-fold higher compared to 4°C for both LCC and AIC cultures. This is reflective of the ~2-fold higher TEER observed under 4°C compared to 37°C. Permeability of LCC for either 4°C or 37°C resulted in -3.6 to 4-fold higher compared to AIC culture. This is also reflective of the TEER for AIC compared to LCC cultures. A plot of 1/TEER as a function of Papp shown Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in Fig. 4-15 indicates that the extent of mannitol transport is linearly dependent on 1/TEER (r2 = 0.9856). 2.1.2. L-Arginine Transport L-Arginine, a model substrate for the amino acid transporter was used to determine the transport process of this solute in cultured RCrECL. The cumulative appearance of L-arginine in the receiver fluid was linear for the 4 hr duration of the experiment for both LCC and AIC cultured RCrECL. Permeability in the AB direction was significantly higher by -15 and 4-fold from that in the BA direction for LCC and AIC cultured RCrECL, respectively. Permeability in the AB direction at 37°C was significantly higher by ~16-fold and 13-fold compared to 4°C for LCC and AIC cultured RCrECL, respectively. Unlike mannitol that exhibited bi-directional symmetry in transport, L-arginine resulted in a net vectorial preference for AB transport in both culture systems. A plot of 1/TEER as a function of Papp shown in Fig. 4-15 indicate that the extent of L-arginine transport is not linearly dependent on 1/TEER (r = 0.3724) indicating the possibility of the involvement of other transport mechanism(s) (e.g. amino acid transporter) besides the passive transport mechanism. 2.1.3. L-Carnitine Transport Like L-Arginine, L-camitine is also an amino acid transporter substrate (e.g. the B0 ,+ system) but it is also considered an organic cation substrate (for the OCTN2 type). The cumulative appearance of L-camitine in the receiver fluid was linear for the 4 hr 113 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. duration of the experiment for both LCC and AIC cultured RCrECL. Permeability in the AB direction was significantly higher by ~2-fold from that in the BA direction for LCC cultured RCrECL and no difference was observed for the AIC cultured cells. Similarly, permeability in the AB direction at 37°C was significantly higher by ~2-fold compared to 4°C for LCC culture and no significant difference was observed in AIC cultured cells. A plot of 1/TEER as a function of Pa p p shown in Fig. 4-15 indicates that the extent of L-camitine transport is linearly dependent on 1/TEER (r2 = 0.7469). The fact that permeability across AIC cultured cells resulted in values to the same extent as mannitol may indicate that L-camitine is mainly transported through the paracellular pathway whereas, AB and BA transport of L- camitine was ~6-fold and 3-fold higher compared to mannitol transport, respectively. 2.1.4. Cyclosporin A Transport The bidirectional transport of cyclosporin A (CsA) was determine across cultured RCrECL in order to evaluate the possible role of the 170 kDa p-glycoprotein efflux pump in restricting comeal absorption of lipophilic drugs. The cumulative appearance of CsA in the receiver fluid was linear for the 4 hr duration of the experiment for both LCC and AIC cultured RCrECL. At a donor concentration of 100 nM H-CsA, CsA flux exhibited directional asymmetry to the same extent for both LCC and AIC RCrECL culture. The apparent permeability coefficient (Pap p , cm/s) in either direction was estimated from the slope of the linear portion of a plot Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of cumulative amount (pmoles) of H-CsA appearing in receiver chamber as a function of time. The BA Pa p p of 0.13 ± 0.02 x 10'6 was 1.8-fold higher than in AB direction (0.07 ± 0.002 x 10'6 cm/s) (p<0.05) (Table 4-12) for LCC cultured RCrECL as well as AIC cultured cells. The results indicate that the leakiness of LCC RCrECL did not affect the passive penetration of CsA. In contract, the bidirectional transport of mannitol, a paracellular marker, was symmetric at 100 pM (Table 4-12) for both LCC and AIC culture conditions. Interestingly, there was a difference of 4- fold between LCC and AIC mannitol permeability. This indicates that the leakiness of LCC culture condition resulted in 4-fold higher paracellular permeability compared to AIC culture condition. 2.1.5. Guanidine Transport Guanidine is known to be a substrate for the organic cation transporter (of the OCT type). The cumulative appearance of guanidine in the receiver fluid was linear for the 4 hr duration of the experiment for both LCC and AIC cultured RCrECL. Permeability in the AB direction was not significantly different from that in the BA direction for both culture systems. Permeability in the AB direction at 37°C was 2-3- fold higher compared to 4°C for both LCC and AIC cultures. A plot of 1/TEER as a function of Pa p p shown in Fig. 4-15 indicates that the extent of guanidine transport is linearly dependent on 1/TEER (r2 = 0.9335). Although, no net directional transport was observed for both culture systems as well as there is a dependence on TEER, the actual permeability values are -9-10 fold higher compared to mannitol. 115 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.1.6. Lactic Acid Transport Lactic acid, a model substrate for the monocarboxylic acid transporter (MCT) was used to determine the transport process of this solute in cultured RCrECL. The cumulative appearance of lactic acid in the receiver fluid was linear for the 4 hr duration of the experiment for both LCC and AIC cultured RCrECL. Permeability in the BA direction was significantly higher by -1.33 and 8-fold from that in the AB direction for LCC and AIC cultured RCrECL, respectively. Permeability in the AB direction at 37°C was significantly higher by -2.5-fold and 3.6-fold compared to 4°C for LCC and AIC cultured RCrECL, respectively. Unlike mannitol that exhibited bi directional symmetry in transport, lactic acid resulted in a net vectorial preference for BA transport in both culture systems. A plot of 1/TEER as a function of Pa p p shown in Fig. 4-15 indicate that the extent of lactic acid transport is not linearly dependent on 1/TEER (r2 = 0.506) indicating the possibility of the involvement of other transport mechanism(s) (e.g. MCT) besides the passive transport mechanism. 2.1.7. Uridine Transport Uridine, a model substrate for the nucleoside transporter (NT) was used to determine the transport process of this solute in cultured RCrECL. The cumulative appearance of uridine in the receiver fluid was linear for the 4 hr duration of the experiment for both LCC and AIC cultured RCrECL. Permeability in the AB direction was significantly higher by 6-fold from that in the BA direction for LCC and no significant difference was observed for AIC culture. Permeability in the AB 116 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. direction at 37°C was significantly higher by 10-fold and 9-fold compared to 4°C for LCC and AIC cultured RCrECL, respectively. Unlike mannitol that exhibited bi directional symmetry in transport, uridine resulted in a net vectorial preference for AB transport in LCC culture but no in AIC culture. A plot of 1/TEER as a function of Pa p p shown in Fig. 4-15 indicate that the extent of uridine transport is not linearly dependent on 1/TEER (r2 = 0.4604) indicating the possibility of the involvement of other transport mechanism(s) (e.g. NT) besides the passive transport mechanism. These results also indicate that culture condition (AIC vs. LCC) may modulate transporter expression. 2.2. Evidence for Oligopeptide Transporter on the Basolateral Membrane 2.2.1. Effect of pH on Gly-Sar Transport The transepithelial transport of gly-sar was determined at pH 6.0 and pH 7.4 in AIC cultured RCrECL. Transport in the AB direction was significantly lower by 2-fold compared to BA direction and this net BA directional transport does not seem to be affected by dosing with either pH 6.0 or pH 7.4. In all conditions, bioelectric properties remained similar throughout the duration of the study. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4-12: Summary of Permeability Coefficient (Pa p p ) of Various Model Solutes Across AIC and LCC Cultured RCrECL. Compounds LCC Culture P a p p (x1 O '6 cm/s) AIC Culture P a p p (x1 O '6 cm/s) 4°C 37°C 4°C 37°C 1 4 C-Mannitol AB BA 0.096 ± 0.006 n.d. 0.18 ±0.01 0.20 ± 0.01 0.024 ± 0.005 n.d. 0.050 ± 0.008 0.048 ± 0.005 3 H L-Arginine AB BA 0.85 ± 0.033 n.d. 13.9 ±0.65 0.93 ± 0.37* 0.19 ±0.044 n.d. 2.39 ± 0.068 0.63 ± 0.026* 3 H L-Camitine AB BA 0.66 ± 0.03 n.d. 1.09 ±0.034 0.63 ± 0.048* 0.054 ± 0.007 n.d. 0.048 ± 0.002 0.038 ± 0.002 3 H-CsA AB BA 0.065 ± 0.004 0.058 ±0.012 0.073 ± 0.002 0.129 ±0.015* 0.045 ± 0.002 0.040 ± 0.006 0.077 ± 0.007 0.125 ±0.023* 1 4 C-Guanidine AB BA 1.02 + 0.08 n.d. 1.6 ± 0.15 2.1 ±0.14 0.13 ±0.005 n.d. 0.43 ± 0.03 0.50 ± 0.04 1 4 C-Lactic Acid AB BA 0.48 ± 0.06 0.54 ± 0.04 1.2 ±0.07 1.6 ± 0.13* 0.053 ± 0.005 0.046 ± 0.006 0.19 ±0.02 1.50 ±0.19* 3 H-Uridine AB BA 0.18 ±0.02 n.d. 1.80 ± 0.30 0.30 ± 0.005* 0.031 ± 0.003 n.d. 0.29 ± 0.04 0.20 ± 0.01 Transport of 14C-mannitol (5 pCi/ml = 100 pM), 3H-arginine (2 pCi/ml +100 pM unlabeled arginine = 100 pM), 3H-carnitine (1 pCi/ml = 12.4 nM), 3H-CsA (0.8 pCi/ml = 100 nM), 14C-guanidine (1 pCi/ml = 18.2 pM), 14C-lactic acid (1 pCi/ml = 18.2 pM) and 3H-uridine (1 pCi/ml = 23 pM) acro ss LCC and AIC RCrECL. Values are m ean ± s.e.m ., sam ple size (n) = 4-11. n.d. = not determ ined AB = apical-to-basolateral transport. BA = basolateral-to-apical transport. Statistical significance w ere determ ined by One-way ANOVA followed by Student-Newman-Keuls multiple com parison tests w ere used to determ ine statistical difference am ong culture condition groups (LCC and AIC at 4°C and 37°C) and p<0.05 w as considered statistically significant. * Significantly different com pared to AB transport. 118 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4-13: Effect of pH on Transepithelial Permeability of [3 H]Gly-Sar Across AIC Cultured RCrECL. Dosing Transepithelial transport pH Direction Pa p p (x 10'6 cm/sec) pH 6.0 AB 0.13 ±0.017 BA 0.25 ±0.016* pH 7.4 AB 0.14 ±0.006 BA 0.28 ± 0.040* Each data point represents the m ean ± s.e.m . for sam ple size (n) = 4-7. Asterisk (*) indicates statistical difference from AB permeability (P<0.05). Statistical analyses w ere performed by one-way analysis of variance followed by Student’s multiple com parison test. AB and BA denote apical-to-basolateral and basolateral-to-apical directions, respectively. Papp denotes apparent permeability coefficient. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mannitol 0.4 - i R = 0.9856 0.2 - a a 0.25 0.5 0.75 1/TEER (1/kohms.cm2 ) Guanidine R = 0.9335 o . 0.25 0.5 0.75 1/TEER (1/kohms.cm?) Lactic Acid R = 0.506 Q. 0.25 0.5 1/TEER (1/kohms.cm2 ) 0.75 Uridine R = 0.4604 a. 0.5 - a. 0.25 0.5 0.75 1/TEER (1/kohms.cm2 ) L-Arginine 15 - i R = 0.3724 0.25 0.5 0.75 1/TEER (1/kohms.cm2) L-Carnitine R = 0.7469 0.5 - 0.25 0.5 0.75 1/TEER (1/kohms.cm2 ) f 0.20 i o 0.15 - T - X 0.10 - o. a 0.05 - < 0 a. 0.00 4 R = 0.2246 0 0.25 0.5 0.75 1 1/TEER (1/kohms.cm2 ) Fig. 4-15: Relationship between drug permeability (Pa p p ) and transmembrane electrical conductance for LCC and AIC cultured RCrECL under 4°C and 37°C. Corneal cell layers were cultured to confluency (day 7-8) and transport of each solute was carried out in LCC and AIC culture conditions at 4°C and 37°C. Data points are means ± s.e.m., n = 4-11. 120 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.2.2. Time Course and Temperature Dependence [3 H]gly-sar uptake into the apical (Fig.4-16A) and basolateral (Fig. 4-16B) corneal epithelial cell layers was time and temperature dependent. At 37°C, uptake from apical and basolateral fluid did not show difference. Only after 30 min uptake did basolateral begin to show higher uptake compared to apical. At 4°C, apical and basolateral uptake process was significantly lower (by 13 to 56% and 78 to 90%, respectively) than that at 37°C at all time points. 2.2.3. Effect of Extracellular pH on Gly-Sar Uptake Figure 4-17 shows the effect of extracellular pH in the range of 5.0-8.0 on [3 H]gly- sar uptake by the apical and the basolateral membrane. [3 H] gly-sar uptake from the apical fluid was found to be slightly higher at pH 6.0 compared to other tested pHs. Similar results were also observed for basolateral uptake. 2.2.4. Energy Dependence Apical [3 H]gly-sar uptake was not affected by either 2,4-dinitrophenol (DNP) or sodium azide (Fig. 4-18A). On the contrary, basolateral gly-sar uptake at 37°C was inhibited by 2,4-DNP (by 28 + 1.4%) and sodium azide (by 49 + 1.9%) (Fig. 4-16B). At 4°C, basolateral gly-sar uptake was inhibited by 91 ± 0.7%. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.2.5. Concentration Dependence The [3 H]gly-sar uptake by the basolateral membrane was saturable within the concentration range of 0.05-10 mM (Fig.4-19B) compared to apical uptake (Fig.4- 19A). The specific uptake was calculated by subtracting the nonspecific uptake, which was estimated in the presence of 4°C, from the total uptake. The apparent Michaelis-Menten constant (Km ) value for the basolateral peptide transporter was 1.49 ± 0.06 mM and a maximum uptake rate (V m ax) of 104 ± 0.93 pmol/mg protein/min (r2 = 0.99). 2.2.6. Substrate Specificity Apical uptake of [3 H]gly-sar by RCrECL was not inhibited by 20 mM amino acids such as L-histidine and P-alanine; dipeptides such as L-camosine, gly-sar and gly- phe; tripeptide such as thyrotropin-releasing hormone (TRF1; L-pyroglutamyl-L- histidyl-L-prolinamide) (Fig. 4-20). In contrast, basolateral uptake of gly-sar was inhibited by L-camosine (by 68 ± 1.2%), gly-sar (by 75 ± 2.9%), gly-L-phe (by 84 ± 0.9%), gly-D-phe (by 52 ± 1.3%) and TRH (by 39 + 2%). Stereoselectivity of gly- sar uptake was assessed by the inhibitory effect of gly-L-phe being three times more potent than gly-D-phe. 2.2.7. Differential Recognition of p-Lactam by RCrECL Figure 4-21 shows the inhibition of [3 H]Gly-sar uptake by basolateral peptide transporter in the absence and presence of increasing concentrations (0.05 to 10 mM) 122 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of cyclacillin and cefadroxil. Uptake of gly-sar in the absence of inhibitors was taken as 100% (2.04 ± 0.05 pmol/mg protein/20 min). The estimated inhibition constant (Ki) values for cyclacillin and cefadroxil were estimated to be 121 and 1739 pM, respectively. Cyclacillin was found to be 14-fold more potent than cefadroxil indicating the possible expression of a PepTl-like transporter on the basolateral membrane. 8 o 6 o 1 5 I * 4 m < > O 0 | 2 0 40 60 0 20 Time (min) 8 o Q. o > 4 | o E 9 C L 2 0 0 60 20 40 Time (min) Fig. 4-16: Time course of apical (A) (apical pH 6.0, basolateral pH 7.4) and basolateral (B) (apical pH 7.4, basolateral pH 7.4) uptake of 3 H-gly-sar (5 pCi/ml = 1.25 pM) by AIC cultured RCrECL. Each point represents the mean ± s.e.m. value of 3-4 determinations. Where no error is shown, it falls within the symbol. Key: ♦, 37°C (solid lines); 0, 4°C (dashed lines); •, net active uptake (solid lines). 123 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. c E o C V I c ■ 5 4 - 4 o 1 - Q . O ) E o E Q . £ (0 * 4 Q . Z > 3 2 1 0 5.0 6.0 7.4 8.0 c 1 o C \l c ■ 5 4 -4 o i- Q . O ) E 3 $ 0 E a. 1 (0 3 2 1 0 5.0 6.0 7.4 8.0 8.0 Extracellular pH Fig. 4-17: Effect of pH of apical (A) and basolateral (B) bathing fluid on 3 H-gly-sar (5 pCi/ml = 1.25 pM) uptake by AIC cultured RCrECL (receiving fluid pH 7.4) at 37°C. Each bar graph represents the mean ± s.e.m. value of 4 determinations from two cultures. Statistical significance was tested by one way ANOVA. Group means were contrasted for significant differences using the Fisher’s FLSD post hoc test. Statistical significance was set at p<0.05. *p<0.01 comparison between pH 6.0 and pH 7.4 for apical uptake; p<0.05 comparison between pH 6.0 and pH 8.0 for basolateral uptake. 124 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. c I o CM 2.5 2.0 1.5 ■ 2 1.0 c © o 1 - Q . G > E !s ? o E a 2 re ■ * - * a 3 0.5 0.0 Control 0.1 mM 2,4-DNP 2 mM NaN, 4°C 2.5 2.0 1.5 c I o CM c © o 1- Q. O ) E 0 ? o E Q . © 0.5 < o Q . 3 ■i 1.0 0.0 B Control 0.1 mM 2,4-DNP 2 mM NaN, 4°C Fig. 4-18: Effect of metabolic inhibitors on gly-sar uptake (5 pCi/ml = 1.25 pM) by rabbit corneal epithelial cell layers (apical pH 7.4, basolateral pH 6.0). Metabolic inhibitors were applied apically (A) and basolaterally (B) 30 min prior to the gly-sar uptake studies. Each column represents the mean + s.e.m. (n=4). * Significant difference compared to control (p<0.01). 125 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 800-| e E c 600- Q . U ) | 400- o £ 3 3 200 - (0 C L 3 10 8 0 2 4 6 Concentration (mM) 200 - | c E c Km = 1.49 ±0.06 mM Vmax = 104 + 0.93 pmol/mg protein/min a > o X m 150- a. O ) | 100- o E a. 50- (0 4- * a . =) 2 4 6 8 10 0 Concentration (mM) Fig. 4-19: Concentration dependence of [3 H]gly-sar uptake by apical (A) and basolateral (B) membrane of the primary cultured pigmented rabbit corneal epithelial cells. [3 H]Gly-sar (5 pCi/ml = 1.25 pM) uptake was carried out in the presence of an inwardly directed proton gradient (apical pH 7.4, basolateral pH 6.0). Each point represents the mean ± s.e.m. values of 5-6 determinations from two cultures. Overall uptake (■) was corrected for simple diffusion determined at 4°C (□) to yield carrier-mediated uptake (•), which was fitted to Michaelis-Menten kinetics by non-linear regression analysis. Where no error is shown, it falls within the symbol. Insert: Eadie-Hofstee plot of gly-sar uptake after correction for nonsaturable component. V, uptake rate (pmol/mg protein/min); S, gly-sar concentration (mM). 126 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. c E 2.5 i o C M C 2.0 © O Q . o > 1.5 E o 1.0 E >& a > 0.5 (0 C L D 0.0 X A I" t — “—r c E o C \l c '< 5 * - • o i_ Q . O ) E o E > & a > .* (0 4 - > a z > 2.5 2.0 1.5 1.0 0.5 0.0 B t — —r Fig. 4-20: Effect of unlabeled amino acids, excess gly-sar, dipeptides and drugs structurally similar to di/tripeptides at 20 mM final concentration on 5 pCi/ml (1.25 pM) [3 H]gly-sar uptake by (A) apical and (B) basolateral AIC cultured RCrECL. Each column represents the mean ± s.e.m. value of 3-5 determinations from 1 -2 cultures. ‘ Statistical significance were determined by Student’s f test and p<0.005 was considered statistically significant unless indicated otherwise. TRH = thyrotropin-releasing hormone (L-pyroglutamyl-L-histidyl-L- prolinamide). 127 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. c 100 o o * 801 2 re 4- 1 a. 3 i_ re V) > > 0 I CO 601 40 20 =// j g- Cyclacillin K , = 121 pM I T iin iin i m u m Cefadroxil K j = 1739 jliM mp o - i - / / - 0 1 0 1 1 C P 101 1 tf 1 C P 10* 1 0 3 Inhibitor Concentration (^M) Fig. 4-21: Differential recognition of (3-lactam antibiotics by basolateral RCrECL. Uptake of [3 H]gly-sar (5 pCi/ml = 1.25 pM) was measured for 20 min at pH 6.0 in the absence and presence of increasing concentration of cyclacillin or cefadroxil (0.1 to 10,000 pM). Uptake of gly-sar measured in the absence of the inhibitors was taken as 100% (2.04 ± 0.05 pmol/mg protein/20 min). Key: • , cefadroxil; o , cyclacillin. Each point represents mean ± s.e.m. of 3-6 determinations. 128 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.3. Evidence of Nucleoside Transporters in RCrECL 2.3.1. Temperature dependent time course of uridine uptake in RCrECL [3 H]uridine uptake into the apical (Fig. 4-22A) and basolateral (Fig. 4-22B) AIC RCrECL showed temperature dependency at all time points evaluated. At 37°C, uptake from basolateral fluid was 2-fold higher than that from apical fluid. At 4°C, both apical and basolateral uptake process were significantly lower (by 72 to 87%) than that at 37°C at all time points. 2.3.2. Inhibition of uridine uptake in apical and basolateral RCrECL Figure 4-23 shows inhibition of uridine uptake by 4°C, 0.2 mM unlabeled uridine, 1 mM unlabeled uridine, 2 mM unlabeled uridine (Fig. 4-23B), sodium-free condition, effect of 1 pM and 10 pM NBMPR in apical (Panel A) and basolateral (Panel B) AIC RCrECL. RCrECL was pre-incubated with NBMPR and sodium-free condition for 30 min before uptake. As seen in Figure 4-23A, 4°C, Na+ -free, 0.2 mM and 1 mM unlabeled uridine inhibited [3 H]uridine uptake to 75% (baseline). There was no significant NBMPR effect on apical uridine uptake. In Figure 4-23B, 4°C and 2 mM unlabeled uridine inhibited uridine uptake to baseline, whereas, Na+ -free, 0.2 mM and 1 mM inhibited uridine uptake by 25, 20, and 50%, respectively. In addition, NBMPR decreased uridine uptake by 40 and 70% at 1 and 10 pM concentration, respectively. Inhibition of apical uridine uptake by 100 pM and 200 pM purine and pyrimidine by -40 and 60% indicate the presence of nuceloside transporter 129 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. involvement (Fig. 4-24). In addition, uracil and D-ribose at 200 pM concentration did not affect uridine uptake. 2.3.3. Na+ -dependent kinetics in apical and basolateral RCrECL As shown in Fig. 4-24A, Na+ -dependent uridine uptake is present in both the apical (-100%) and basolateral (-25%) membrane of RCrECL. Both apical (Panel A) and basolateral (Panel B) uptake processes consist of both a saturable (Na+ -dependent) and a non-saturable component (Na+ -free condition) (Fig. 4-24B). The I n dependent saturable component (denoted by open circle and a solid line in the figures) was calculated by subtracting the linear non-saturable component from total uptake, and was used to estimate the kinetic parameters of uridine uptake. For apical uptake (0-80 pM), Km was 28.14 ± 8.8 pM and Vm a x was 8.4 ± 1.1 pmol/mg protein/min (Fig. 4-25 A). For basolateral uptake (0-1.5 mM), Km was 0.70 ± 0.2 mM and Vm a x was 646 ± 85 pmol/mg protein/min (Fig. 4-25B). Since five Na+ -dependent nucleoside transport systems have been described based on substrate selectivity, we tested the effect of purine and pyrimidine nucleosides on Na+ -dependent [3 H]uridine uptake in both the apical (Fig. 4-27B) and basolateral (Fig. 4-27A) membranes. All basolateral uridine uptake were carried out in the presence of 0.2 mM NBMPR to avoid any contribution from Na+ -independent transport systems. As shown in Fig. 4-27A, [3 H]uridine uptake was inhibited by -25% by pyrimidine nucleosides, cytidine and thymidine (100 pM). In contrast, the purine nucleosides, guanosine and inosine (100 pM) inhibited [3 H]uridine uptake by 130 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ~ 50 and 75%, respectively. This suggested that the purine nucleoside-selective transport system is the dominant system expressed in the basolateral membrane of the RCrECL. Our results also suggest a minor contribution by the pyrimidine nucleoside-sensitive system or the broadly selective N3-like system. 2.3.4. Basolateral Na+ -Independent Uridine Uptake - NBMPR Dose Response For basolateral Na+ -independent saturable component (Fig. 4-26), Km was 2.08 ± 0.28 mM and Vm a x was 2.4 ± 0.21 nmol/mg protein/min. To test whether the basolateral Na+ -independent nucleoside transporter consisted of both es and ei, the es system was pharmacologically defined as the nucleoside uptake that was inhibited by 100 nM up to 1 pM NBMPR and the NBMPR-insensitive activity (resistant to 100 nM but sensitive to 0.2 mM) as ei uptake. As shown in Fig.4-23, es uptake system contributed -30% and ei uptake system contributed -70% at 10 pM NBMPR. To further confirm that the basolateral RCrECL contained both es (-30%) and ei (-70%) systems, the dose response of NBMPR inhibition of [3 H]uridine uptake was determined (Fig.4-26). This dose-response curve was biphasic with a plateau between 10 nM and 1 pM, consistent with the coexistence of es and ei. If this dose- response curve was dissected into es (sensitive to 100 nM) and ei (resistant to 100 nM) uptake components, which when fit to a two-site competition model, resulted in I C 5 0 values of 4.5 nM and 13 pM, respectively. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.3.5. Molecular Identity of Nucleoside Transporters in Corneal Epithelium Since the Na+ -dependent purine nucleoside-selective N1 (CNT2) and Na+ - independent EI (ENT2) nucleoside transporter systems have been cloned (Accession # AF323951, Wu et al., 2000), RT-PCR was used to confirm the expression of CNT2 and ENT2 (Fig. 4-28). The rabbit intestine cDNA was used as a positive control because both of these nucleoside transporters have been reported in this tissue. As predicted, PCR product of ENT2 (495 bp) was amplified by RT-PCR from RNA isolated from tissue comeal epithelium. In contrast, there was no amplified CNT2 PCR product from tissue comeal epithelium. Therefore, these results complemented our functional studies that the RCrECL expressed ENT2. 2.3.6. Nucleoside Transporter in LCC Cultured RCrECL The presence of a high affinity Na+ -dependent nucleoside transporter appears to be located on the apical membrane and low affinity Na+ -independent nucleoside transporters appears to be located on the basolateral membrane of the AIC cultured RCrECL. The presence of these transporters is probably the reason why there is a lack of net directional uridine transport observed in AIC culture (Table 4-12). The BA permeability value for AIC culture appears to be in the same range as LCC culture but the AB permeability for AIC culture appears to be significantly lower by 6-fold. These results indicate that the nucleoside transporter(s) present on the apical membrane of either LCC or AIC may be different or the transporter(s) expression may have been affected by the culture conditions. Fig. 4-29 shows that uridine Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. apical uptake in LCC cultured RCrECL is completely inhibited by the presence of sodium-free condition as well as in the presence of 1 mM excess unlabeled uridine. In addition, Na+ -dependent uridine uptake is saturable and exhibits high affinity (Km =25 + 8.8 pM) and low capacity (Vm a x = 17 ± 2.6 pmol/mg protein/min) kinetic parameters. The Km for AIC and LCC cultured cells are similar but the Vm a x is 2- fold higher for LCC compared to AIC cultured cells. This indicates that the I n dependent nucleoside transporter on the apical membrane of RCrECL may have been downregulated in AIC cultures or that LCC cultures have upregulated the transporter expression. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ■ § 0.16 Q . g> 0.12 o E S 0.08 = 0.04 © c ;5 = 0.00 120 90 30 60 0 Time (sec) c 0.32 a > o A S 0.24 E o E S 0.16 J § ! (0 H M = 0.08 a > c ' ■ 5 3 0.00 120 0 30 60 90 Time (sec) Fig. 4-22: Time course of (A) apical and (B) basolateral uptake of [3H]uridine uptake (5 pCi/ml = 0.12 pM) by AIC cultured RCrECL. Data points represent mean ± s.e.m.. n = 3-4. Where no error is shown, it falls within the symbol. Key: ■, 37°C (solid line); □, 4°C (solid line); O, net active uptake (dash line). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 0 ) (0 75 0 ) c ' ■ 5 NaMree NBM PR NBM PR Control 4°C 0.2 m M 1 m M (1 pM) (10 pM) uridine uridine 100 75 a. 50 Control 4°C 0.2 m M 1 m M 2 m M NaMree NBM PR NBM PR uridine uridine uridine (1 pM ) (10 pM ) Fig. 4-23: Inhibition of (A) apical and (B) basolateral [3 H]uridine uptake by 4°C, 0.2 mM unlabeled uridine, 1 mM unlabeled uridine, Na+ -free condition, 1 pM and 10 pM NBMPR. The corneal cell layers were preincubated with NBMPR and sodium-free condition for 30 min before uptake. Each bar graph represent means ± s.e.m., n = 4. ‘ Statistical significance was tested by one-way ANOVA. Group means were contrasted for significant differences using the Student-Newman-Keuls multiple comparison test (p<0.05). NBMPR = nitrobenzylthioinosine. 135 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 4-24: Inhibition of [3 H]uridine apical uptake (5 pCi/ml = 0.12 pM) by purines and pyrimidines in AIC cultured RCrECL at 100 pM (light bars) and 200 pM (dark bars) concentrations. Each bar graph represent means ± s.e.m., n = 4. 136 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Km = 28.1 ± 8.8 pM V m ax = 8.4 ±1.1 pmol/mg protein/min Total 20 Na(-) Q . a. Na(+) 0 40 60 Uridine concentration (pM) c Km = 2.08 ± 0.28 mM V m ax = 2.40 ±0.21 nmol/mg protein/min Uridine concentration (mM) Fig. 4-25: Na+ -dependent [3 H]uridine uptake (1 min) by AIC cultured RCrECL from the apical (A) and basolateral fluid (B), respectively, as a function of dosing concentration, in the presence and absence of sodium. The net Na+ -dependent uptake represents the difference between total uptake and Na+ -free condition. For basolateral (B), values have been subtracted from nonspecific binding as determined by uridine uptake in the presence of Na+ - free and 0.2 mM NBMPR preincubated for 30 min before uptake. Data points represent m e a n ± s.e .m ., n = 4. Error b a rs w h en not s e e n a r e sm a lle r th a n th e sy m b o ls. K eys: ♦ , total uptake; 0, uptake in the presence of Na+ -free condition; • , net Na+ -dependent uptake. 137 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. <D (0 + * Q. 3 V c 100 80 60 40 20 0 ICR n = 4.5 nM t — r - 1 — i— i— i— i T T T T T T T T T T T -2 -1 1 2 3 Log NBTI (nM) Fig. 4-26: Biphasic dose-response of NBMPR (or NBTI) inhibition on Na+ -independent uridine uptake in basolateral AIC RCrECL. Initial rate (1 min) of [3 H]uridine uptake (5 pCi/ml = 0.12 pM) was measured in the absence of Na+ and in the presence of varying concentration of NBMPR (0 to 100 pM). RCrECL was preincubated in Na+ -free BRS containing varying concentrations of NBMPR for 30 min before uptake. Data points represent mean ± s.e.m., n = 4. 138 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 o « 75 Q. g 50 '■ g = 25 i * 0s 4? « / : 4 > ® \ 0 p* Fig. 4-27: Inhibition of basolateral (A) Na+ -dependent and (B) Na+ -independent [3 H]uridine (5 pCi/ml = 0.12 pM) uptake by purines and pyrimidines in AIC cultured RCrECL. In Na+ -dependent uptake, RCrECL was preincubated in 0.2 mM NBMPR Na+ containing buffer for 30 min before uptake. Uptake was then measured without (control) or with simultaneous addition of 100 pM of nonradioactive cytidine, thymidine, guanosine, or inosine. As for Na+ -independent uptake, RCrECL was preincubated in Na+ -free buffer for 30 min before uptake in the presence of Na+ -free without or with simultaneous addition of 1 mM of non radioactive cytidine, thymidine, uridine, adenosine, guanosine or inosine. Data represent mean ± s.e.m., n = 3-4. 139 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. rbCNT2 1 2 3 4 5 1. 100 bp ladder 2. Corneal epithelium (cDNA) 3. Corneal epithelium RNA neg ctr 4. Small intestine (cDNA) 5. Small intestine RNA neg ctr rbENT2 1 2 3 4 5 Fig. 4-28: RT-PCR analysis of (A) concentrative nucleoside transporter 1, CNT2 and (B) equilibrative nucleoside transporter 1, ENT2 mRNA in the corneal epithelium. PCR products are seen only in reactions using oligonucleoside primer pair for ENT2 but not CNT2. Positive controls with corneal epithelium cDNA indicate the expected size of amplified fragment (ENT2, 495 bp). DNA (100 bp) size markers are indicated (/eft). The two specific primers used for amplifying rbCNT2 were CTGCTTCTGTGATGGGTGCC (sense primer corresponding to nucleotides 1143-1162) and GATTGGCAAATCCACAGAGTG (antisense primer corresponding to nucleotides 1647-1667), which generated a 525-bp rbCNT2 PCR product. The primers were used in PCR amplification with corneal epithelium cDNA as the template or RNA as the negative control under the following conditions: 94°C for 1 minute; 55°C for 1 minute; 72°C for 1 minute for 35 cycles for the rbCNT2 specific primers. 140 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Km = 25 ± 8.8 nM Vmax = 17 ± 2.6 pmol/mg protein/min < 8 ® I --------- 1 20 40 60 80 Uridine concentration (uM) Control 1 m M Na (-) Fig. 4-29: Evidence of Na+ -dependent high affinity nucleoside transporter on the apical membrane of LCC cultured RCrECL. Inset: Na+ -dependent [3 H]uridine uptake (at 1 min) by LCC RCrECL from the apical membrane as a function of dosing concentration, in the presence and absence of sodium. The net Na+ -dependent uptake represents the difference between total uptake and Na+ -free condition. Data points represent mean ± s.e.m., n = 4. Error bars when not seen are smaller than the symbols. Keys: ♦, total uptake; 0, uptake in the presence of Na+ -free condition; • , net Na+ -dependent uptake. 2.4. Evidence for an Amino Acid Transporter 2.4.1. Time Course of L-Carnitine Uptake [3 H]L-camitine uptake into the apical (Fig. 4-30A) and basolateral (Fig. 4-30B) AIC RCrECL showed temperature dependency at all time points evaluated (up to 15 min). At 37°C, uptake from basolateral fluid up to 10 min was comparable to that from apical fluid but after 10 minutes, apical seems to be higher than basolateral (seem to be saturating). At 4°C, both apical and basolateral uptake process were significantly lower (by > 90%) than that at 37°C at all time points. 141 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.4.2. Concentration Dependency - L-Carnitine Uptake L-Camitine taken up from the apical (Panel A) membrane of AIC cultured RCrECL consist of a saturable component whereas, the basolateral membrane have not reach saturation up to 1 mM L-camitine concentration (Fig. 4-31). The saturable component (denoted by solid circle and a solid line in Fig. 4-31 A) was calculated by subtracting the linear non-saturable component from total uptake, and was used to estimate the kinetic parameters of L-camitine uptake. For apical uptake (0-1000 pM), Km was 148 ± 53 pM and Vm a x was 129 ± 17 pmol/mg protein/5min. For basolateral uptake saturation has not been reached at 1 mM concentration and thus, kinetic parameters could not be calculated. 2.4.3. Inhibition of L-Carnitine Uptake Inhibition of [3 H]L-camitine by L-arginine, L-camitine and some organic cations were determined in both apical and basolateral AIC cultured RCrECL (Fig. 4-32). Of the compounds tested, only arginine inhibited apical [3 H]L-camitine uptake in AIC cultured cells by 35% (Fig. 4-32A). Interestingly, L-arginine, L-camitine, TEA, verapamil, choline, cimetidine and cyclacillin inhibited basolateral [3 HJL-camitine uptake in AIC cultured cells by 26, 24, 13, 15,12, 44 and 9%, respectively (Fig. 4- 32B). In addition to L-arginine inhibition (63%) of apical [3 H]L-camitine uptake as seen in AIC cultured RCrECL, excess L-camitine also inhibited [ H]L-camitine uptake by 35% in LCC cultured RCrECL (Fig. 4-33). 142 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.06-1 ■ * .E *S v § ■ o 0.04- o °- .E o > 5 E C i s $ E 0.02- i. CL 0.00 10 15 0 5 Time (min) 0.06-| o 0.04- L _ < D O- = O ) a E c ^ t ° ( 0 c u = 0.02- Q . 0.00 0 5 10 15 Time (min) Fig. 4-30: Time course of (A) apical and (B) basolateral uptake of [3 H]L-carnitine (2 pCi/ml = 25 nM) by AIC cultured RCrECL. Data points represent means ± s.e.m.. n = 4. Where no error is shown, it falls within the symbol. Key: ■, 37°C (solid line); □, 4°C (dash line); • , net active uptake (solid line). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 750 n Km = 148 ± 53 |iM V m a x = 129 ± 17 pmol/mg protein/5min c E re tp ° r ~ 500- Z 3 < u O o C L - 4 ? & " E o) ,» I 250- “ i Q . 1000 500 750 0 250 Concentration (fiM) 600 c < D E re in 400- 3 0 ) s s IE £ L E o) ® J 200- Q . •O - 0 250 500 750 1000 Concentration (|j.M) Fig. 4-31: Concentration dependence of [3 H]L-carnitine uptake (2 pCi/ml = 25 nM + 10-1000 pM unlabeled L-carnitine) by apical (A) and basolateral (B) membrane of AIC cultured RCrECL. Each point represents the mean ± s.e.m., n = 4. Overall uptake (■ ) was corrected for simple diffusion determined at 4°C (□) to yield carrier-mediated uptake (•), which was fitted to Michaelis-Menten kinetics by non-linear regression analysis. Where no error is shown, it falls within the symbol. Insert: Eadie-Hofstee plot of gly-sar uptake after correction for nonsaturable component. V, uptake rate (pmol/mg protein/min); S, gly-sar concentration (mM). 144 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 4-32: Inhibition of [3 H]L-carnitine (2 pCi/ml = 25 nM) uptake by 1 mM unlabeled L- carnitine, guanidine, TEA, verapamil, choline, cimetidine, cyclacillin and 2 mM L-arginine from (A) apical and (B) basolateral membrane of AIC cultured RCrECL. Each bar represents the mean + s.e.m., n = 4. 145 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 4-33: Inhibition of [3 H]L-carnitine (2 pCi/ml = 25 nM) uptake by 1 mM unlabeled L- carnitine, guanidine, TEA, verapamil, choline, cimetidine, cyclaciliin and 2 mM L-arginine from apical membrane of LCC cultured RCrECL. Each bar represents the mean ± s.e.m., = 4. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. D. IN VITRO-IN VIVO CORRELATION 1. In Vitro-In Vivo Correlation 1.1. Ocular Absorption and Tissue Distribution of P-Adrenergics The tissue-concentration profiles of p-adrenergics in ocular tissues are shown in Figure 4-34. Most of the topically applied drugs have a peak corneal absorption time of -30 minutes. This peak absorption was confirmed by determining drug distribution/absorption into intraocular tissues. The results indicate that the absorption of p-adrenergics into the ocular tissues was rapid with very low drug concentration in the plasma. The % fraction drug absorbed after 30 minutes was normalized by the wet weight or volume of each tissue. The average wet weight of conjunctiva (Cj), comeal epithelium (CE), comeal-stroma-endothelium (CS) and iris- ciliary body (IC) was 501 ±44, 11.3 ± 1.4, 53.6 ± 3.1 and 42.3 + 2.5 mg, respectively (n=44). The results indicate that most of the P-blockers absorbed was optimal in the comeal epithelium after 30 minutes except nadolol and levobunolol which reside mainly in aqueous humor (AH) (Fig.4-34). HPLC analysis of the absorbed nadolol samples revealed no evidence of metabolite formation but levobunolol was metabolized completely into dihydrolevobunolol. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. S otalol Pindolol Atenolol Nadolol Timolol Acebutolol Levobunolol Metoprolol Labetalol Alprenolol B etaxolol Propranolol Fig. 4-34: Ocular absorption and tissue distribution of p-adrenergics following a single 25 nL ophthalmic administration of 15 mM p-adrenergic drugs. Note: conjunctiva (Cj), corneal epithelium (CE), corneal stroma-endothelium (CS), aqueous humor (AH), iris-ciliary body (1C) and plasma. 0 0 1.2. Effect of Lipophilicity on In Vivo Drug Absorption The effect of drug lipophilicity was examined by evaluating the in vivo corneal epithelium absorption characteristics of a series of P-adrenergics that vary in the log octanol/pH 7.4 buffer partition coefficient (log P). These P-adrenergic drugs evaluated were the same compounds tested in the in vitro permeability. Here, we will determine if the in vivo corneal absorption method will also lead to a similar sigmoidal relationship observed in in vitro permeability as a function of lipophilicity. The in vivo comeal epithelial % fraction absorption was determined for each p- adrenergic drug (Table 4-14) with the highest absorption from betaxolol resulting in 0.803 ± 0.097 % fraction absorbed into the comeal epithelium and with the least absorption from sotalol resulting in 0.094 ± 0.009 % fraction absorbed into the comeal epithelium. This is about a 8.5-fold difference. Figure 4-35 shows a plot of % fraction absorbed as a function of logP for all P-adrenergic drugs absorbed into the comeal epithelium. HPLC analysis of the absorbed samples revealed no evidence of metabolite formation for all evaluated P- adrenergic drugs except levobunolol as observed in in vitro permeability with metabolite, dihydrolevobunolol. Only about 50% of the levobunolol analyzed from the comeal epithelium were metabolized to dihydrolevobunolol whereas, in the aqueous humor was 100% of the metabolite. The trend in the influence of increasing solute logP on the comeal epithelial absorption, is best described by a sigmoidal relationship (Fig. 4-35)(r2 =0.97, p<0.01). From the nonlinear least square curve 149 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. fitting, the parameter estimates of the minimum Absm m was 0.806 ± 0.35 % fraction absorbed, the maximum Absm a x was 13 ± 3.0 (x 10'6 cm/sec), the logP at the half- maximal Pa p p was 2.58 + 0.21, and the transitional slope representing the gradual increase in Pa p p between Pappmin and Pappm a x was 0.27 ± 0.22. The values represent predicted permeability’s ± standard error means of the parameter estimates listed in Table 4-15. Table 4-14: In Vivo Corneal Epithelial Absorption of p-Adrenergics Following a Single 25 pL ophthalmic administration of 15 mM p-Adrenergics. Drug LogP3 MWb In Vivo CE % Fraction Absorbed Sotalol -0.62 308.8 0.094 ± 0.009 Atenolol 0.16 266.3 0.105 ±0.018 Nadolol 0.93 309 0.122 ±0.003 Pindolol 1.75 248.3 0.190 ±0.033 Acebutolol 1.77 372.9 0.056 ± 0.028 Metoprolol 1.88 684.8 0.124 ±0.011 Timolol 1.91 432.5 0.148 ±0.044 Levobunolol 2.40 291 0.266 ±0.013 Labetalol 2.55 364.9 0.198 ±0.044 Alprenolol 2.61 399.4 0.629 ± 0.075 Propranolol 3.21 259 0.698 ± 0.040 Betaxolol 3.44 347 0.803 ± 0.097 aLog octanol/pH 7.4 buffer partition coefficient. bMolecular weight. cAcetonitrile (Acne). dlnternal standard (IS). Mobile phase: 0.2% TEA, pH 3.0 (adjust with phosphoric acid). Flow rate = 1 ml/min for all com pounds except sotalol and atenolol (1.5 ml/min). Injection volume = 100 pi; Run time = 10-15 min; Internal standards = 20 pg/ml. Dual UV and fluorescence detection w as se t up for analysis of levobunolol and dihydrolevobunolol (metabolite). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ill 0 § S 1 £ T S d ) n i — o < / > . q a a > » o ■ a c o * -* o (0 1.00-1 0.75- 0.50- 0.25- 0 .00 - 12 r i 6 -1 0 “I 4 LogP Fig. 4-35: In vivo corneal epithelial absorption of p-adrenergic drugs as a function of logP. Key in increasing order lipophilicity: 1-sotalol, 2-atenolol, 3-nadolol, 4-pindolol, 5-acebutolol, 6-metoprolol, 7-timolol, 8-levobunolol, 9-labetalol, 10-alprenolol, 11-propranolol, 12-betaxolol. The effective half-maximal LogP was found to be 2.90 ± 0.27 (^=0.97). Where no error is shown, it falls within the symbol. Table 4-15: Parameters for Sigmoidal Relationship Between p-Adrenergic In Vivo Corneal Epithelial Absorption with its Corresponding Lipophilicity. In Vivo Absorption (%) Parameters Mucosal Absm jn 0.806 ± 0.35 AbSm a x 13.26 ±0.63 LogP at Abs5 0 2.90 + 0.27 Transitional slope 0.106 ±0.026 r2 0.9680 Papp = Absm in + AbSm a x - AbSm in 1 + exp [(LogP at Abs5 0 - LogP)/Slope)] A b sm in represent the minimum predicted permeability that estim ates the paracellular transport of hydrophilic molecules, A b S m ax is an estim ate of the maximal transepithelial permeability that represents the transcellular permeability of lipophilic molecules, LogP at Absso is the logP corresponding to the half maximal permeability, and the slope is the transition between A b sm in and A b S m ax , respectively. Note: The transitional slope represents the gradual increase in Papp. P aram eters for the sigmoidal relationship betw een Abs and LogP were performed using G raphPad Prism version 3.00 for Windows, G raphPad Software, San Diego California USA, www.graphpad.com". 151 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.3. In Vitro-In Vivo Correlation for All p-Adrenergics As shown in Fig. 4-36, in vivo absorption of 12 P-Adrenergic drugs with diverse lipophilicity characteristics resulted in a good linear correlation (r2 =0.953) with that of in vitro cell culture permeability. o (0 y = 0.0488x + 0.0894 R2 = 0.9527 1,2,3,5 0 2 4 6 8 10 12 In-vitro culture Pa p p (x10'6 cm/s) 14 Fig. 4-36: In vitro-in vivo correlation for all p-adrenergic drugs. Key in increasing order lipophilicity: 1-satolol, 2-atenolol, 3-nadolol, 4-pindolol, 5-acebutolol, 6-metoprolol, 7-timolol, 8-levobunolol, 9-labetalol, 10-alprenolol, 11-propranolol, 12-betaxolol. CE = corneal epithelium. 152 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.4. Ocular Absorption and Tissue Distribution of Paracellular and Carrier- Mediated Substrates The tissue-concentration profiles of paracellular and carrier-mediated substrates in ocular tissues are shown in Figure 4-37. Consistent with what was observed with 0- adrenergics, most of the topically applied drugs have a peak corneal absorption time of ~ 30 minutes. This peak absorption was confirmed by determining drug distribution/absorption into intraocular tissues. The results indicate that the absorption of paracellular and carrier-mediated substrates into the ocular tissues was rapid with very low drug concentration in the plasma. The % fraction drug absorbed after 30 minutes was normalized by the wet weight or volume of each tissue. The average wet weight of conjunctiva (Cj), corneal epithelium (CE), comeal-stroma- endothelium (CS) and iris-ciliary body (IC) was 501 ± 44,11.3 ± 1.4, 53.6 ±3.1 and 42.3 ± 2.5 mg, respectively (n=44). Bidirectional in vitro transport and in vivo topical absorption of paracellular and carrier-mediated model drugs by the corneal epithelium were determined (Table 4-16). Of the carrier-mediated model drugs evaluated, lactic acid, gly-sar, valacyclovir, verapamil and CsA were among the drugs that resulted in net directional transport in the BA direction except valacyclovir with a net directional transport in the AB direction (~ 1.3-fold higher). As for in vivo absorption of these compounds, cidofovir, lactic acid and gly-sar were among ones that absorbed to the same extent as mannitol. L-camitine, L-arginine, CsA and uridine were among ones that resulted in 1.4, 1.5, 2.7 and 4.4% fraction dose absorbed into the comeal epithelium, which were 153 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. significantly higher than mannitol absorption by 242, 257, 459 and 750-fold, respectively. 1.5. In Vitro-ln Vivo Correlation for all Transport Mechanisms in the Corneal Epithelium The Pa p p of all compounds (n=23) evaluated fell within the range of 0.05 ± 0.008xl0'6 cm/sec (mannitol) and 12.9 ± 0.05xl0'6 cm/sec (betaxolol) (Fig. 4-38). The corresponding % dose absorbed was in the range of 0.0058 ± 0.0016% (mannitol) and 4.35 ± 0.69% (uridine), with betaxolol being the fifth highest. A good in vitro-in vivo linear correlation was observed (y=0.049+0.052x, r2 =0.94, n=23), except for the actively transported uridine (a nucleoside transporter substrate), CsA (a Pgp substrate), L-arginine (an amino acid transporter substrate) and L- camitine (an organic cation transporter substrate). The ocular absorption of these four compounds was underestimated according to the above correlation equation. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Fig. 4-37: Ocular absorption and tissue distribution of paracellular and carrier-mediated substrates in rabbits following a single 25 pl_ ophthalmic administration of 3H or 4 C-drugs. The following ocular tissues were assessed: conjunctiva (Cj), corneal epithelium (CE), corneal stroma- endothelium (CS), aqueous humor (AH), iris-ciliary body (1C) and plasma. 50 Mannitol Guanidine L-carnitine 2 50 - 200 - T I 1 5 0 ' [fl 5 100 - Id 5 0 L J o I * ■ i B , ■ O '' Cidofovir E 0 0.8 1 0 6 - T K ill _ifc_ Acyclovir Uridine o 500 Gly-sar « 12 - 5 » 0.8 - Valacyclovir L-Arginine S 1 0 0 - Q CE CS A H 1 C Plasma Lactic Acid o 0.8 - » 0.6 Verapamil c f o Cyclosporin A = 2 0 0 LA LA Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 4-16: Summary of In Vitro Permeability Coefficient (Papp) and In Vivo Absorption of Various Paracellular and Carrier- Mediated Compounds. Drug MW In Vitro Pa p p (x 10'6 cm/sec) In Vivo CE Absorption AB BA % Fraction dose absorbed 1 4 C-mannitol (LogP = -3.1) 182 0.05 ± 0.008 0.048 ± 0.005 n.d. 3 H-mannitol (LogP = -3.1) 182 0.083 ± 0.01 0.11 ±0.012 0.0058 ± 0.002 3 H-acyclovir 225 1.07 ±0.028 0.97 ± 0.048 0.104 ±0.019 3 H-L-arginine 174 2.39 ± 0.068 0.63 ± 0.026 1.49 ±0.161 Cidofovir (clogP = -2.43) 315 0.17 + 0.02 0.15 ±0.009 0.0068 ± 0.002 3 H-L-carnitine HCI 198 0.048 ± 0.002 0.038 ± 0.002 1.401 ±0.269 1 4 C-lactic acid 90.1 0.19 ±0.20 1.50 ±0.19* 0.006 ± 0.002 1 4 C-guanidine 95.5 0.43 ± 0.03 0.50 ± 0.04 0.043 ± 0.01 3 H-gly-sar 146 0.14 ±0.006 0.28 ± 0.04* 0.0056 ± 0.001 3 H-uridine (LogP = -1.98) 244 0.29 ± 0.04 0.20 ± 0.01 4.352 ± 0.687 3 H-valacyclovir 397 0.71 ± 0.059 0.54 ± 0.009** 0.045 ±0.014 3 H-verapamil (LogP = 3.8) 491 4.8 ± 0.05 5.8±0.11* 0.245 ± 0.045 3 H-cyclosporin A (LogP = 3.0) 1203 0.077 ± 0.007 0.125 ±0.023* 2.66 ± 0.59 AB = Apical-to-basolateral direction. BA = Basolateral-to-apical direction. Values represent m ean ± s.e.m . (sam ple size) from two or more culture preparations. Sam ple size (n)=4-9. * Significant difference betw een Papp in AB and BA (p<0.05). Log P = log octanol/ pH 7.4 buffer partition coefficient ** Statistical significance w as tested by one way ANOVA and the Student-Newm an-Keuls post test (p<0.05) for acyclovir and valacyclovir. U \ O s 6 Uridine 4 CsA 2 r2 = 0.9397 j L-Arginine L-Carnitine 0 15 11 13 9 3 5 7 1 1 I n - v itr o culture Papp (x10"6 cm /s) Fig. 4-38: Correlation between in vivo fraction absorbed in corneal epithelium (CE) after 25 pL topical administration (expressed as % fraction of the administered drug dose) and in vitro permeability in AIC cultured RCrECL for all compounds evaluated. 1.6. In Vivo Inhibition of Actively Absorbed Compounds In vivo inhibition of 3 H-mannitol, 3 H-uridine, 3 H-L-arginine and 3 H-CsA absorption were determined using 1 mM unlabeled excess substrates for all compounds except 0.5 mM propranolol for inhibition of 3 H-CsA (Fig. 4-39). Ocular tissues collected for analysis include Cj, CE, CS, AH, IC and plasma. In vivo corneal epithelial absorption of 3 H-mannitol appears to be 1,000-, 250-, and 450-times lower compared to 3 H- uridine, 3 H-L-arginine and 3 H-CsA, respectively. In addition, 3 H-uridine, 3 H-L- arginine and 3 H-CsA corneal epithelial absorption were inhibited by excess substrates or propranolol by 90, 81, and 47%, respectively. Inhibition of the corneal epithelium also inhibited absorption into subsequent tissues (e.g. CS, AH, IC, etc.) significantly as seen in Fig. 4-39. 157 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2. Possible Reasons for the Lack of In Vitro-In Vivo Correlation 2.1. Culture Age on Transporter Activity In vitro-In vivo correlation for all compounds evaluated (Fig. 4-38) resulted in significant higher in vivo absorption for uridine, CsA, L-camitine and L-arginine than expected for the in vitro permeability values. One possible reason for this difference is that the transporter expressions are downregulated as a function of culture condition or age. Thus, culture age dependent uridine, CsA, L-camitine and L-arginine uptake was determined for both LCC and AIC cultured RCrECL. In these cultures, the cells reached at least 100 flcm 2 by day 4 and thus were considered tight enough for drug uptake evaluation. L-arginine uptake in LCC cultured cells appears to decrease from days 6 to 10 by -75% (Fig. 4-40). L-arginine uptake in AIC cultured cells appears to peak on day 8 and decreases to half of the uptake activity by day 10. Using verapamil to inhibit CsA efflux, allowed for enhanced CsA uptake into the cell layers. For both culture conditions, CsA efflux activity was the highest on day 6 with an efflux inhibition of 70-80% with verapamil (Fig. 4-41). This activity was not observed at all by day 10 for both LCC and AIC cultures. Fig. 4-42 shows the effect of culture age on L-camitine uptake in both LCC (panel A) and AIC (panel B) cultured RCrECL. L- camitine uptake activity appears to be completely abolished by day 8 in LCC and by day 6 in AIC cultured RCrECL. Uridine uptake in LCC cultured cell does not appear to change from days 4 to 9, whereas, uridine uptake in AIC cultured RCrECL decreased from day 4 to 8 by 90% (Fig. 4-43). 158 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mannitol Mannitol (1 mM-excess) Uridine (1 mM-excess) Uridine L-Arginine (+1 mM excess) L-Arginine CE CS AH IC Plasma CsA (+ 0.5 m M Propranolol) 100% IC Plasma Fig. 4-39: In vivo inhibition of mannitol (1 pCi/eye = 2.01 pM), uridine (1 pCi/eye = 0.97 pM), L- arginine (1 pCi/eye = 0.70 pM), and CsA (0.5 pCi/eye = 2.20 pM) absorption by 0.5 mM or 1mM excess unlabeled substrates into conjunctiva (Cj), corneal epithelium (CE), corneal stroma-endothelium (CS), aqueous humor (AH), iris-ciliary body (IC) and plasma following a 25 pi single topical instillation into pigmented rabbit eyes. NS = not significant. 159 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.0 o > (0 a 3 0 ) c E O ) c E o c o k. Q . O ) E ^ 0.5 H o E a 1.5 1.0 - 0.0 A T X 1 II 10 Culture Day © (0 c E o a 3 C a ) ® *; E 2 c Q - D ) O ) 2.0 i 1.5 - 1.0 - o °-5 H E a 0.0 B o T 8 Culture Day 10 Fig. 4-40: The effect of culture age on [3 H]L-arginine (2 (iCi/ml = 35 nM) uptake in (A) LCC and (B) AIC cultured RCrECL. The nonspecific binding was determined by 5 mM inhibition with 5 mM unlabeled L-arginine and subtracted from the net L-arginine uptake. Each bar graph represent mean + s.e.m., n=4. 160 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. £ £ o © s: X. £ « © 3 < < o o) O E 4 0 0 300 - 2 200 ■ 5 100 E a 8 Culture day m 10 1500 £ E o © ^ ■tt £ S © < “ • t n o> ° I o E Q . 1000 500 B I rTl m Culture day 10 Fig. 4-41: The effect of culture age on [ H]CsA (4 nCi/ml=0.5 |uM ) uptake in (A) LCC and (B) AIC cultured RCrECL. Inhibition of [3 H]CsA efflux was determined by 0.5 mM unlabeled verapamil (seen as the darker color bars). Each bar graph represent mean ± s.e.m., n=4. 161 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3 i < c 0.10 i 0.08 ■ 5 0.06 c c c O ■ d > o s_ Q. a> £ < / > ( D O E a c f l > "g ^ c (0 L O Q . C 3 * 5 5 I t )E a c o > E (0 O I X C O (0 a > o E Q. 0.04 -J ® 0.02 0.00 0.10 0.08 0.06 0.04 0.02 0.00 fh m 6 8 Culture day B Jri jt l 6 8 Culture day Fig. 4-42: The effect of culture age on [3 H]L-carnitine (2 jj.Ci/ml= 25 nM) uptake in (A) LCC and (B) AIC cultured RCrECL. The nonspecific binding was determined by inhibition with 1 mM unlabeled L-carnitine (seen as the lighter color bars). Each bar graph represent mean ± s.e.m., n=4. 162 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 (0 * - » a 3 < D c 0.4 -| c £ 0.3 - c a > * - > o a 0 .2 - D ) | I 0.1 - a 0.0 9 8 4 6 < D X. ■ o 3 Culture day 0.4 -i jc a. 0.2 - o > | 0 1 0.1 - 0.0 4 6 8 9 Culture day Fig. 4-43: Effect of culture age on [ H]uridine (5 |xCi/ml=0.12 ^M) uptake in (A) LCC and (B) AIC RCrECL. The nonspecific binding was determined by inhibition with 1 mM unlabeled uridine and subtracted from the net uridine uptake. Each graph represent mean ± s.e.m., n=4. 163 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.2. Effect of Antibiotics and Antifungal on Transporter Activity Another possible reason for the lack of in vitro-in vivo correlation is the effect of antibiotics and antifungal present in the culture media everyday until cell confluency. As seen in Fig. 4-44, the lack of amphotericinB (AmB), gentamicin (Gent) or streptomicin-penicillin (Strep-pen) did not affect L-arginine uptake in either LCC or AIC cultured RCrECL. Similarly, this was also seen for CsA uptake (Fig. 4-45). The effect of not having AmB, Gent or Strep-pen appears increase uridine uptake by 30, 47 and 60% in LCC cultured RCrECL (Fig. 4-46). As for AIC culture RCrECL, the effect of not having AmB, Gent or Strep-pen also increased uridine uptake by 50, 60 and 40%. Although, LCC and AIC appears to increase uptake to similar degree, in general, uridine uptake in AIC cultured RCrECL in all conditions were lower than LCC cultured cells. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. £ • i Q. £ 3 'a > * 1 £ Q . o > 2 ? < 1 _j o E £ 3 - 2 - 1 - Control AmB (-) Gent (-) Strep-pen (-) £ a > ^ ■ S 2 Q . " e 3 0 I I c a. " 5 > o) < 1 _j o E 3 - 2 - 1 - B T T i , Control AmB (-) Gent (-) Strep-pen (-) Fig. 4-44: Effect of antibiotic and antifungal agents as a function of [ H]L-arginine (2 nCi/ml = 35 nM + 100 |iM unlabeled) uptake in (A) LCC and (B) AIC RCrECL. The corneal cells were cultured without amphotericinB, gentamicin or streptomicin-penicillin from day 1 to culture confluency. Control cells were grown in the presence of all compounds. Each bar graph represent mean ± s.e.m., n=4. 165 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. < D , X C s ® 300 E ? 200 a 3 o < “ ■ (0 O ) O E < 100 - o E Q . Control AmB (-) Gent (-) Strep-Pen (-) 300 B c E o o _ c + 5 '3 a. 3 ” 200 o < “■ V) D > I < 100 o E Q . T JL 0 Control AmB (-) Gent (-) Strep-Pen (-) Fig. 4-45: The effect of antibiotics and antifungal on [3 H]CsA (4 p,Ci/ml=0.5 pM) uptake in (A) LCC and (B) AIC RCrECL. The corneal cells were cultured without amphotericin B, gentamicin or streptomicin-penicillin from day 1 to culture confluency. Control cells were grown in the presence of all compounds. Each bar graph represent mean ± s.e.m., n=4. 166 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. c 1 1 .0 i 0.8 © , * c ® © o 6 3 2 0 ) CL .E O ) 2 E 0.4 3 O E a w 0.2 A * T * X * T 0.0 Control AmB (-) ■ Gent (-) Strep-pen (-) 1.0 B ^ 0.8 © S o 0 6 § ■ o ° 'b J a. Q ) Q . .E o > 2 ^ 0.4 D O E Q . 0.2 - 0.0 Control * T * T AmB (-) Gent (-) Strep-pen (-) Fig. 4-46: The effect of antibiotics and antifungal on [3 H]uridine (5 pCi/ml=0.28 pM) uptake in (A) LCC and (B) AIC RCrECL. The corneal cells were cultured without amphotericin B, gentamicin or streptomicin-penicillin from day 1 to culture confluency. Control cells were grown in the presence of all compounds. Each bar graph represent mean ± s.e.m., n=4. 167 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. V. DISCUSSION Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A. DEVELOPMENT AND CHARACTERIZATION OF A CORNEAL EPITHELIAL CELL CULTURE MODEL 1. Primary Culture of Corneal Epithelial Cell Layers One of the most important factors in culturing corneal epithelial cells is isolating the right cell type from the cornea. Much of the earlier attempts of isolating corneal epithelial cells have utilized the stripping method (Fig. 3-1) (Stocker et al., 1958; Khodadoust, 1967; Wilcox, 1969; Gipson and Grill, 1982) or the tissue explant procedure (Yanoff and Cameron, 1977). In both cases, there is a great chance of contamination of the corneal epithelial cells with fibroblast or endothelial cells. Therefore, our first objective was to develop a better method for isolating the corneal epithelium from the cornea. This was achieved by developing an eyeball holder (Fig. 3-1) that allowed for the exposure of the corneal and limbal epithelium to the digestive enzyme which, greatly limited contamination with fibroblast or corneal endothelium. The exposure of the limbal region to protease allows for the extraction of corneal epithelial stem cells (Schermer et ah, 1986). After 1 hour exposure, the epithelium is gently scrapped off with the dull end of the blade. The Bowman’s membrane (Fig. 1-1), separating the epithelium from that of the stroma, prevents the contamination of corneal fibroblasts. This is evident by the high transepithelial resistance and the purity of the three distinct epithelial cell types ascertained by morphological characterization. In addition, other advantages include a 50% savings in time and possibly greater viability of the corneal epithelial cells due to a steady 169 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. supply of nutrients from the aqueous humor integral to the intact eye during cell harvesting. 2. Bioelectric Parameters of Corneal Epithelial Cell Layers The three electrical parameters used to assess the integrity and activity of corneal epithelial cell layers are the transepithelial electrical resistance (TEER), the potential difference (PD) and the equivalent short-circuit current (Ieq). The TEER is a measure of the resistance to the passive flow of charge and an index of barrier properties or integrity of the cell layers. The spontaneous PD is the difference between the apical and basolateral membrane potential generated by the flow of cations and anions. The Ieq current is an index of the overall capacity of a cell layer to transport ions. An important finding in the present study is that implementing the air- interface condition on day 1 of seeding is critical to the formation of tight junctions as indicated by TEER (Fig. 4-1) and the cell layers’ permeability to paracellular markers (Table 4-1). On this basis alone, our primary corneal epithelial cell culture grown on a permeable support is superior to that reported by Kawazu et al. (1998). Their culture model was characterized by a peak TEER of 144 Q.cm , 35 times lower than that afforded by our AIC model but comparable to that afforded by our LCC model. Therefore, it is not surprising that their drug permeability values varied from as low as 2-fold (e.g. propranolol) to as high as 37-fold (e.g. atenolol) higher than those in the excised tissue (Huang et al., 1983). 170 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The key differences between Kawazu et al.’s model (1998) and our AIC model include: (a) source of corneal epithelial cells, (b) condition of these cells before and during culture preparation, (c) seeding density, (d) substrata, (e) media supplements, and (f) air-interface condition. Specifically, whereas Kawazu et al. (1998) used corneal epithelial cells that were supplied frozen in 10% DMSO and 10% FBS in RCGM2 by Kurabo Industries Ltd. (Osaka, Japan), we used freshly isolated pigmented rabbit corneal epithelial cells. In terms of substrata, whereas, Kawazu et al. (1998) precoated their collagen Transwell-COL with fibronectin, we precoated ours with collagen, fibronectin and laminin. Although we did not carry out an extensive study on the effect of substrata, it is important to note that this factor may have played an important role in the overall result. In terms of seeding density, Kawazu et al. (1998) used 4 x 104 cells/cm2, with cell detachment occurring at higher values. In contrast, we used a seeding density of 0.9 x 106 cells/cm2 and did not observe cell detachment at any seeding density. 3. Comparison of AIC vs. LCC Cultured RCrECL As observed in Fig. 4-1, AIC compared to LCC cultured RCrECL resulted in improved bioelectric parameters that are similar to in vivo condition. It is an established relationship that an increase in the overall cell differentiation of cultured epithelial cells reflects improved electrical properties (Yamaya et al., 1992). This was observed in the growth of tight corneal epithelial cell layers under AIC condition, reaching ~24-fold higher TEER than LCC cultures. This may be the 171 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. result of a less differentiated superficial cell layer known to contain tight junctional proteins in the LCC cultured cell layers. As for cellular activity, although Ieq is comparable between AIC and LCC cultures, this appears to be the result of the estimated ratio of PD to that of TEER being the same, but this may be misleading due to the higher TEER and PD in AIC cultured cells. The 26-fold higher PD in AIC cells indicates a distinct vectorial movement of ions, a greater capacity for ion transport compared to LCC cells. Another method used to assess cellular integrity is by determining paracellular permeability using known hydrophilic markers such as mannitol, fluorescein, and FITC with varying molecular weight. The resulting AIC cultured cell layers demonstrated the expected inverse dependence of permeability to the m.w. of paracellular markers, as in the case of conjunctival epithelial cell layers (Yang et al., 2000). Moreover, the 5 times lower Pa p p for mannitol in the cultured corneal epithelial cell layers than that in cultured conjunctival epithelial cell layers is consistent with their ~5 times larger TEER. We further evaluated the AIC cultured cell layers based on these findings between AIC and LCC cultures. Culturing epithelial cell layers at the air-interface has been attempted in the airway (Adler et al., 1990; Kondo et al., 1991; Mathias et al., 1995), epidermis (Chapman et al., 1989; Kennedy et al., 1996), and conjunctiva (Meller and Tseng, 1999; Yang et al., 2000). To date, air-interface corneal epithelial cell cultures have not been attempted, although the air-interface corneal organ cultures have been reported to show improvements in comeal morphology (Richard et al., 1991) and 172 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lipoxygenase activities (Liminga and Oliw, 2000). In the case of airway epithelial cells, culturing under the AIC condition resulted in enhanced morphological features, polarity, and ion transport characteristics, comparable to those of the native tissue (Kondo et al., 1991; Yamaya et al., 1992; Johnson et al., 1993; Robison and Kim, 1994). This was attributed to improved oxidative metabolism made possible by improved access of the epithelial cells to oxygen when cultured in the air-interface configuration (Johnson et al., 1993; Kondo et al., 1997). In other words, epithelial cells cultured under the LCC configuration may suffer from slower cell growth that result in less differentiated cell types. In addition, LCC condition can also result in hypoxia (Robin et al., 1984), which has been shown to impair cation transport in A549 cells and rat alveolar epithelial cells (Mairbaurl et al., 1997). Comeal hypoxia associated with contact lens wear (especially when the eyes are closed) (Bonanno and Poise, 1987) has been found to slow down epithelial mitotic activity (Hamano et al., 1983) and reduce epithelial wound healing rate (Mauger and Hill, 1992). We evaluated the effect of culturing comeal epithelial cells under air-interface condition on days 1, 2, and 3. Placing these cells under air-interface condition on day 1 compared to day 3 was found to facilitate higher TEER and PD (Table 4-4). Therefore, based on the AIC and LCC bioelectric parameters as well as the permeability data, the key to the formation of a tight, epithelial barrier seems to be direct exposure of the superficial epithelial cells to air, which ensures an adequate oxygen supply. 173 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The major difference between AIC and LCC culture conditions is the ability of the AIC cultured cells to obtain an adequate supply of oxygen which is equivalent to obtaining an adequate supply of ATP. MDCK cells (epithelial cells derived from dog kidney) have been extensively utilized for the study of tight junction formation (Kovbasnjuk et al., 1998; Gonzalez-Mariscal et al., 1999). Recent work with the ATP depletion-repletion model in MDCK cells resulted in the observation that transepithelial resistance dropped rapidly and reversibly in parallel with declining ATP levels (Mandel et al., 1994; Gorodeski and Goldfarb, 1997). 4. Effect of Culture Media Composition on the Bioelectric Parameters In terms of media composition, Kawazu et al. (1998) used DMEM/F12 supplemented with 5% FBS, 10 ng/mL EGF, 0.1 pg/mL cholera toxin, 0.5 pg/mL hydrocortisone, 5 pg/mL insulin, 100 IU/mL penicillin G, and 100 pg/mL streptomycin. We also used DMEM/F12 but without supplementation with FBS and cholera toxin. We found that, supplementing the medium with 1 ng/mL EGF resulted in a 2-fold increase in PD and Ieq , when compared with no EGF (Table 4-1). At higher EGF concentrations (5 and 10 ng/mL), the TEER decreased by >2-fold, whereas the PD and Ie q remained unchanged. Corneal epithelial cells have been grown in 10% (Sun and Green, 1977), 5% (Jumblatt and Neufeld, 1983) and 0% (Hackworth et al., 1990; Castro-Munozledo et al., 1997) fetal bovine serum for assessing cell growth requirement for proper growth of corneal epithelium. Although the effect of FBS on tight junction formation has 174 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. not been determined in corneal epithelial cells, it has been found to alter bioelectric properties of canine (Mortell et al., 1993) and rabbit (Mathias et al., 1995) cultured tracheal epithelial cells. In both reports, TEER was reduced by -25-35% and Ieq increased by -40-50% in the presence of 5% FBS. This effect has been seen in distal human fetal lung (Barker et al., 1995), choroid plexus (Hakvoort et al., 1998), and retinal pigment epithelial cell cultures (Chang et al., 1997). Serum contains a number of polypeptide growth factors, hormones, and soluble matrix components and may influence the state of differentiation of epithelial cells in general. In addition, the amount of these components in sera of different species may differ. This was observed in a case where the rate of rat alveolar epithelial cells type II differentiation toward the type I cell phenotype was significantly modulated by soluble factor(s) present in rat serum (Borok et al., 1995). Given the variability of commercially available fetal bovine serum, it’s propensity to affect the differentiation states of the corneal epithelium and modify epithelial bioelectric properties, a more defined nutritional supplement such as pituitary extract was used in the growth of corneal epithelial cells. In our study, serum was found to inhibit formation of tight junction (as observed by a lower TEER) in AIC cultured corneal epithelial cells. Moreover, we used a 10 times lower EGF concentration and included in our media bovine pituitary extract (BPE), bovine serum albumin (BSA), L-glutamine, transferrin, selenious acid, linoleic acid, gentamicin, and amphotericin B. Compared with DMEM/F12 175 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. non-supplemented media, our supplemented DMEM/F12 + 1 ng/mL EGF prototype media yielded up to 60-and 45-fold higher TEER and PD, respectively (Table 4-1). We evaluated the effect of culturing corneal epithelial cells under air- interface condition on day 1, 2, and 3. Placing these cells under air-interface condition on day 1 compared to day 3 was found to facilitate the formation of tight junctions (Table 4-3). 5. Morphology of Cultured Corneal Epithelial Cell Layers Ultrastructural examination of tissue comeal cell layers by transmission electron microscopy (TEM) allows for the assessment of cellular components, including microfilaments, mitochondria, nuclei, endoplasmic reticulum, and cellular junctions whereas, the scanning electron microscopy (SEM) allows for the assessment of the top cell layers. SEM of comeal epithelial cells a day after seeding reveals that the cells are well attached and are beginning to migrate (Fig. 4-2). At day 7 after seeding, the individual cells seen in day 1 are no longer visible. They have migrated out to become more flattened cells (characteristics of the superficial cell layer). This can also be seen from the transmission electron micrograph (Fig. 4-3), where basal cells are cuboidal in shape with middle layers of cells taking on more of a wing shape and the most outer layer, the superficial cells, the most flattened cell layer. These characteristics as well as the microvilli found on the surface of superficial cell layer and the numerous desmosomes and cytoplasmic filaments are well characterized in tissue comeal epithelium (Hazlett et al, 1980; Doughty, 1997). 176 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6. Immunohistochemical Staining of Corneal Specific Keratin Epithelial cells express a pattern of keratin(s) that defines not only the epithelial tissue type, but also the differentiation status of the cells. Using a commercially available monoclonal antibody, AE5, which is highly specific for a 64,000 molecular weight corneal keratin, designated RK3, immunohistological analysis of freshly isolated corneal tissue compared with that of the cultured corneal epithelial cell layers supports the conclusion that the cultured cells express corneal specific cytokeratin filaments. They similarly express in common the 64-kD ‘corneal’ antigen and this filament appears to be mainly expressed in the superficial cell layers. The use of this 64-kD cytokeratin as an identification marker for the corneal epithelium is well established (Schermer et al., 1986; Rodrigues et al., 1987; Kurpakus et al., 1990). 7. Ion Transport Processes in AIC Cultured RCrECL We found in primary AIC cultured RCrECL derived from pigmented rabbits that Is c is contributed by active Na+ and Cl' transport under short-circuit condition at 64+1.5% (Fig. 4-5) and 44+4.6% (Fig. 4-6), respectively. Since the apical Na+ -free condition (Fig. 4-5) resulted in 64% Is c inhibition and 45% Isc inhibition from amiloride treatment, suggested that about 15-20% of corneal Na+ -absorption is due to other electrogenic, amiloride-insensitive processes (e.g., Na+ -amino acid cotransporter (Liaw et al., 1991) or slightly electrogenic amiloride-insensitive Na+ /Ca2 + exchanger (Rich and Rae, 1995)). Other possible contributions may result 177 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. from Na+ -dependent carrier mediated transport processes (e.g., Na+ -dependent nucleoside (Hosoya et al., 1998), monocarboxylate (Horibe et al., 1998), L-arginine (Hosoya et al., 1997) or organic cation (i.e. OCTN2 type) transporter(s) (Ueda et al., 2000)) that are electrogenic and are present on the rabbit conjunctival epithelium. These transporters have yet to be determined in the comeal epithelium. Placing both apical and basolateral side with Na+ -free condition (Fig.4-5) resulted in a complete Isc inhibition. Placing basolateral side with Na+ -free condition inhibited Is c by -15% which is consistent with the reported 15 ± 5% Isc inhibition of Na+ (K+ )C1' cotransporter with bumetanide on the basolateral membrane of rabbit comeal epithelium (Bonanno et al., 1989). The presence of mucosal Cl' conductive pathway was assessed with NPAA, a known inhibitor of Cl' channel activity in other epithelial tissues (Stutts et al., 1990; Barker et al., 1992), which inhibited Isc by 46-53% in RCrECL (Fig. 4-6). NPAA may also have inhibited electrogenic processes such as Ca2 + -channels (Reinsprecht et al., 1995) on the apical membrane, as Ca2 + entry in rabbit comeal epithelial cells (Rich and Rae, 1995) and frog comeal epithelium (Huff and Reinach, 1985) has been detected. Net Na+ absorption and Cl' secretion were confirmed by bi-directional flux measurements using 2 2 Na and 3 6 C 1 radionuclides. Net baseline Na+ absorption and Cl' secretion rates in rabbit cornea are 0.062 ± 0.006 and 0.046 ± 0.008 pEq/cm2 /hr, respectively. A range of baseline values for various comeal tissues have been reported as 0.023-0.54 pEq/cm2 /hr for Na+ and 0.042-0.057 pEq/cm2 /hr for Cl' (Donn et al., 1959; Friedman and Kuper, 1960; Green, 1965; Ehlers and Ehlers, 178 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1968; Klyce et al., 1973; Fischer et al., 1974; Van Der Heyden et al., 1975). Our results are comparable to the reported values for rabbit corneal tissues, indicating that the RCrECL culture model derived from pigmented rabbits faithfully replicates the intact corneal tissue. Slow inhibition of the Na+ /K+ -pump is characteristic of rabbit and frog corneal epithelium (Klyce and Wong, 1977; Candia, 1972), which we also observed. Ouabain inhibited at 100 min Isc by 86 ±1.5% in pigmented RCrECL cultures with a half-time (Ti/2 ) for ouabain inhibition of Is c of 25 min. The Na+ (K+ )2C1- cotransporter activity was inhibited by bumetanide to 13 ± 1.9% for RCrECL derived from pigmented rabbits. Interestingly, a similar result with 10 pM bumetanide has also been reported for excised cornea from albino rabbits (Bonanno et al., 1989). The existence of basolateral K+ conductance has been reported in frog (Reinach and Nagel, 1985), rabbit (Rae et al., 1990) and human (Rae et al., 1990) corneal epithelium. This channel is selectively blocked by external Ba2 + in the range of 0.1- 10 mM. Its novel properties, including stretch activation and large conductance (Rae et al., 1990), can pose a problem when handling the excised corneal tissues. We have provided indirect evidence for this channel with the use of 2 mM Ba2 + applied to the basolateral side of RCrECL, which inhibited Isc by 18% (with a Tm of 54 min) (Fig. 4-8) for RCrECL from pigmented rabbits. The rate-limiting step of sodium absorption in a number of tight epithelia appears to be the apically localized Na+ -channel activity. The amiloride-sensitive epithelial sodium channel (i.e., ENaC) is functionally distinct from the tetrodotoxin 179 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (TTX)-sensitive, voltaged-gated Na+ -channels in excitable cells. Tetrodotoxin (TTX)-sensitive Na+ -channel transcript has been reported to be absent in the rabbit corneal epithelium (Shepard and Rae, 1998). In our study, addition of 10 pM amiloride to the apical fluid of pigmented rabbit corneal cultures inhibited 45±1.2% of total Is c under short-circuit conditions. The existence of amiloride-sensitive Na+ - channel was further evaluated with a more potent inhibitor, benzamil, which is an amiloride derivative substituted on the guanidine moiety. Benzamil (IC5o=0.1 pM) was found to be 10-fold more potent than amiloride (IC5o=l pM) confirming the presence of amiloride sensitive Na+ -channel on the mucosal membrane of the corneal epithelium. These results on amiloride and benzamil are consistent with an earlier observation with excised bovine cornea tissue (Midelfart, 1987), where a rapid decrease in Isc due to amiloride inhibition of the apical Na+ entry was found. Midelfart reported an IC50 of 2 pM for amiloride, which is similar to our IC50 of 1 pM in cultured RCrECL. In addition to this pharmacological evidence for an amiloride-sensitive Na+ -channel on the apical membrane of excised bovine corneal epithelial cell layers (Midelfart, 1987), some recent reports using in situ hybridization (Matsuo, 1998) and immunohistochemical localization (Mirshahi et al., 1999) techniques have also provided the evidence of the amiloride-sensitive sodium channel in rodent and human comeal epithelium. The presence of ENaC was shown by immunofluorescence, confocal, and electron microscopy to be almost exclusively 180 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. at the plasma membranes, which is regulated by steroid hormones in cells of the human eye (Mirshahi et al., 2001). Since albino rabbits are commonly used for studies, it is important to ask whether the lack of pigment will affect the type of reaction that might be expected in man, a species that usually has a well pigmented eye. In this respect, we’ve compared various active ion transport processes in both corneal culture models derived from pigmented and albino rabbits (Table 4-7). Of the five ion transport properties evaluated in albino derived RCrECL, the degree of Is c inhibition by NPAA and BaCl2 on Cl' and the K+ -channels, respectively, were significantly different compared to pigmented derived RCrECL. Ouabain inhibited at 100 min Is c by 85 ± 0.7% in albino RCrECL with a Tl A of 37 min. This is similar to the observed inhibition in pigmented RCrECL. The Na+ (K+ )2Cl'-cotransporter activity was inhibited by bumetanide to the same extent for RCrECL derived from albino (15 ± 3.5%) and pigmented (13 ± 1.9%) rabbits. We also evaluated the possible existence of Na+ -channels in RCrECL from albino rabbits with amiloride in the same manner as pigmented rabbit corneal cultures. The extent of Isc inhibition was similar in both rabbit strains. These results indicate that there is no difference between RCrECLs from pigmented and albino rabbits in the existence of Na+ -channels on the apical membrane. In addition, they contribute to the Isc to the same extent. It should be pointed out that an earlier study with excised corneal tissues from albino rabbits found amiloride insensitivity on Is c (Marshall and Klyce, 1983). However, our data indicates that bioelectric parameters 181 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (TEER, PD, and Isc) for RCrECL from albino rabbits are similar to those from pigmented rabbits. Moreover, the corresponding sensitivity towards amiloride and ouabain was about the same in both RCrECLs from pigmented and albino rabbits, suggesting that active Na+ absorption is present in comeal epithelial cells of both rabbit species. In summary, active ion transport for Na+ and Cl" mechanisms have been found in both primary cultured RCrECL derived from pigmented and albino rabbits. This is the first report of evaluating active ion transport mechanisms in primary cultured rabbit comeal epithelial cell layers. This primary culture model exhibited ion transport properties similar to those reported for the excised comeal epithelium. These properties may be regulated in conjunction with modulation of comeal drug transport, which remains to be determined. Apical (Tears) Basolateral (Stroma) ci- NPAA — | Ouabain Na 1 — BaCI2 Cl- Amiloride Benzamil | — Bumetanide * Na ci- Fig. 5-1: Summary of the active transport processes across cultured rabbit corneal epithelial cells. (Note: the Na+ -channel present on the apical membrane of the corneal epithelium is controversial). 182 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. B. IN VITRO DRUG TRANSPORT PROCESSES 1. Effect of Lipophilicity on Drug Transport Based on bioelectric parameters, paracellular and passive transcellular permeability, tight AIC compared to leaky LCC cultured RCrECL appeared to be the model of choice. We further characterized the effect of lipophilicity on drug transport across AIC cultured corneal epithelial cell layers. This was determined in both the AB and BA directions using a homologous series of P-adrenergics of comparable molecular weights (-300 daltons) and pKa’s (-9.2). The Pa p p of hydrophilic p-adrenergics such as atenolol (MW 266) and sotalol (MW 308.8) are similar to that of mannitol (MW 183) as shown in Table 4-1. This result was also observed in cultured alveolar (Saha et al., 1994), tracheal (Mathias et al., 1996) and conjunctival epithelium (Saha et al., 1996). The moderately lipophilic p-adrenergics (nadolol, pindolol, acebutolol, metoprolol, and timolol) with LogP between 0.93-1.91 and the highly lipophilic p- adrenergics (levobunolol, labetalol, alprenolol, propranolol, and betaxolol) with LogP between 2.40-3.44 were transported more rapidly and to a greater extent than the hydrophilic p-adrenergics (Table 4-8). The effect of drug lipophilicity on transepithelial transport is best described by a sigmoidal relationship which has been well documented in excised tissue cornea (Huang et al., 1983; Wang et al., 1991) and in tissue conjunctival epithelium (Wang et al., 1991; Yang et al., 2000) as well as primary culture of conjunctival epithelium (Saha et al., 1996; Yang et al., 2000). Other reports of such a relationship include excised hamster buccal epithelium (Tavakoli-Saberi and Audus, 1989), rabbit intestinal mucosa (Narawane et al., 1992), 183 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cultured bovine cerebrovascular endothelial cells (van Bree et al., 1988), Caco-2 cell monolayers (Artursson, 1990), TR146 human buccal epithelium (Nielsen and Rassing, 2000), primary culture of alveolar epithelial cells (Saha et al., 1994) and primary culture of tracheal epithelium (Mathias et al., 1996). The effective half-maximal Pa p p occurs at a logP of about 2.58 and 2.61 for AB and BA transport, respectively (Fig. 4-13). These values are slightly higher than reported comeal tissue values of 2.45 and 2.02 (Table 4-11; Fig. 4-14). The lower effective half-maximal Pa p p observed in other tissues such as conjunctival tissue (by 30-50%) or cultured (by 53%) epithelium is a result of the slightly more leaky barrier properties compared to the comeal epithelium. The closeness of these values to each other indicates that the P-adrenergics are transported passively across the comeal epithelium. Although this appears to be the case, few of the p-adrenergics resulted in net directional transport. Sotalol, acebutolol, timolol, levobunolol, labetalol and propranolol all resulted in a net directional transport in the BA direction except for timolol and levobunolol (net AB direction). Of these compounds, sotalol resulted in the largest (~6-fold) difference between AB and BA transport with a net directional transport in the BA direction. The net BA directional transport of sotalol was associated with a 22% decrease in Ieq whereas, this was not observed for AB transport. This Ieq inhibition is due to the inhibition of K+ -channel on the basolateral membrane of RCrECL by sotalol, which has been reported to inhibit K+ -channels in guinea pig’s heart (Tribulova et al., 1999). This effect may have indirectly affected 184 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. paracellular permeability of sotalol across RCrECL. Another possibility is the involvement of Pgp in the efflux of sotalol but this has never been reported. Many p-adrenergics such as celiprolol (Karlsson et al., 1993), nadolol (Terao et al., 1996), propranolol (Yang et al., 2000), timolol (Terao et al., 1996), acebutolol (Terao et al., 1996) and bunitrolol (Matsuzaki et al., 1999) have been reported to involve Pgp efflux. This is probably the reason for the net BA directional transport for acebutolol and propranolol. Permeability of acebutolol in the BA direction resulted in ~3.5-fold higher compared to AB direction. Taylor et al. (1985) first observed an unexpected low permeability across the intestinal epithelial cells. Later, using Caco-2, in vivo rat intestinal tissue, and a multidrug-resistant cancer cell line (K562/ADM) that overexpressed Pgp, Terao et al. (1996) determined that this low permeability was due to the involvement of Pgp. Similarly, propranolol has also been reported as a Pgp substrate (Yang et al., 2000). Interestingly, timolol has been reported to be transported out of the intestinal epithelial cells by Pgp (Terao et al., 1996) but this was not observed in RCrECL. Instead, there’s a net AB directional transport (~1.7-fold). The reason is unclear but there are a few plausible explanations. Timolol, at 100 pM has been found to inhibit corneal epithelium Na+ /K+ -ATPase activity by 68% (Whikehart et al., 1991). This may have indirectly decreased BA transport across RCrECL. Another reason may be due to the presence of a novel cationic drug transporter on the apical membrane of RCrECL that compete with Pgp in the transport/efflux of timolol. A recent report of such a transporter in human retinal pigmented epithelia cells showed inhibition of verapamil uptake by 185 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. timolol and propranolol (Han et al., 2001). Thus, it is unclear if one of these mechanisms or multiple mechanisms are contributing to the transport of timolol. Table 4-8 shows that Pa p p increased with increasing lipophilicity of the P- blockers. The Pa p p value for acebutolol and labetalol are not as high as would have been predicated from the lipophilicity. P-adrenergics have been reported to differ in their ability to be absorbed by epithelial tissues depending partly on their lipophilicity. On the basis of correlations between permeabilities and lipid solubilities, however, some p-adrenergics have unexpectedly low permeability across the cell layer. All p-adrenergic drugs seem to follow a general trend of increasing transport with increasing lipophilicity with the exception of acebutolol and labetalol. Both of these drugs were underestimated in both AB and BA direction for RCrECL, in in vivo absorption into the corneal epithelium, and in transport across the tissue cornea as reported in the literature (Table 4-8). In general, the degree of absorption and permeability of these P-blocker are not due to molecular weight difference (range 248 to 433 with the exception of metoprolol, with a m.w. of 685), molecular charge (all +1 except labetalol) or degree of ionization of compound since most of the chosen drugs have pKa values of ~9. HPLC analysis of the transport samples revealed no evidence of metabolite formation for p-adrenergic drugs except for levobunolol, which was reduced to dihydrolevobunolol by ketone reductase enzymes present in the comeal epithelium (Tang-Liu et al., 1986; Lee et al., 1988). Metabolism of levobunolol is one explanation for the net AB directional transport observed across the RCrECL. 186 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In spite of the net directional transport observed for sotalol, acebutolol, timolol, levobunolol, labetalol and propranolol, which suggests the involvement of other transport mechanism(s) besides the passive transport, the permeability of p- adrenergics in AIC cultured RCrECL showed a similar sigmoidal relationship to drug lipophilicity and are well correlated with such reporting in excised corneal tissue from 2 different research labs (Fig. 4-15). In conclusion, paracellular and transcellular passive permeability across the AIC cultured RCrECL appears to behave in the same manner as do native rabbit cornea and other native/cultured epithelial models. 2. Effect of Culture Conditions on Drug Transport The effect of culture conditions on drug transport across the corneal epithelial cell layers is relatively unexplored. The purpose of specific aim #2 is to elucidate the different transport mechanisms with the focus of identifying carrier-mediated transport processes across the cultured RCrECL using various known drug substrates. These compounds were divided into two groups, those that are actively transported and those that are transported by passive diffusion, by either the transcellular or the paracellular route. By assessing the permeability characteristics of AIC and LCC in both directions (AB and BA), we aim for a better understanding of the effect of culture conditions on drug transport mechanisms (paracellular and transcellular-passive and active) across the corneal epithelium. 187 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.1. Paracellular Transport The presence of tight junctions around the cells acts to restrict passive diffusion of hydrophilic solutes across the paracellular route. The paracellular pathway across AIC and LCC cultured RCrECL was determined using mannitol, a widely used paracellular marker, under bidirectional (AB vs. BA) and temperature (37°C vs. 4°C) dependent transport (Table 4-12). As expected, mannitol permeability was independent of direction at 37°C for either AIC or LCC cultures. Mannitol permeability was ~4-fold higher in LCC compared to AIC cultures at 37°C. Interestingly, corneal epithelial cells placed under 4°C decreased paracellular permeability by ~2-fold in either AIC or LCC cultures. The observed permeability of mannitol reflected the measured TEER across comeal epithelial cell layers under various culture conditions. At peak bioelectric parameters (days 7-8), LCC cultures (< 1 kQ.cm2 ) were at least 3-fold lower in TEER compared to AIC cultures (>3.5 kQ.cm2 ). In addition, placing either LCC or AIC cell at 4°C increased TEER by > half the original value for 1-2 hours before the start of the experiment (-2.22 and 5.88 kQ.cm2 for LCC and AIC cultures, respectively). Similar temperature effect on TEER and mannitol permeability across Caco-2 monolayers and rabbit jejunum has been reported (Delie et al., 1997). Similar to our finding, TEER and mannitol permeability across rabbit jejunum and Caco-2 monolayer resulted in ~2-fold lower at 4°C compared to 37°C (Delie et al., 1997). These observations have several implications in the interpretation of permeability studies at low temperature, i.e. 4°C. One implication is that a decrease 188 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in flux of small molecules at 4°C, when compared to 37°C, may not necessarily indicate the presence of an active transport component. Another implication is that TEER is greatly dependent on temperature, therefore, utilizing Pa p p values taken at 4°C as a function of TEER, can be assessed to determine passive or active transport mechanism. As expected, permeability of mannitol at 4°C and 37°C for LCC and AIC cultured RCrECL correlated linearly with the transmembrane electrical conductance values (inverse resistance; see Table 4-23) (r2=0.99). Since the electrical conductance is a measure of charge flow across the membrane via paracellular pathway, this suggests that mannitol is transported via paracellular pathway (Fig. 4-23). In addition, these results showed that TEER and mannitol permeability across LCC and AIC cultured RCrECL are greatly affected by temperature changes. At low temperature, the paracellular space appeared to be less permeable to hydrophilic markers. Unfortunately, the molecular structure involved in the closure of paracellular space at low temperature is unclear. Of the hydrophilic compounds evaluated, guanidine resulted in a similar permeability vs. electrical conductance relationship (r2=0.93) profile as mannitol indicating that guanidine transport across the corneal epithelium is mainly by paracellular diffusion. In addition, guanidine permeability was independent of direction. Guanidine (pKa = 12.5), a primary amine that almost exist exclusively as the quanidinium ion at physiological pH, is a known substrate for characterization of organic cation transporter of OCT type in tissues such as human placenta (Prasad et al., 1992), kidney (Chun et al., 1997), rabbit alveolar (Shen et al., 1999) and 189 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. conjunctiva (Ueda et al., 2000). Although guanidine is transported across the corneal epithelium by passive diffusion mechanism, permeability at 37°C in both AB and BA directions in either LCC or AIC cultured cells resulted in ~ 10-fold higher permeability compared to mannitol. Similarly, permeability at 4°C was 5 and 10 fold higher for guanidine compared to mannitol in AIC and LCC cell layers, respectively. The reason for the enhanced observed permeability is probably due to the positive charge of guanidine. Cationic drugs have been found to traverse epithelial cell membranes of various tissues including epidermal (Green et al., 1989; Hatanaka et al., 1996), intestine (Karlsson et al., 1999), alveolar (Dodoo et al., 2000) as well as cell lines such as MDCK (Colegio et al., 2002) and Caco-2 (Knipp et al., 1997) better than uncharged and negatively charged compounds. In addition, Rojanasakul and colleagues (Rojanasakul et al., 1992) assessed charge selectivity in rabbit tissues including the nasal, tracheal, bronchial, buccal, rectal, vaginal, corneal, epidermal, duodenal, jejunal, ileal, and colonic epithelia and found that all were highly selective for the absorption of positively charged solutes (Rojanasakul et al., 1992). In the case observed in Caco-2 cells, atenolol, a positive charged beta-adrenergic under physiological pH, resulted in ~7-fold higher permeability than mannitol (Knipp et al., 1997). The reason for the enhanced affinity for cationic compounds is unclear. A recent explanation was proposed by Colegio and colleagues (Colegio et al., 2002), who hypothesized that members of the claudin family of tight junctional proteins create charge selective channels in the paracellular space. At the same time, 190 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Amasheh et al. (2002) specifically identified claudin-2 expression as the key factor in the formation of cation-selective channels in tight junctions of MDCK cells. 2.2. Active Transcellular Transport In general, solute transport across the corneal epithelium operates mainly via the passive transcellular and paracellular pathway (through intercellular space), which has been a subject of intensive study (Huang et al., 1989; Huang et al., 1990; Huang et al., 1991; Reinach et al., 1991; Hamalainen et al., 1997; Maurice, 1997). However, little is known about the active transcellular pathway; either absorptive (influx) or secretive (efflux) directions. In the present study, the presence of active carrier-mediated transport mechanism(s) were evaluated, namely the amino acid (ATB°’ + ), organic cation (OCT and OCTN2 types), monocarboxylic acid (MCT), nucleoside transporters (CNT and ENT) and oligopeptide transporters (PepT). It was previously determined in the paracellular permeability evaluation that OCT type transporter does not appear to be present on the RCrECL. 2.2.1. Amino Acid Transporter Unlike mannitol or guanidine transport, L-arginine AB permeability across LCC and AIC cell layers at 37°C compared to 4°C was -16 and 13-fold higher, respectively. This is - 6-8 fold higher compared to mannitol. The 4°C AB permeability for both culture conditions is comparable to guanidine, indicating that the positive charge on L-arginine may have resulted in enhanced paracellular permeability as observed with 191 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. guanidine. In addition, a net directional transport favoring the AB direction was observed in both cultures. The LCC culture resulted in 15-fold higher AB permeability compared to BA, whereas, AIC culture only resulted in 3.8-fold higher in the AB direction. This is in contrast to no net directional transport for mannitol in either culture conditions. Unlike mannitol and guanidine, permeability of L-arginine was not linearly proportional to the electrical conductance for all culture conditions evaluated (r = 0.37) (Table 4-23). These results indicate the possibility of a carrier- mediated mechanism(s) for the transport of L-arginine across the apical membrane (tear side) of the corneal epithelium. The results also indicate that culture conditions may affect L-arginine permeability as observed by the 4-fold difference in AB/BA transport at 37°C between LCC and that of AIC cultured cell layers. It has been proposed that L-arginine and N°-nitro-L-arginine (L-NA) transport across the rabbit conjunctiva via the amino acid transporter, ATB0 ,+ (Hosoya et al., 1991; Hosoya et al., 1998). The L-arginine transport in conjunctiva is associated with both a high and a low affinity process, with a corresponding Km value of 0.07 mM and 5.90 mM (Hosoya et al., 1997), whereas, L-NA is characterized by a single kinetic process with a Km value of 0.35 mM (Hosoya et al., 1998). Although, carrier-mediated uptake of L-tyrosine was recently reported in the SIRC cell line, it is unclear if this is also the case for native comeal epithelium (Balakrishnan et al., 2002). The presence of amino acid transporter(s) has not been actively explored in the comeal epithelium but it is conceivable that the ATB0 ,+ is 192 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. also likely to be present in the comeal epithelium, like conjunctiva, to salvage and utilize amino acids present in tear. 2.2.2. Organic Cation Transporter L-Camitine (3-hydroxy-4-trimethylaminobutyrate), an organic cation transporter (OCTN2 type) substrate is endogenously produced and shuttled from tissue to tissue for cellular energy production. Permeability of L-camitine across RCrECL resulted in net AB direction for LCC cultured cells (1.7-fold) at 37°C but no net directional transport was observed for AIC cultured cells. The permeability of L-camitine as a function of electrical conductance relationship (r2 = 0.75) was not as clear as mannitol, guanidine or L-arginine (Table 4-23). These results indicate that culture condition affected L-camitine permeability across comeal epithelial cell layers. It is unclear whether this is a result of modified paracellular or active transcellular transport. The active component was further evaluated by determining if L-camitine uptake was temperature dependent, saturable and inhibitable by L-arginine and various cationic substrates. L-Camitine uptake from both the apical and basolateral membrane was rapid and temperature dependent (Fig. 4-38). Apical L-camitine uptake consists of a saturable component that conforms with Michaelis-Menton kinetics, to a Km of 148 pM and Vm a x of 129 pmol/mg protein/5 min (Fig. 4-39). In contrast, there does not appear to be a basolateral saturable component up to 1 mM L-camitine concentration. Next, we assessed the L-camitine inhibition profile. First, 193 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. inhibition was determined in AIC cultured RCrECL. Apical L-camitine uptake was significantly inhibited by L-arginine by 35%, whereas, the basolateral L-camitine uptake was inhibited by L-arginine, L-camitine, TEA, verapamil, choline, cimetidine and cyclacillin by 26, 24, 13, 15, 12,44 and 9%, respectively (Fig. 4-40). Interestingly, apical L-camitine inhibition in LCC cultured RCrECL resulted in 63 and 35% inhibition by L-arginine and L-camitine (Fig. 4-41). These results indicate that the transporter mediating L-camitine apical uptake also have affinity for L- arginine. Basolateral inhibition indicates that there may be more transporters involved than apical membrane. This is not surprising since it has always been thought that nutrients are largely supplied from the aqueous humor. Since, LCC cultured RCrECL represents a less differentiated culture model, these results may indicate that there may be a cell type dependent functional affinity for L-camitine uptake and that this affinity may decrease with more differentiated cell types. The data suggest the lack of OCTN2 activity on the apical membrane of RCrECL since L-camitine uptake was not inhibited by other organic cationic drugs and P-lactam antibiotic (Ganapathy et al., 2000). Recently, Nakanishi et al. (2001) demonstrated that the amino acid transporter ATB0 ,+ is able to mediate the transport of carnitine in a sodium- and chloride-coupled manner by mouse ATB0 ,+ cloned from colon and human ATB0 ,+ cloned from a mammary tumor cell line. Carnitine transport in HRPE cells was inhibited by zwitterionic as well as cationic amino acids such as Lys and Arg, which inhibited carnitine transport by 80-90%. The concentrative capacity of ATB0 ,+ for 194 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. carnitine was found to be much greater than that of OCTN2. However, ATB0 + is a low-affinity transporter for carnitine (Kt = 1-2 mM). In contrast, OCTN2 is a high- affinity transporter for carnitine (.Kt = 5-15 pM) (Tamai et al., 1998; Wu et al., 1999). Carnitine concentration in blood is quite high (range of 10-50 pM) and therefore, OCTN2 is thought to be more important than of ATB0 ,+ for cellular uptake of carnitine in most tissues under physiological conditions. Interestingly, the tissue distribution of the two transporters is quite different. ATB0 ,+ is likely to play a significant role in tissues in which it is expressed since it is expressed primarily in the mammary gland, lung and intestinal tract (Sloan and Mager, 1999), whereas OCTN2 is ubiquitously expressed. ATB0 ,+ , is already known as a high-affinity transporter of cationic and neutral amino acids (Km values in the low micromolar range for many substrates) (Sloan and Mager, 1999), which may function as a major route of amino acid absorption. ATB0 ,+ transports both carnitine and propionyl-L-camitine with lower affinity (Km values of the order 0.6-0.9 mM) than does OCTN2 but, unlike the latter transporter, ATB0 ,+ is a very poor transporter of acetyl-L-camitine. The tissue distributions of the two cloned carnitine transporters are also distinct from one another. The ATB0 ,+ transporter is expressed primarily in the intestinal tract, lung and mammary gland. In contrast, OCTN2 is poorly expressed in the intestine, particularly in adult animals. It seems likely that both OCTN2 and ATB0 ,+ will contribute to intestinal carnitine absorption. The oral bioavailability of carnitine was only reduced by about 50% in mice with a genetic defect in OCTN2 transport (Yokogawa et al., 1999), lending 195 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. support to the idea that multiple transporters absorb carnitine from the gut lumen. Carnitine is not absorbed with extreme efficiency by the mammalian intestine in vivo; indeed only about 55-85% of dietary carnitine is absorbed in man. Studies of carnitine absorption in intact intestine (Rebouche and Seim, 1998) have revealed both active transport and putative ‘passive diffusion’ mechanisms (the latter becoming increasingly important in the adult). The active transport mechanism has functional characteristics resembling OCTN2, but it is tempting to speculate that the presumed ‘passive diffusion’ mechanism includes a major contribution from the low- affinity ATB0 ,+ carnitine transporter. We might also predict that carnitine absorption through ATB0 ,+ would achieve greater importance when OCTN2 was genetically or pharmacologically compromised and during therapeutic supplementation with doses of L-camitine. ATB0 ,+ is expressed in the colon as well as the small intestine and may therefore play an additional important role in scavenging carnitine from the distal intestine, by virtue of its high concentrative capacity. The tear L-camitine concentration has not been reported. Although, L- camitine appears ubiquitously distributed in rabbit eye tissues in the micromolar concentrations (cornea, aqueous humor, vitreous humor, iris, ciliary body, choroid- retina and lens) with the lowest in vitreous humor and highest in the lens (Pessotto et al., 1994). It is unclear where the ocular source is coming from and what purpose it serves. The ATB0 ,+ transporter appears likely to serve a dual role in comeal absorption of amino acids and carnitine (although apparently not of acetyl-camitine) like that in the intestine. The extent to which different ATB0 ,+ substrates will compete 196 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. for absorption in vivo requires more detailed investigation of the functional properties and localization of the transporter protein. 2.2.3. Monocarboxylic Acid Transporter Lactic acid, a known monocarboxylic acid transporter (MCT) substrate was used to evaluate the existence of MCT transporter(s) in comeal epithelium. For both LCC and AIC cultured cells, lactic acid appeared to favor a net directional transport in the BA direction, indicating the presence of a basolateral MCT transporter in the comeal epithelium. Although there is a net directional transport in the BA direction, it is interesting to note that the AB/BA difference between AIC (7.9-fold) and LCC (1.3- fold) is quite large. This difference is mainly due to the extent of AB transport and not BA transport since the extent of BA transport in either culture condition is comparable. One possible reason for this may be the differences in transporter (MCT) expression as a function of culture conditions. Bonanno (1990) has previously identified a lactate-H+ cotransporter located on the basolateral membrane of the rabbit comeal epithelium, a highly glycolytic tissue (Riley, 1969). His evidence for this transporter was supported by changes in intracellular pH due to lactate addition, inhibition of lactate-induced changes in pHi by known lactate-H+ cotransporter inhibitors and pHi changes due to stimulation of glycolysis (i.e. lactate production) by hypoxia and the effects of cotransporter inhibitors on hypoxia-induced pHi changes. Bonanno proposed that this transporter is present on the comeal epithelium for the removal of metabolic waste products 197 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (e.g., lactic acid) so that further production of ATP is not inhibited by high cellular or acidic pH. It has been shown by Mathias (1997) that LCC cultured primary tracheal epithelial cell culture exhibited 43% higher anaerobic cellular respiration (measured by the amount of lactate secreted to the bathing media) than in AIC cultures. The 43% higher production of lactate in LCC compared to AIC may be the reason for the observed difference in lactic acid permeability in RCrECL. Similar to L-arginine permeability, lactic acid Pa p p across RCrECL under various culture conditions does not appear to be linearly related to the electrical 'y conductance (r = 0.506) indicative of MCT transporter(s) present in corneal epithelium. For some tissues such as red blood cells, white muscle, tumor cells and comeal epithelium with few or no mitochondria, lactate is produced in quantity as an end product of glycolysis and so must be expelled from the cell. This will also be true of most cells under conditions of hypoxia, which is known to occur in comeal epithelium during limited supply of oxygen. 2.2.4. Nucleoside Transporter Uridine, a known nucleoside transporter (NT) substrate was used to evaluate the existence of NT transporter(s) in comeal epithelium. For LCC cultured cells, uridine appears to favor a net directional transport in the AB direction. In contrast, AIC cultured cells do not appear to exhibit a net directional transport at 37°C. Despite the lack of directionality for AIC cultured cells, AB transport at 37°C for either culture 198 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. condition was ~ 10-fold higher in permeability compared to 4°C. This is at least ~5- fold higher compared to mannitol. Assessing the relationship between uridine permeability as a function of inverse TEER (r2 = 0.460) indicates the lack of linear correlation (r2 = 0.46) which is indicative of the involvement of carrier-mediated transport mechanism(s). The result suggests the presence of an apical nucleoside transporter in LCC cultured RCrECL and that this transport activity is significantly less compared to AIC cultured cells. The net AB directional and temperature sensitive transport of uridine across LCC cultured RCrECL was the initial evidence that nucleoside transport was mediated by an active carrier-mediated process (Table 4-12). In addition, this transport process is affected by culture condition as seen by the 6-fold difference in AB uridine transport between LCC and AIC cultured cells, whereas, only 1.5-fold difference was observed for BA transport. Since nucleoside transporters are known to exist on both apical and basolateral membrane, this was further evaluated and characterized in AIC cultured RCrECL. Uridine uptake from both the apical and basolateral membrane were rapid and temperature dependent with the basolateral uptake being two times higher than from the apical membrane (Fig. 4-30). This is in contrast to no directionality observed in the 4 hour transport. Apical uridine uptake was strongly dependent on the presence of Na+, being nearly abolished when Na+ was replaced with choline in the bathing medium. In addition, uptake at 4°C as well as in the presence of 0.2 mM unlabeled uridine nearly abolished uridine uptake but inhibition by coronary 199 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. vasodilators such as nitrobenzylthioinosine (NBMPR) did not occur. The maximum apical inhibition of 3H-uridine (75%) was observed under 4°C, 0.2 mM uridine, 1 mM uridine and Na+ -free conditions. The remaining 25% probably represent the non-specific absorption of 3H-uridine. The other physiological nucleosides at 200 pM inhibited uridine uptake to a similar extent (60-75%). These results are consistent with the notion that uridine uptake from the apical membrane of comeal epithelium is mediated by an electrogenic carrier-mediated transport process. In contrast, basolateral uridine uptake was not strongly dependent on the presence of Na+ but was completely abolished at 4°C. Inhibition of basolateral uridine uptake by 0.2 mM unlabeled uridine only inhibited 3H-uridine by -20% and completely abolished uridine uptake at 2 mM uridine. Basolateral uridine uptake was inhibitable by NBMPR at both 1 and 10 pM, which indicates the presence of NBMPR-sensitive equilibrative nucleoside transporter. Kinetic analysis revealed two distinct types of nucleoside transporters, the high affinity transporter (Km<60 pM) present on the apical membrane and low affinity transporters (Km>60 pM) (Paterson et a l, 1981) present on the basolateral membrane of AIC cultured RCrECL. The high affinity transport mechanism is driven by Na+, whereas, the low affinity transporters is not dependent on Na+. A saturable component that conforms to Michaelis-Menton kinetics, with a Km of 28.1 ± 8.8 pM and Vm a x of 8.4 + 1.1 pmol/mg protein/min, was determined for the high affinity transporter on the apical membrane of AIC cultured cells. Similarly, a 200 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. saturable component was found on the basolateral membrane with a Km of 154 ± 26.6 pM and Vm a x of 113 ± 14 pmol/mg protein/min. To test whether the basolateral Na+ -independent nucleoside transporter consisted of both es and ei, the es system was pharmacologically defined as the nucleoside uptake that was inhibited by 100 nM up to 1 pM NBMPR and the NBMPR-insensitive activity (resistant to 100 nM but sensitive to 0.2 mM) as ei uptake. As shown in Fig. 4-31, es uptake system contributed -30% and ei uptake system contributed -70% at 10 pM NBMPR. To further confirm that the basolateral RCrECL consist of both es (-30%) and ei (-70%) systems, the dose response of NBMPR inhibition of 3H-uridine uptake was determined (Fig. 4-34). This dose- response curve was biphasic with a plateau between 10 nM and 1 pM, consistent with the coexistence of es and ei systems. If this dose-response curve was dissected into es (sensitive to 100 nM) and ei (resistant to 100 nM) individual uptake components, when fit to a two-site competition model would result in I C 5 0 values of 4.5 nM and 13 pM, respectively. Since the Na+ -dependent purine nucleoside-selective N1 (CNT2) and Na+ - independent EI (ENT2) nucleoside transporter systems have been cloned (accession # AF323951), RT-PCR was used to confirm the expression of CNT2 and ENT2 (Fig. 4-36). The rabbit intestine cDNA was used as a positive control because both of these nucleoside transporters have been reported in this tissue. As predicted, PCR product of ENT2 (495 bp) was amplified by RT-PCR from RNA isolated from tissue comeal epithelium. In contrast, there was no amplified CNT2 PCR product from 201 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tissue comeal epithelium. Therefore, these results support the functional findings that the basolateral ENT is of ENT2-type and that the apical CNT is not of CNT2- type transporters. The presence of a high affinity Na+ -dependent nucleoside transporter appeared to be located on the apical membrane and low affinity Na+ -independent nucleoside transporters appeared to be located on the basolateral membrane of the AIC cultured RCrECL. The presence of these transporters is probably the reason there is a lack of net directional uridine transport observed in AIC culture (Table 4- 12). The BA permeability value for AIC culture appeared to be in the same range as LCC culture but the AB permeability for AIC culture appeared to be significantly lower by 6-fold. These results indicate that the nucleoside transporter(s) present on the apical membrane of either LCC or AIC may be different or the transporter(s) expression may have been affected by the culture conditions. Fig. 4-37 shows that uridine apical uptake in LCC cultured RCrECL is completely inhibited by the presence of no sodium as well as in the presence of 1 mM uridine. In addition, Na+ - dependent uridine uptake is saturable and exhibits high affinity (Km= 25 ± 8.8 pM) and low capacity (Vm a x = 17 ± 2.6 pmol/mg protein/min) kinetic parameters. The Km for AIC and LCC cultured cells are similar but the Vm a x is 2-fold higher for LCC compared to AIC cultured cells. This indicates that the Na+-dependent nucleoside transporter on the apical membrane of RCrECL may have been downregulated in AIC cultures or that LCC cultures have upregulated the transporter expression. 202 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Our study demonstrates that the corneal epithelium expresses apical I n dependent nucleoside transporter that mediates the transport of nucleoside substrates with high affinity and low capacity, whereas the basolateral Na+ -dependent and Na+ - independent (es and ei-types) transporters mediates the transport of nucleosides with low affinity and high capacity. Based on this data, we hypothesize the concentrative transporter(s) in the apical membrane and both the concentrative and equilibrative transporters in the basolateral membrane are arranged in series in the corneal epithelium to allow for efficient vectorial transport of nucleosides across the tight comeal epithelium. To our knowledge, this is the first functional evidence of nucleoside transport systems in the comeal epithelium. The comeal epithelium, relative to other tissues, is particularly unique with respect to the transport of nucleosides. It is a resilient, self-renewing cellular barrier that serves as the first line of defense against infection, which has a high turnover rate; replacing cells that are sloughed off from the apical membrane. The time for the replacement of the entire population of epithelial cells has been thought to be six- seven days but recently shown to take >2 weeks for complete renewal (Haddad, 2000). The high cell turnover rate indicates that the cornea has a high intrinsic requirement for nucleosides, which are required for cell DNA and RNA synthesis. In most animal tissues such as that in liver (Zollner, 1982), nucleosides can be formed by de novo pathway. However, in certain tissues such as the intestine, including the comeal epithelium, de novo synthesis is probably limited, and a salvage pathway appears to predominate as in the case for the intestine (Sonoda and 203 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tatibana, 1978). This may be cell type dependent (Haaskjold et al., 1990) or circadian dependent in the corneal epithelium (Haaskjold et al., 1992). In the salvage process, cells utilize preformed nucleosides available from extracellular sources. Exogenous comeal sources of nucleosides include the break-down products of sloughed-off epithelial cells. Therefore, the comeal epithelium with its high intrinsic requirements for nucleosides and active salvage pathways may be significantly different from other tissues with respect to the transport of nucleosides and nucleoside analogues. The presence of different nucleoside transporter systems on the comeal epithelium may indicate different physiological roles for them. A possible physiological role for the high affinity apical Na+ -dependent nucleoside transporter process in the comeal epithelium may be to salvage nucleosides from the nucleotides, such as DNA and RNA, produced from the desquamating comeal and conjunctival epithelial cells and subsequently hydrolyzed by ectoenzymes to nucleosides. Similarly, possible role for the predominantly low affinity Na+ - independent nucleoside transporter systems (both es and ei-type) in the basolateral membrane indicate the need for maximal capacity efflux of nucleosides into the comeal stroma/comeal endothelium and subsequently into the aqueous humor or for maximal absorption of nucleoside into the comeal epithelium from the basolateral membrane to meet the high energy demands of the comeal epithelium. Purine adenosine and its metabolite inosine in the aqueous humor up to micromolar levels (Crosson and Petrovich, 1999) have been reported. Although 204 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. purines appear to be major regulators of aqueous humor formation, their physiological source is unknown. In addition, diadenosine tetraphosphate (AP4A) and ApsA were determined in rabbit tear under basal conditions at 2.92 pM and 0.58 pM, respectively (Pintor et al., 2002). These dinucleotides were found to evoke tear secretion suggesting that they play a role in the regulation of corneal hydration. Recently, Ap4A and ApsA were determined in rabbit aqueous humor at 0.34 pM and 0.08 pM, respectively (Pintor et al., 2003). It was found that AP2A, AP3A and ApsA increases intraocular pressure. In contrast, AP4A decreases intraocular pressure and was suggested as a therapeutic agent for treating elevated intraocular pressure (Pintor et al., 2003). It is conceivable that these dinucleotides are generated from break down products of sloughed-off cells from either the corneal or conjunctival epithelium. It is possible that these dinucleotides can potentially be broken down further for the generation of nucleosides to be actively transported across the corneal epithelium and into the aqueous humor. From a therapeutic point of view, the nucleoside transporter(s) may be a conduit for the entry of antiviral compounds through the corneal route into the anterior region. These transporter systems would allow for an efficient vectorial transport of antiviral compounds across the corneal epithelium from topical dosing. The Na+ -dependent concentrative nucleoside transporter system is involved in the transport of nucleosides across the apical comeal epithelial cell membrane. Once inside, the basolateral Na+ -dependent, NBMPR-sensitive and NBMPR-insensitive Na+ -independent processes efflux the intracellular nucleosides across the basolateral 205 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. membrane into the comeal stroma/endothelium. This has great therapeutic potential for delivering nucleosides and nucleoside analog drugs to the anterior regions of the eye. In conclusion, rabbit comeal epithelium possess both Na+ -dependent and Na+ -independent nucleoside transport systems. These transporters may be important in supplying nucleosides to the higher energy demands of the comeal epithelium from either the tear or aqueous side. The Na+ -dependent processes may also play a role in the overall maintenance of Na+ homeostasis. 2.2.5. Oligopeptide Transporter The initial evidence of a carrier-mediated dipeptide transport process was obtained from the finding that gly-sar transport favored net BA direction (Table 4-13). In addition, gly-sar uptake was rapid, temperature sensitive and requires energy (Fig. 4- 26). The peptide uptake process was evaluated on both the apical and basolateral membrane of primary cultured rabbit comeal epithelial cell layers using gly-sar as a model substrate. The net BA directional transport of gly-sar in RCrECL suggests the presence of a basolateral oligopeptide transporter. Based on our functional evidence, the apical dipeptide transporter does not appear to be present in RCrECL. One major difference between the comeal basolateral peptide transporter and the known PepTl and PepT2 transporter is its proton-coupled driven process. We have found it to be much less sensitive to extracellular medium pH. Similar results have been reported in the basolateral membrane of Caco-2 cells (Saito and Inui, 206 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1993). It has been demonstrated that the intestinal (Caco-2 cells) basolateral peptide transporter is functionally distinguishable from the apical H+ -peptide cotransporter (PepTl) and involves the transepithelial transport of small peptides and peptide-like drugs (Inui et al., 1992; Matsumoto et al., 1994; Saito and Inui, 1993; Terada et al., 1999). Since corneal gly-sar uptake from the basolateral membrane was not driven by a proton-coupled process, we evaluated the requirement of metabolic energy for this process. Gly-sar uptake was energy dependent, indicated by the 28% and 49% inhibition by pretreating the corneal epithelial cell layers with 0.1 mM 2,4- dinitrophenol and 2 mM NaN3, respectively. In addition, gly-sar uptake was inhibited by 89% upon lowering the temperature to 4°C. The basolateral peptide transporter in the Caco-2 cell line was characterized as a low-affinity facilitative transporter (Saito and Inui, 1993; Terada et al., 1999). Interestingly, a renal basolateral peptide transporter has also been demonstrated (Fonteles et al., 1983; Lowry et al., 1985) and was found to be functionally distinguishable from known peptide transporters (PepTl and PepT2) as well as the intestinal basolateral peptide transporter (Terada et al., 2000). Terada et al. determined Km values of gly-sar for the renal basolateral peptide transporter (55 pM in rat renal cortical slices and 71 pM in MDCK cells) (Terada et al., 2000) were much smaller than that of the intestinal basolateral peptide transporter of Caco-2 cells (2.1 mM) (Terada et al., 1999). It was thought that the low affinity intestinal basolateral peptide transporter was important for the efficient efflux of substrates into the blood, whereas, the high affinity renal basolateral peptide transporter may 207 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. contribute to the uptake of small peptides from the circulation. Like the intestinal PepTl (Fei et al., 1994; Liang et al., 1995), the basolateral peptide transporter can also recognize and transport nonpeptidic compounds (Irie et al., 2001) and thus, play an important role for the transfer of peptide-like and nonpeptidic drugs to achieve high oral bioavailability. The inhibition study provides information about the specificity of the comeal basolateral transporter. The amino acids L-histidine and (3-alanine did not significantly inhibit the uptake of gly-sar. PHT1 is a Na+ -independent, proton- coupled transporter that transports peptides as well as L-histidine. The lack of inhibition of gly-sar uptake by saturating concentration of L-histidine indicates that PHT1 is not involved. Gly-sar uptake was inhibited by excess amount of gly-sar and inhibited by L-camosine but not by its constituent amino acids, L-His and P-Ala, consistent with the substrate selectivity of PepTl and PepT2. Inhibition of gly-sar uptake by L-His rules out the possibility of this transporter being the peptide- histidine type transporter (PHT). Inhibition of gly-sar uptake by gly-L-phe and gly- D-phe indicates a preference for the L-configuration compared to the D- configuration. The ability of gly-D-phe to significantly inhibit gly-sar uptake also indicates that this basolateral peptide transporter does recognize one amino acid of a dipeptide in the D-configuration. Interestingly, inhibition of gly-sar uptake by TRH was found to be competitive and different compared to a similar study in primary cultured apical conjunctival epithelial cell layers. Whether TRH is a substrate for the peptide 208 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. transporter has been a controversial issue. Some reports indicate that TRH was predominantly transported by the paracellular pathway across Caco-2 cell monolayers (Gan et al., 1993; Thwaites et al., 1993) and across alveolar epithelial cell monolayers (Morimoto et al., 1994) which PEPT2 mRNA and protein have been identified (Groneberg et al., 2001). Others report TRH to be transported by a saturable carrier-mediated transport process in addition to passive paracellular transport in certain clone of Caco-2 (Yokohama et al., 1984; Yokohama et al., 1984; Walter and Kissel, 1994). This is also in agreement with the report that TRH is taken up in rat intestinal mucosal cells by a dipeptide transporter (Tanaka et al., 1996). Recently, TRH was reported not to inhibit L-camosine uptake in the apical membrane of conjunctiva epithelial cells (Basu et al., 1998). In our study, TRH did not significantly inhibit apical uptake of Gly-Sar but inhibited basolateral uptake by -38%. It is unclear what the reasons are for this observed discrepancy. In order to determine if this basolateral oligopeptide transporter was more like PepTl or PepT2, a differential recognition experiment was performed on AIC culture RCrECL. The Ki values for cyclacillin and cefadroxil were found to be 121 and 1739 ± pM, respectively (Fig. 4-29). Cyclacillin was found to be 14-fold more potent than cefadroxil, indicating the possible function of a PepTl-like transporter on the basolateral membrane of corneal epithelium. These inhibitory results of cefadroxil and cyclacillin are particularly interesting, because for the first time, this may shed some light on the relative likeness of this basolateral transporter to that of PepTl in the corneal epithelial cells. Ganapathy et al. (1995) have shown that 209 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cyclacillin is approximately 9 times more selective of PepTl than of PepT2, whereas cefadroxil is approximately 14 times more selective of PepT2 than of PepTl based on its 50% inhibitory concentration (IC50) and inhibition constant (Ki) values. 2.3. Active Transcellular Efflux The most extensively studied active efflux mechanism is associated with the presence of Pgp and to a lesser extent, the MDR protein expressed in either normal or cancer cells (McNamara et al., 1996; Borst et al., 2000). The apically localized Pgp have been found in ocular tissues such as retina (Holash and Stewart, 1993), iris (Holash and Stewart, 1993), conjunctiva (Saha et al., 1998; Yang et al., 2000), cornea (Kawazu et al., 1999) and functionally identified in the blood-aqueous barrier (Kajikawa et al., 1999). Unlike Pgp, MRP exists in the apical, basolateral or cytoplasmic membrane of cells and has been reported in retina (Aukunuru et al., 2000) and conjunctiva (Yang and Lee, 2000). Cyclosporin A (CsA), a known Pgp substrate, is a popular immunosuppressive agent used in the treatment of various ocular diseases, such as some forms of uveitis, keratoconjunctivitis, immune- mediated keratitis, necrotizing scleritis, Behcet’s syndrome, Sjogren’s syndrome, and comeal transplantation (Foets et al., 1985; Faulds et al., 1993; Smet et al., 1993). The ocular pharmacokinetics of CsA has been determined by a variety of investigators (Bell and Hunnisett, 1986; Wiederholt et al., 1986; Kaswan, 1988; BenEzra and Maftzir, 1990; Oh et al., 1995). In addition to the involvement of Pgp, the lipophilic property of CsA as well as the hydrophilic property of the stroma was 210 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. among other reasons for the low intraocular penetration of CsA that ultimately resulted in low ocular bioavailability. As mentioned previously, Pgp has been implicated in the secretion of p- blockers such as acebutolol, celiporolol, nadolol, propranolol and timolol. Propranolol has been implicated in such mechanism in liver canalicular membrane vesicle (Kwon et al., 1996), CEM/VLB100 cells (Zamora et al., 1988; Wigler, 1996), Caco-2 (Karlsson et al., 1993) and conjunctival epithelial cell layers (Yang et al., 2000). Propranolol BA transport across the conjunctival cell layers was ~2.4-fold higher compared to AB direction (Yang et al., 2000). In the present evaluation, the initial evidence of Pgp involvement was observed in the net BA directional transepithelial transport of propranolol (13% higher) across the AIC cultured RCrECL (Table 4-8). Pgp involvement was further evaluated with CsA as the model substrate (Table 4-12). Net BA directional transport (~1.8-fold) was observed for CsA transport across LCC and AIC cultured RCrECL. Regardless of culture condition, BA transport of CsA was ~2-fold higher compared to AB transport at 37°C, whereas, no directionality was observed at 4°C. These results confirm Kawazu’s (1999) finding of Pgp activity present on the tear membrane of rabbit comeal epithelium. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. B. IN VITRO-IN VIVO CORRELATION 1. In Vitro-ln Vivo Correlation Outcome 1.1. Passively Absorbed Drugs Thirty minutes in vivo comeal epithelium absorption was carried out for all compounds evaluated and was correlated with in vitro permeability across the AIC cultured RCrECL. Thirty minutes in vivo absorption time was chosen because that is approximately the peak cornea (Kleinberg et al., 1979) and aqueous humor concentration time for many drugs (Leibowitz et al., 1977; Kleinberg et al., 1979). The correlation obtained between the percentage of absorbed drag after topical administration of P-adrenergics in rabbits and apparent permeability coefficients obtained in AIC cultured RCrECL, is illustrated in Fig. 4-44. The effect of lipophilicity on drag absorption into in vivo comeal epithelium was assessed by the same homologous series of P-adrenergics used in the in vitro cell culture assessment. Intraocular tissue distribution of each p-adrenergic indicates that peak comeal epithelium absorption occurs approximately 30 minutes after topical administration. Of the compounds evaluated, nadolol and levobunolol peak concentration was observed in aqueous humor and not the comeal epithelium. Topical dosing of levobunolol resulted in its metabolism to dihydrolevobunolol by the presence of ketone reductase in the comeal epithelium (Tang-Liu et al., 1988). Present in the comeal epithelium was mainly dihydrolevobunolol (-50%), whereas, the aqueous humor was 100% metabolite. This may have been the reason for the higher aqueous humor concentration compared to comeal epithelium. This was not 212 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the reason for nadolol since analysis of nadolol by HPLC did not suggest the metabolism of nadolol. The significantly higher penetration of nadolol into the aqueous humor may be the use of ethanol in solubilizing 15 mM nadolol and thus, change its absorption distribution profile. Similar to in vitro permeability, in vivo absorption of p-adrenergics into comeal epithelium was found to be dependent on drug lipophilicity and is best described by a sigmoidal relationship. The effective half-maximal in vivo absorption occurs at a logP of about 2.90 ± 0.27 (Table 4-15). This is comparable to the AB and BA in vitro effective half-maximal values of 2.58 and 2.61 (Table 4-9). Acebutolol resulted in both lower expected in vivo absorption and in vitro permeability. Permeability of acebunolol across comeal tissue has previously been reported to fall significantly below than the expected distribution coefficient compared to 12 other P-blockers (Huang et al., 1983). Huang suggested that from a structural point of view, acebutolol might show exceptional hydrogen bonding ability compared to the other 11 p-blockers agents (atenolol, sotalol, nadolol, metoprolol, timolol oxprenolol, levobunolol, propranolol, bevantolol, bufuralol and penbutolol). Similarly, labetalol also resulted in both lower expected in vivo absorption and in vitro permeability. A Similar result has also been reported in TR146 cell culture and in the porcine buccal mucosa models (Nielsen and Rassing, 2000). Stmctural differences were also the reason given for the lower observed permeability/absorption. Labetalol has two benzene rings and on one of these there is a phenolic group in the vincinity of another electron withdrawing group (-NH2), 213 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. which may result in a lower pKa value than 9 (pKa=7.38). In addition, labetalol has the highest polar water accessible surface area, which probably contributes to the relatively low absorption and permeability rates across the two experimental conditions. The molecular weight of labetalol is 328 and thus, does not deviate from the range of these compounds. The LogP for labetalol is 2.55, which is closest to alprenolol, which is 2.61. Nielsena and Rassing reported a 4.4-fold higher permeability for alprenolol compared to labetalol in their TR146 culture model. Ours resulted in a 73-fold higher in vitro permeability and 2-fold higher in vivo absorption. As expected, in vitro permeability and in vivo absorption of passively diffused drugs resulted in good linear correlation (r2 =0.95) (Fig. 4-44). Although there is no data of such correlation reported for corneal epithelium, there have been in vitro-in vivo correlation evaluations of passively absorbed/transported P- adrenergics in tissues such as intestinal (Artursson, 1990). The permeability of such cultures correlated reasonably well with those of in vivo and other in vitro models for drugs that opt for the paracellular pathway. 1.2. Actively Absorbed Drugs Among the different drugs investigated, when compared to other drugs that are equally permeable in cultured RCrECL, CsA, uridine, L-camitine and arginine showed significantly higher in vivo corneal epithelial absorption than would have been predicted by its in vitro counterpart (Fig. 4-46). One may speculate that the 214 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. following reasons partly or mainly contribute to the observed apparent discrepancy between in vitro expectations and in vivo results. First, the dosing concentration employed for in vitro and in vivo were different depending if the drug was radiolabeled or not. This will affect the rate and extent of drug penetration/ absorption across the corneal epithelium depending if it is passive or active transport. Second, when the diffusional (passive) permeability is high, the transport, especially efflux effect may tend to become inconsequential (Chiou et al., 2000a, Spahn- Langguth et al., 1998). This may apply to the extent of in vivo absorption of CsA. The high lipophilicity of CsA allows for good penetration across the lipophilic comeal epithelium but CsA appears to be accumulating mainly in the comeal epithelium (2.4%) with 0.35% CsA detected in the comeal-stroma-endothelium and no amount detected in the aqueous humor. This accumulation has been reported to occur with CsA binding to a cytosolic protein, cyclophilin (Kawazu et al., 1999). Third, the transporter expression in cell culture may greatly differ compared to in vivo comeal epithelium. Reports of in vitro-in vivo correlation have been observed in the human epithelial intestinal cell line, Caco-2 (Chong et al., 1996; Lennermas et al., 1996; Gres et al., 1998) and TC-7 cells (Gres et al., 1998). In recent years, the use of Caco-2 cell monolayers has gained in popularity and momentum as an in vitro human absorption surrogate. Moreover, the monolayers are generally accepted as a primary absorption screening tool in several pharmaceutical companies. In order to validate the use of this culture model for anticipating in vivo absorption in humans, 215 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. various attempts have been made at establishing a relationship between in vitro permeability of Caco-2 monolayer and human absorption in vivo. There are several examples of successful application of these monolayers for correlation with human absorption (Artursson and Karlsson, 1991; Rubas et al., 1996; Yee, 1997). Others have reported that paracellular and transcellular, either passive or active is less efficient than in vivo (Hidalgo et al., 1989; Cogbum et al., 1990; Dantzig and Bergin, 1990; Artursson, 1991; Rubas et al., 1993; Stewart et al., 1995; Lennermas et al., 1996). Although such relationships have been established by these authors, the minimum apparent permeability coefficient values reportedly required to anticipate a 100% absorption in humans, varied between 1 x 10'6 cm/s (Hidalgo et al., 1989) and 6 x 10'5 cm/s (Rubas et al., 1993). This variation may be a consequence from either cell culture conditions (media, culture supplements, cell seeding, confluency or cellular differentiation), cell characteristics or varying experimental conditions. These cell culture and experimental conditions may affect passive and/or active transport across the monolayer. Moreover, subclones from the Caco-2 parental cell line have been shown to display increase/decrease expression of transporters such as taurocholic acid (Woodcock et al., 1991), sucrase isomaltase (Chantret et al., 1994), glucose transporters (Mesonero et al., 1994) and dipeptide transporter (Gres et al., 1998). Although much has been reported for in vitro-in vivo correlation for oral absorption, little has been reported on reasons of lack of correlation. From the correlation and functional evaluation, there appears to exist nucleoside transporter(s), an amino acid transporter(s) and possibly organic cation of 216 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0CTN2 type transporter present on the apical membrane of the corneal epithelium. The model carrier-mediated substrates that fall within the passive diffusion correlation may indicate that these transporters are not present/active on the tear membrane of the corneal epithelium. These transporters include the monocarboxylic acid transporter, organic cation of OCT-type transporter and dipeptide transporter. Consistent with Bonanno’s (1990) results, monocarboxylic acid transporter is present on the basolateral (aqueous humor) membrane of the corneal epithelium and not on the apical membrane (tear side). The in vitro permeability also suggests the presence of a basolateral MCT on both LCC and AIC cultured RCrECL. Contrary to our result, functional evidence of oligopeptide transporter has recently been reported in the tear side of the rabbit cornea using L-valacyclovir (Anand and Mitra, 2002). Their functional studies were carried out in whole comeal tissue. It is difficult to determine what influence the stroma and endothelium had on the permeability of various substrates used in their evaluation. Our In vitro functional characterization indicates the existence of a basolateral oligopeptide transporter and no evidence of apical membrane transporter. This correlates well with our in vivo absorption of gly- sar and L-valacyclovir appears to fall in the passive diffusional correlation. For the first time, the in vitro-in vivo correlation exhibit the possible existence of an apical Na+ -dependent nucleoside transporter and an L-arginine amino acid transporter present in rabbit comeal epithelium. In addition, the presence of a Pgp efflux mechanism previously reported to be present in the comeal epithelium (Kawazu et al., 1999) is also present in the culture model and in vivo. The in vivo 217 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. evidence of these transporter activities can be seen by inhibition of uridine, L- arginine and CsA absorption. Inhibition of mannitol absorption with ImM excess unlabeled mannitol did not change the extent of corneal epithelium, stroma- endothelium, iris-ciliary body and plasma absorption. In contrast, inhibition of uridine and L-arginine absorption with 1 mM excess unlabeled substrates decreased comeal epithelial absorption by 90% and 81%, respectively (Fig. 4-47). In addition to inhibiting comeal epithelial absorption, absorption of substrates into the stroma- endothelium, aqueous humor, iris-ciliary body and plasma were inhibited to a great extent (19-90%). Inhibition of CsA efflux with 0.5 mM propranolol increased comeal epithelial and stroma-endothelial absorption by 47% and 60%, respectively. These results suggested the presence of a Na+ -dependent nucleoside transporter, an amino acid transporter and the Pgp efflux pump using model substrates. However, the capacities of the carriers for these substrates are reduced and thus, lead to lower permeability rates across the cultured cell layers. The carrier- mediated transport mechanisms in cultured cell layers has been shown but fraction absorbed after topical administration of well actively transported compounds may be underestimated from the cell culture model (Fig. 4-46). This appears to be the case for L-arginine, L-camitine and uridine. As for CsA, the in vivo absorption of CsA appears to be predominantly passive transcellular diffusion. Although Pgp appears to be present in the cell culture and in vivo (Table 4-47), it does not appear to possess the trend seen by L-arginine, L-camitine and uridine. At first glance from the in vitro-in vivo correlation, Pgp appears to be overly expressed in cell culture but this is 218 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. not the case since the in vitro permeability of CsA only resulted in ~2-fold higher BA transport. One possible explanation for this is that binding of CsA to a cytosolic protein, cyclophilin (Kawazu et al., 1999) in cell culture vs. in vivo may be different, depending on the expression of this protein. The high lipophilicity of CsA allows for quick absorption across the apical corneal epithelial membrane. Once inside, in addition to the solubility factor of CsA, cyclophilin may also act as a depot for CsA and depending on the number of cyclophilin protein produced in the cytosol, binding of CsA may vary. Indirect evidence for difference in CsA binding between LCC and AIC cultured RCrECL can be seen from the 4 hr 3 H-CsA AB and BA transport. 3 H- CsA uptake after 4 hr (steady-state flux) from the apical and basolateral membrane for AIC cultured cells was 24.4 ± 0.82% and 8.18 ± 0.13%, respectively. As for LCC cultured cells, apical and basolateral 3 H-CsA uptake was 14.0 ± 0.51% and 8.05 ± 0.31%, respectively. These results suggest that the capacity of the corneal epithelium to accumulate CsA from the apical membrane compared to basolateral membrane is quite different, indicating the existence of different mechanisms. 2. Possible Reasons for Lack of Correlation Cell cultures are a great tool for understanding various physiological mechanisms such as ion and drug transport processes. Little is known about cornea transporter activity in tissue and culture due to the curvature of the tissue and the lack of a functional culture model, respectively. However, it has been shown in cultured bovine corneal endothelial cells that Na+ /H+ exchange activity is only about half of 219 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. its value in its freshly isolated counterpart (Bonanno and Giasson, 1992). Similarly, the corneal epithelial Na:K:2Cl cotransporter (NKCC) expression in freshly isolated cells is apparently higher than in their cultured counterpart (Jelamskii et al., 2000). In addition, after corneal epithelial wounding, mRNA and protein expression for GLUT1 is rapidly enhanced (Takahashi et al., 1996). These reports indicate that membrane protein activity exhibited in the cornea may be altered based on factor(s) in culture conditions or based on the demands of the cells (normal vs. diseased state). Recently, Bildin et al. (2001) compared the Na:K:2Cl cotransporter, glucose transporter 1 and aquaporin 1 expression in freshly isolated and cultured bovine comeal endothelium and epithelial cells. They found that these transporter expressions are greatly reduced in culture and that the comeal endothelial cell culture is passage-dependent with a 2- to 3-fold decrease in mRNA levels by passage 3. 2.1. Culture Age In light of what has been reported, it is not surprising that the in vitro protein activity may not correlate to the extent observed in in vivo. The initial indication of the effect of culture age on transporter expression is from the observed differences in permeability of model compounds across LCC and AIC cultured RCrECL (Table 4- 12). LCC cultured cells are considered a less developed/differentiated cell culture (Fig. 4-1) due to the slow growth rate associated with reduced oxygen supply as compared to AIC cultured RCrECL. Therefore, the permeability differences for L- camitine, L-arginine, CsA and uridine in different culture conditions suggests that 220 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the stage of cell development may have affected the amino acid transporter, nucleoside transporter, efflux activity and possibly the OCTN-type organic cation transporter in vitro. This was further supported by determining the Km and Vm a x of uridine uptake in AIC (Fig. 4-33A) and LCC (Fig. 4-37) cultured RCrECL. The Km for both cultures is similar (-26 pM) but Vm a x for LCC cultured RCrECL is 2-fold higher compared to AIC cultured cells. These results indicate that the high affinity Na+ -dependent nucleoside transporter activity is present in both LCC and AIC cultured RCrECL but the capacity of this transporter is lower in AIC compared to LCC cultured cells. This was further evaluated by determining culture age dependence on L-arginine (Fig. 4-48), CsA (Fig. 4-49), L-camitine (Fig. 4-50), and uridine (Fig. 4-51) uptake across both AIC and LCC cultured RCrECL. The effect of culture age on L-arginine uptake activity was evaluated on days 6, 8 and 10 in both LCC and AIC cultured cells. In LCC cultured cells, L-arginine uptake decreased by 40% and 75% from days 6 to 8 and from 6 to 10. In contrast, L- arginine uptake in AIC cultured cells increased by 80% from days 6 to 8 but decreased by 50% from days 8 to 10. The amino acid transporter activity appears higher in the earlier days for LCC cultured cells but appears to be optimal on day 8 for AIC cultured cells. The effect of culture age on Pgp efflux activity was determined by CsA uptake and inhibition of CsA uptake by verapamil. Pgp efflux activity appears to be highest in earlier culture days (day 6) for both LCC and AIC cultured RCrECL (Fig 4-49). In addition, this efflux activity appears to decrease as culture age with completely no activity by day 10. Similar age dependent expression 221 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of Pgp in Caco-2 cell monolayer was previously reported to be optimally expressed during day 17-27 of culture (Hosoya et al., 1996). The effect of culture age on L- camitine uptake activity was evaluated on days 4, 6, 8 and 9 in both LCC and AIC cultured cells. L-camitine uptake activity was completely abolished by day 8 and day 4 of LCC and AIC cultured RCrECL, respectively. The L-camitine transporter activity appears to be higher in the earlier days for LCC and AIC cultured cells. The effect of culture age on uridine uptake activity was evaluated on days 4,6, 8 and 9 in both LCC and AIC cultured cells. Uridine uptake activity appears to be optimal on day 8 for LCC cultured cells, whereas uptake was highest at the earlier days in AIC cultured cells. This culture age dependency has also been observed in primary cell cultures such as the retinal epithelial membrane protein (REMP) (Philp et al., 1995) and conjunctival epithelial cell layers (Basu, 1999). The REMP shares ~33% sequence identity with monocarboxylic acid transporter isoform MCT1, was found to be upregulated during retinal development but completely downregulated in primary culture (Philp et al., 1995). In contrast, the dipeptide transport activity increases from day 2 to day 7 after seeding (Basu, 1999). This apparently was due to possible downregulation of PepTl and upregulation of PepT2 during the culture days. Such increases in transport activity with Caco-2 culture age have also been reported for Na+ -glucose (Blais et al., 1987) and H+ -dipeptide (Saito and Inui, 1993) transporter. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. It is unclear if the age dependent transport regulation is a factor of cell type, culture conditions (e.g., media components) or secreted factors by peripheral tissues (e.g., conjunctiva, lacrimal gland, etc.). The cause of age related downregulation of specific transporter(s) is unclear. This may be an intrinsic factor that occurs normally, diseased condition or the lack of specific factors that are secreted by peripheral tissues. Currently there are age- related diseases such as the age-related polyuria (impaired ability of the aging kidney to retain water), which is associated with a downregulation of aquaporin-2 (AQP2) and -3 (AQP3) expression in the medullary collecting duct (Preisser et al., 2000). 2.2. Effect of Antibiotics and Antimycotic DMEM/F12 base media was used in the culturing of the RCrECL. This media contains inorganic salts, amino acids, vitamins, D-glucose, HEPES as a buffering agent, linoleic acid, lipoic acid, phenol red, Na putrescine, Na pyruvate, thymidine and Na hypoxanthine. In addition, it is supplemented with bovine pituitary extract, hydrocortisone, EGF, insulin, transferrin, selenious acid, bovine serum albumin and linoleic acid. Like all cell culture systems, antibiotics and antifungal agents were also added to the media. Of the supplements added, the antibiotics and antifungal agents are not naturally present in vivo. Specifically, gentamicin, penicillin, streptomycin and amphotericin B were added for the purpose of keeping the culture free from bacterial and fungal growth. Therefore, we further evaluate the effect(s) of these agents on the transport activity. For assessing the effect of gentamicin, 223 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. penicillin, streptomycin and amphotericin B on the transporter activity, the corneal epithelial cells were cultured in the absence of each agent. The agents did not appear to affect L-arginine uptake in either LCC or AIC cultured RCrECL (Fig. 4-52). As for assessing the effect of these agents on Pgp activity, in the absence of streptomycin-penicillin, CsA uptake appeared to increase by 29% (although not statistically significant), indicating a slight downregulation of Pgp activity in both LCC and AIC cultured RCrECL (Fig. 4-53). In the absence of amphotericin B and gentamicin, CsA uptake was not affected. Interestingly, uridine uptake appears to be affected by all agents and differently depending on culture conditions (Fig. 4-54). Since it was established that apical uridine uptake is mediated predominately by the Na+-dependent nucleoside transporter, the effect of antibiotics and antifungal agents observed are predominately attributed to that transporter. Secondary effects may also occur with ion and other carrier mediated transporters. For LCC cultured RCrECL, uridine uptake decreased by 32%, 41% and 53% in the absence of amphotericin B, gentamicin, and streptomicin-penicillin, respectively. As for AIC cultured RCrECL, uridine uptake decreased by 46%, 64% and 38% in the absence of amphotericin B, gentamicin, and streptomicin-penicillin, respectively. It is not surprising to see the effects of these antibiotics and antifungal agents on the transporter activity since gentimicin and amphotericin B has been reported to reduce aquaporin water channels in rat kidney (Lee et al., 2001; Kim et al, 2001). 224 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Of the transporter activities evaluated, it appears that the nucleoside transporter(s) present on the apical membrane of the comeal epithelium is sensitive to both culture age and antibiotics/antifungal agents in the culture media. The regulatory aspects of nucleoside transporters are well reported in the literature (Fig. 5-1). In addition, hypoxia has been implicated in the reduced function of the rat equilibrative nucleoside transporter, rENTl (Kobayashi et al., 2000). The numerous factors regulating this class of transporter makes it not at all surprising to see the regulatory response from the comeal epithelium. Starvation Nucleotide Deprivation Jejunal Enterocyte (Soler eta/., 2001) B-Cell lines Starvation Development TNFa CNT2| |EN T1|© Hepatic regeneration CNT2 . Differentiating / Agents (Dx+T4) Cell cycle progression Hepatocyte and Hepatocyte-derived Cell lines Activation (IFNy) © Proliferation (MCSF) Macrophages Fig. 5-2: Known regulatory features of nucleoside transporters in jejunal epithelial cells, hepatocytes and hepatocyte-derived cell lines, B cell lines and bone marrow macrophages. Only those transporters known to be regulated are presented in each cell model, (information from Pastor-Anglada et al., 2001). Note: © Simulation and © inhibition. 225 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VI. CONCLUSION 226 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A. SUMMARY OF FINDINGS A relatively simple method for cultivation of a tight primary RCrECL was developed and characterized. The growth conditions that would favor the development of a tight functional RCrECL grown on a permeable support were assessed and identified. Culturing corneal epithelial cells under an air-interfaced condition was found to be a key growth factor that resulted in a tight comeal epithelial cell layers and was found to resemble tissue comeal epithelium characteristics in bioelectrical (barrier integrity), cell architect (morphology) and ion transport properties (functional activity). In addition to culturing the epithelial cells under an air- interfaced condition (AIC), omitting fetal bovine serum and keeping the concentration of epidermal growth factor at 1 ng/ml were equally important. Compared with cells cultured under the liquid-covered (LCC) condition, those cultured under AIC exhibited a significantly higher peak transepithelial electrical resistance (TEER) of up to 5 kQ.cm2 , a higher potential difference (PD) of up to 26 mV, and an estimated short-circuit current (Ie q ) of 5 p.A/cm2 after 7-8 days of culture. These values were comparable to those in the excised cornea. Consistent with the TEER, the AIC cell layers were 4-40 times less permeable to paracellular markers than their LCC counterpart. Phenotypically, the AIC cell layers were found to express cornea-specific 64 kD keratin. In addition, AIC cultured RCrECL exhibited all three cell types that are unique to the comeal epithelium (basal, wing and superficial cell layers). The baseline Isc was contributed by active Na+ and Cl' 227 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. transport under a short-circuit condition at 64 and 44%, respectively. In addition, the presence of apical Na+ and CF channels and basolateral Na+ /K+ ATPase, K+ channel and Na+ /Cl' cotransporter were also present in the AIC cultured RCrECL. Taken together, the improved growth and differentiation of the air-interfaced cultures, along with bioelectric, morphology and ion transport properties that resemble the native comeal epithelium, make this the model of choice for the study of dmg transport processes. In the second stage of this research, the in vitro dmg transport mechanisms were evaluated. First, the passive transcellular mechanism was evaluated. The influence of solute lipophilicity in AIC cultured RCrECL was compared with literature values reported for LCC cultured comeal epithelial cells and excised comeal tissue. Using P-blockers of same molecular weight, permeability was found to be dictated by lipophilicity. As expected, AIC cultured cells resulted in a sigmoidal relationship between dmg permeability and lipophilicity with a half maximal permeability occurring at a log partition coefficient of 2.58. These permeability characteristics are similar to that found in the native tissue or other cell culture/tissue, which validates this culture model for studying diffusional dmg transport processes. The next step was to probe for active transport mechanisms across AIC and LCC cultured RCrECL employing widely used substrates such as L- arginine (amino acid substrate), L-camitine (OCTN type substrate), gly-sar (oligopeptide substrate), guanidine (OCT substrate), lactic acid (monocarboxylic acid substrate), uridine (nucleoside substrate), and CsA (efflux substrate). Mannitol 228 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. was used as a paracellular marker. The permeability characteristics of these substrates were evaluated in both LCC and AIC cultured RCrECL for the determination of the effect of culture conditions on transporter activity. From the initial permeability characteristics (directionality, temperature effect, and comparison with mannitol), it appears that amino acid, nucleoside, monocarboxylic acid and Pgp efflux transporters may be present on the cultured corneal epithelium. In addition, the extent of transport appears to be governed by culture conditions (AIC vs. LCC). In order to determine if these permeability values were reliable in predicting possible transporter activity, uridine, gly-sar and L-camitine were functionally evaluated. The initial evidence of the oligopeptide transporter was from the result of a net BA transport of gly-sar across AIC cultured RCrECL. Functional characteristics of gly-sar uptake was determined in both apical and basolateral AIC cultured RCrECL. The basolateral uptake of gly-sar was rapid, temperature sensitive and requires energy. In addition, this uptake was not affected by extracellular pH and is inhibited by excess gly-sar and structurally similar di/tripeptides. Basolateral uptake of gly-sar was saturable (Km = 1.49 ± 0.06 mM and Vm a x = 104 ± 0.93 pmol/mg protein/min). In addition, the differential recognition of cyclacillin and cefadroxil results indicated that this basolateral oligopeptide transporter might be more like PepTl than PepT2. Cyclacillin was found to be 14-fold more potent than cefadroxil indicating the possible function of a PepTl-like transporter on the basolateral membrane of the corneal epithelium. Unfortunately, the molecular identity of known basolateral oligopeptide transporters (e.g., intestine) has not been elucidated. There 229 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. does not appear to be an apical dipeptide transport activity present in AIC cultured RCrECL. From the permeability characteristics, it appears that there exists nucleoside transporter(s) on both the apical and basolateral membrane. Although AB and BA transport of uridine across AIC cultured RCrECL did not result in net directional transport, the permeability values were significantly higher compared to mannitol permeability and under 4°C indicating the possible existence of apical and basolateral transporters. The functional activity of nucleoside transporters on both the apical and basolateral membrane of AIC cultured RCrECL was further determined using 3 H-uridine. Uptake of uridine was rapid and temperature dependent into both apical and basolateral uptake. Interestingly, the net basolateral uptake appears to be 2-fold higher compared to apical. This was contrary to the no directional transport in the initial evaluation. Apical uptake was strongly dependent on the presence of Na+ and insensitive to the nucleoside transport inhibitor, NBMPR. Basolateral on the other hand was not dependent on Na+ and strongly dependent on the presence of NBMPR. The apical Na+ dependent uridine uptake was saturable (Km = 28 pM, V m ax = 8.4 pmol/mg protein/min) and insensitive to NBMPR. The basolateral Na+ independent uridine uptake was also saturable (Km = 2.08 mM, Vm a x = 2.40 nmol/mg protein/min). All physiological nucleosides were effective in competing with uridine uptake from both the apical and basolateral compartment. Inhibition of Na+ -independent uridine uptake by NBMPR in basolateral AIC cultured RCrECL resulted in a biphasic dose-response inhibition profile indicating the 230 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. coexistence of es (ENT1) and ei (ENT2) type Na+ -independent nucleoside transporters. Since the nucleotide identity of rabbit ENT2 is known, the molecular identity of this equilibrative nuceloside transporter was further supported by RT- PCR in tissue cornea epithelium cDNA. The predominant Na+ -dependent nucleoside transporter present on the apical membrane appears to be CNT 3 or both CNT1 and CNT3. The Na+-dependent uridine uptake from apical membrane of LCC cultured RCrECL was saturable (Km = 25 pM, V m ax = 17 pmol/mg protein/min). The Km appears to be the same as the AIC cultured cells but Vm a x doubled compared to AIC culture. This result indicates that the apical nucleoside transporter is predominantly the Na+ -dependent type and that the 6-fold lower permeability at 37°C across AIC compared to LCC cultured cells appears to be the result of a lower number of transporter present in AIC compared to LCC culture. The overall nucleoside transport process across the corneal epithelium involves the rapid internalization of nucleosides and nucleoside analogs via the I n dependent nucleoside transporter (CNT3 or CNT1 and CNT3) present on the apical membrane. Once accumulated intracellularly, the efflux of nucleosides down a concentration gradient to the basolateral compartment via the equilibrative nucleoside transporter (both ENT1 and ENT2) will complete the net flux of nucleosides from tear to stroma. The fact that high affinity Na+ -dependent nucleoside transporter is present on the apical membrane and Na+ -independent equilibrative nucleoside transporters on the basolateral membrane provides significant clues to the physiological role of these transporters in meeting the needs 231 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of the comeal epithelium. The Na+ -dependent nucleoside transporter would contribute to the net Na+ -absorption across the apical membrane, which probably contributes to the 15-20% of the amiloride-insensitive Isc observed in the ion transport characterization. In addition, the presence of a high affinity transporter on the apical membrane may indicate that the comeal epithelium can selectively salvage nucleosides from cellular debris to accommodate the high energy requirement of the epithelium. The lack of a bidirectional equilibrative nucleoside transporter on the apical membrane may indicate that the substrate gradient is most likely from apical to basolateral. The bidirectional property of ENTs may also allow for the uptake of nucleosides from the stroma across the basolateral membrane into the comeal epithelium in case the high affinity transporter on the apical membrane is nonfunctional, providing a check and balance system to ensure that nutrients are supplied to the comeal epithelium. Functional characterization of L-camitine suggests the presence of an apical amino acid transporter due to the saturable kinetic value similar to that reported for the amino acid transporter (not the high affinity OCTN transporter), possibly the B0 ,+ system. Apical uptake of L-camitine was inhibited by L-arginine and not by organic cations (L-camitine, guanidine, TEA, verapamil and choline) or p-lactam antibiotics (cimetidine and cyclacillin). Basolateral inhibition of L-camitine uptake by L-arg, organic cations (L-camitine, TEA, verapamil and choline) and P-lactam antibiotics indicate that the basolateral transporter may have broad substrate specificity or that there are different transporters that have overlapping substrate specificity. These 232 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. results suggest the presence of an amino acid transporter, possibly the B0 ,+ system that can mediate absorption of cationic/neutral amino acid as well as L-camitine from the tear membrane. The final stages of this research include the determination of in vitro-in vivo correlation of the transport mechanisms, which were determined for a total of 24 compounds ranging in different lipophilicity as well as various model carrier- mediated substrates. These transport processes include passive paracellular, passive transcellular and active transcellular transport. The passive transport correlation was first determined with the 12 P-blockers and a good linear correlation with an r of 0.95 was established. Subsequently, other polar and carrier-mediated substrates such as mannitol, acyclovir, cidofovir, arginine, carnitine, lactic acid, guanidine, gly-sar, uridine, valacyclovir, verapamil and cyclosporin A were included into the P-blocker in vitro-in vivo correlation. All compounds evaluated fell within the passive transport correlation except L-arginine, L-camitine, CsA and uridine. These compounds showed significantly higher in vivo comeal absorption than would have been predicted by in vitro permeability. To determine if these model substrates were actively absorbed into the comeal epithelium, in vivo absorption of these substrates was inhibited by excess unlabeled substrates. In vivo comeal epithelial absorption of uridine and L-arginine was inhibited by 1 mM unlabeled substrates by 90 and 81%, respectively. No inhibition was observed for mannitol. As for detecting Pgp efflux mechanism, the addition of 0.5 mM propranolol enhanced CsA absorption by 47%. It appears that these transporters are functional in vivo. 233 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Factors such as culture conditions, culture age and the presence of antibiotics and antifungal, were found to possibly affect functional expression of nucleoside, amino acid and Pgp efflux transporters. All activities appear to slowly decrease as a function of culture age. Gentamicin, streptomicin-penicillin and amphotericin B did not appear to affect the amino acid and Pgp efflux transporter function. The nucleoside transporter appears to be quite sensitive to these agents. Apical (Tears) Basolateral (Stroma) Na+ AA Cl- Cationic and neutral Na+ Nucleosides Na+ Na+ K + Di peptides tri peptides peptidom im etics K+ ENT' .e i/e s , H+ * (mct) Lac— ► N ucleosides Monocarboxylic acids / ' n— Na+ < Cl- ► Na+ Cl- Fig. 6-1: Summary of the active ion and drug transport processes in the rabbit corneal epithelium. Pgp = P-glycoprotein; B0 ,+ = amino acid transport(s); CNT = nucleoside transporter(s); ENT = equilibrative nucleoside transporters; OT = Oligopeptide transporter(s); MCT = monocarboxylic acid transporter(s) 234 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. B. SIGNIFICANCE OF THE FINDINGS This research project has resulted in the development of a tight functional primary cell culture model of air-interfaced rabbit comeal epithelium. This culture model exhibits properties unique to native comeal epithelium in cellular architect and function and thus, provides a great tool for evaluating ion and drug transport mechanisms at the biochemical and cellular levels. There are three significant findings that resulted from this dissertation. First, the finding that by culturing the RCrECL under air-interfaced condition which allowed for direct exposure of oxygen, resulted in tight comeal epithelial cell layers. Secondly, the transport pathways (paracellular, transcellular and carrier-mediated) for various model substrates were evaluated as well as the effect of culture condition on these transport pathways. Carrier-mediated transport processes such as the nucleoside and amino acid transport activities were identified on the tear membrane of the comeal epithelium that can be utilized for drug targeting/delivering of nucleoside or amino acid analogs as therapeutics across the tight comeal epithelium. Thirdly, our in vitro culture model resulted in great correlation with in vivo absorption for passively and to some extent actively absorbed drugs. The in vitro model may be used as a predictive model of in vivo absorption. In addition, some of the key regulatory aspects of carrier-mediated transport mechanisms present in the comeal epithelium were identified. Findings from the cell culture development and characterization provided insight into the conventional method of culturing comeal epithelium under the LCC 235 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. condition. Although some investigators have reported enhanced features of AIC comeal cultures (Liminga and Oliw, 2000), it has not been explored extensively. The LCC culture model appears to represent a less differentiated culture model that can be used for studying cell types (e.g. wing or basal cells) other than the superficial cell layers. In addition, this culture model may also represent a condition that occurs when the comeal epithelium is low or deprived of oxygen; as in the case during contact lens wear. Acidification of tissues such as comeal epithelium (Giasson and Bonanno, 1994), stroma (Bonanno and Poise, 1987), endothelium (Giasson and Bonanno, 1994), aqueous humor (Giasson and Bonanno, 1994) and tears (Chen and Maurice, 1990) has been associated with contact lens wear. The cornea is acidified by anaerobic metabolic products and carbon dioxide retention. This acidification is the result of anaerobic metabolic products and carbon dioxide retention, which results in changes in intracellular pH, which can have significant alteration in glycolytic activity, ion transport activity, mitochondrial respiration and possibly carrier-mediated transport mechanisms. In particular, it was first reported that acidosis resulted in an increase in intracellular sodium that is due to the activation of the Na+ /H+ exchanger mechanism (Bonanno and Giasson, 1992; Bonanno and Machen, 1989). It was later clarified that, acidosis resulted in an increase in comeal oxygen consumption which is in part secondary to the activation of the Na+ /H+ exchanger, which then stimulates Na+ /K+ ATPase activity (Harvitt and Bonanno, 1998). It is still unclear whether other ion or drug transport mechanisms are associated with comeal oxygen consumption and thus, both the AIC and LCC 236 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cultured RCrECL models might be useful to understand such cellular processes in healthy and diseased states. The development of a functional culture model set the stage for identifying/determining drug permeability mechanisms in the second phase of this dissertation. Elucidation of the permeability characteristics of the comeal epithelium to drugs of varying model substrates allowed for the identification of various carrier- mediated transport mechanisms such as nucleoside and amino acid transport systems. Using this culture model, concentrative and equilibrative nucleoside transporters were found to be present on the tear and stromal membrane of the comeal epithelium, respectively. In addition, an amino acid transporter also appears to be present on the tear and stromal membrane. Physiologically, this has great implications and suggests that nutrients are supplied from both the precorneal and stromal (or aqueous humor) area. Not surprisingly, there appears to be more transporters on the stromal (basolateral) membrane. In addition, these basolateral transporters appear to be of low affinity types compared to apical (tear) membrane transporters. Therapeutically, these transporters can be dmg delivery targets. As comeal epithelium is the primary barrier to the absorption of hydrophilic drugs after topical administration, the presence of these mechanisms will offer a strategy for the design of transporter-targeted drugs having enhanced permeability. In this dissertation, we have begun to categorize the different classes of ion and dmg transport mechanisms. These results may shed light on the feasibility of carrier- mediated dmg targeting, thus setting the stage for selecting hydrophilic dmg 237 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. candidates that exhibit favorable permeability across the corneal epithelium. Another application of enormous potential is in the field of genomics and proteomics in diagnosis and treatment of corneal related diseases. This will allow for the use of genomics and proteomics in early detection and therapeutic treatments. The application of proteomic technology is generally quite new to ophthalmic research and is associated with confounding problems due in part to the limited availability of tissue and the problems associated with culturing ocular cells. This dissertation has identified some key elements in culturing corneal epithelium and thus, will contribute to proteomics in ophthalmic research. The third area of significant contribution is associated with demonstrating the in vitro-in vivo correlation of the transport mechanisms for a total of 24 compounds ranging in different lipophilicity as well as various model carrier-mediated substrates. These studies indicate that in vivo absorption of passively absorbed drugs can be predicted by in vitro permeability. In addition, the nucleoside and amino acid transporters present on the apical membrane of the AIC cultured RCrECL are also present and functional in vivo. For the substrates that were underestimated by in vitro permeability, it was determined that these transporters did not mimic to the extent of in vivo activity. The fact that in vivo absorption of L-camitine was to the extent that L-arginine was absorbed indicates that both substrates may be absorbed by the same transporter (e.g. ATB0 ,+ transport system). This transporter activity can be affected by culture condition, culture age and possibly by regulatory factors that are present in vitro and not present in vivo. In addition, these transporters can be 238 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. regulated differently based on the functional need of these transporters. For example, the nucleoside transporter was affected by presence of antibiotics and antifungal agents, whereas, the amino acid transporter was not. For all transporters evaluated, culture age appears to affect all transporters. This finding may suggest that as the cells become more differentiated (into superficial cell layer), the transporter activities may naturally decrease or if there is a regulatory factor that is secreted by peripheral tissues that preserve the activities of these transporters regardless of the cell differentiation state. C. FUTURE CONSIDERATIONS This dissertation resulted in the development of a functional tight corneal epithelial cell culture model for assessing penetration mechanism(s) of hydrophilic drugs across the rate-limiting barrier. In addition, this research has provided evidence of carrier-mediated transporters present on the corneal epithelium in both in vitro and in vivo. The following area warrants further consideration for investigation: 1. Cell Culture Model Although in vivo expressed transporters were identified using the functional in vitro model, these transporter activities (e.g., amino acid and nucleoside transporters) appear to be downregulated in culture. It was determined in this dissertation that factors such as culture condition, culture age and antibiotics/antifungal can affect the functional activities of transporters present on the corneal epithelium. This is not 239 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. surprising considering the complexity of in vivo conditions. This suggests that there is room for further improvement on the in vitro culture model. For one, the cells were cultured with complete air-interface condition. When in reality, the superficial cell layer is covered by a layer of precorneal tear film consisting of a water-mucin interface, mucin-aqueous and a lipid layer (Fig. 6-2). The functions of the tear film are (1) optical, (2) protective, (3) moisturizing and (4) it provides trophic support to comeal epithelium (Table 6-1). In addition, the significance that the closing (during sleep) and opening (when awake) of the eye may have on the transporter activity is unclear. Despite many advances in the field of ocular surface disease, the role of the tear fluid in maintaining a healthy ocular surface has not been defined. It is unknown whether tear fluid constituents influence the growth and differentiation of the ocular surface epithelium, and if they do, whether this is a primary or redundant function. Furthermore, it is not clear whether the concentration or activity of biologically active tear factors changes in dmg eye conditions. What we know from basic and clinical research is that the lacrimal gland secretes factors with potential biological activity into the tear fluid, that the concentration of tear fluid cytokines such as EGF is regulated by sensorineural stimulation and the level of aqueous tear production, and that tear fluid has potent antiproliferative activity. The potential role of the tear fluid on the ocular surface can also be inferred from knowledge that disease of aqueous tear deficiency is greatest in the exposure zone that is subject to microtrauma, and that reduced 240 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (triglycerides, tree fatty acids and phospholipids) Mucin-Aquous Layer Water Mucin Interface Lipid V -10 pm Corneal Epithelium Fig. 6-2: The trilaminar components of the precorneal tear film. Corneal and conjunctival epithelial cells secrets mucin rich glycocalyx. The mucin layer imparts a hydrophilic component to the cell surface, facilitating even distribution of the over laying aqueous tear layer produced by the lacrimal gland. Lipids are produced by the meibomian glands of the lids and function to retard evaporation of the aqueous layer. aqueous tear production and/or turnover results in ocular surface inflammation. As in the case for Caco-2 cell culture models that continue to undergo modification (from new technologies to new ideas) to improve its characteristics, this culture model warrants further assessment in the regulatory aspects of carrier- mediated transport mechanisms present on the comeal epithelium. 2. Corneal Epithelial Transporters Both in vitro and in vivo studies in this dissertation indicate the presence of nucleoside and amino acid transporters on the tear membrane and monocarboxylic acid and oligopeptide transporter on the basolateral (stroma) membrane of the comeal epithelium. In addition, the apically determined nucleoside and amino acid transporters also are present on the basolateral membrane. The nucleoside transporters are of concentrative-type on the apical and facilitative-types on the 241 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 6-1: Components of normal tear and serum. C o n c en tra tio n C o m p o n e n ts T e a r S eru m E lectrolytes Na+ 145 mEq/L 135-146 mEq/L K+ 24.1 mEq/L 3.5-5.0 mEq/L Ca2 + 1.5 mM 1.1 mM c r 128 mM 96-108 mM H C 03‘ 26 mM 21-29 mM n o 3‘ 0.14 mM 0.19 mM P 0 4 3' 0.22 mM 1.42 mM S O 42' 0.39 mM 0.53 mM Proteins Total protein 7.37 g/L 68-82 g/L Lysozyme 2.39 g/L 4.0-15 g/L Lactoferrin 1.51 g/L ND Albumin 54 g/L 35-55 g/L IgA 411 g/L 0.9-4.5 g/L igG 32 g/L 8-18 g/L CuZn-SOD 103 ng/mg protein ND (in red blood cells at high conc.) Carbohydrate Glucose 26 mg/L 0.6-1.2 g/L Growth Factors EGF 1.66 ng/mL 0.72 ng/mL TGF-a (male) 247 pg/mL 147 pg/mL TGF-a (females) 180 pg/mL (male and fem ales) T G F - p i (acid activated tear)2 .3 2 ng/mL TGF-P2 55 pg/mL Vitamin Vitamin A 16 ng/mL 883 ng/mL Vitamin C 117 pg/mL 7-20 pg/mL Antioxidant Tyrosine 45 pM 77 pM Glutathione 107 pM ND (in red blood cells at high conc.) CuZn-SOD = Cu.Zn super-oxide dism utase; EGF=epidermal growth factor; TGF=transforming growth factor; ND=not detected. (Table modified from Tsubota eta!., 2002) basolateral membrane. Our functional evidence indicates that the apical amino acid transporter is of the B0 ,+ system but it is unclear what type is on the basolateral membrane. Therefore, the following topics leaves room for further evaluation: 1. Need to identify molecular identity of these transporters. 2. Need to evaluate functional activity of these transporters, especially substrate 242 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. specificity (structural requirement). 3. What is the ocular tissue distribution of these transporters? 4. Is the decrease in transporter activity related to the number of transporters? 5. Are the cell differentiation states (or cell types) a function of inhibition of cellular protein synthesis? By identifying and understanding the transporters at the functional and molecular level would allow for early detection and therapeutic treatment of the eye diseases. It was interesting to see that the B0 ,+ amino acid transporter also transports L-camitine. L-camitine is a small molecule essential for long-chain fatty acid transfer into the mitochondrial where (3-oxidation occurs. Since L-camitine is metabolically stable in eukaryotic cells, L-camitine homeostasis in mammals is regulated by: 1) a modest rate of endogenous synthesis; 2) absorption from dietary sources; and 3) efficient membrane carriers that are able to modulate intracellular and extracellular carnitine pools. Questions such as how is the dual properties of amino acid transporter relevant to the physiological role in the eye-specifically in the comeal epithelium, can be addressed. 3. Normal vs. Disease Condition This dissertation provided evidence of carrier-mediated transport activity in normal situations. It is unclear if disease conditions (e.g. wound healing, diabetes, hypoxia, viral/bacterial infections, etc.) would alter transporter expression or function. Below are proposed issues to be addressed: 243 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (1) Degree of transporter expression or activity in disease states (acute and chronic) compared to normal. (2) What are the effects (e.g. leaky barriers, ion transport alteration, etc.) of the diseased condition? States of acute and chronic diseases such as hyperglycemia can give rise to pathophysiological processes in various structures of the eye. Within the eye, glucose gradients are thought to be maintained by the sodium-independent glucose transporter (GLUT1) (Mueckler, 1990), which maintains corneal endothelial and epithelial functions especially during wound healing. Comeal wound healing involves a number of metabolically demanding processes such as cell migration and proliferation that are in high energy demand provided by glycolysis. This high energy demand was found to be associated with GLUT1 expression (Takahashi et al., 1996). Comeal hypoxia is the result of oxygen deprivation to the cornea, which has been shown to be associated with contact lens wear. It was determined that hypoxia can increase expression and activity of the glucose transporter (GLUT1) in alveolar epithelial cells (Ouiddir et al., 1999). In addition, it was recently determined by analysis of verapamil-inhibitable efflux of digoxin and rhodamine 123 in intact T84 cells that P-gp function can be enhanced by as much as 7-fold over normoxia (Comerford et al., 2002). This study suggests that hypoxia-induced P-gp expression may represent a pathway for resistance of some tumors to chemotherapeutics. 244 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4. Multiple/Selective Targeting In regard to targeting transporters present on the tear surface of the corneal epithelium, the conventional method is to identify a specific transporter for target but it appears that one may better benefit from targeting multiple transporters that have overlapping substrate specificity. For example, nucleoside and amino acid transporter appears to be present on the tear membrane of the comeal epithelium. One may be able to target both transporters by constructing an amino-nucleoside analog that targets both the amino acid and nucleoside transporters. This dissertation lays the foundation for further exploration of carrier-mediated transporters present on the comeal epithelium and to determine substrate specificity for screening of target compounds. The main focus of this dissertation has been to target the anterior chamber related diseases, which makes up -60% of all ocular diseases. The other 40% of eye diseases occur in the posterior chamber. Selective targeting of either the anterior or posterior chamber diseases may allow for a more direct and localized delivery of therapeutics. This would require the knowledge of specific transporters in particular ocular tissues such as the comeal and conjunctival epithelium. The comeal route would allow for dmg delivery to the anterior region of the eye and conjunctiva would allow for targets to the back of the eye. Table 6-2 is a list of ion and dmg transporters identified to date. The ion transport systems can have secondary effect on dmg transporter mechanisms and thus, are important in the targeting selection. 245 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 6-2: Corneal and Conjunctival Transporters for Drug Targeting Location Cornea Conjunctiva Drug Transporter Aquaporin P (Patil etal., 1997) N.D. Na+ amino acid cotransporter A (Liaw et al., 1991) A (Kompella et al., 1995) A/B (this dissertation) A (Hosoya etal., 1997) Dipeptide transporter A (Anand & Mitra, 2002)a A (Basu etal., 1998) Na+ glucose cotransporter NPb A (Kompella etal., 1992) A (Hosoya etal., 1996) Facilitative glucose transporter G (Takahashi etal., 1996) Monocarboxylic acid transporter B (Bonanno, 1990) A (Horibe etal., 1998) B (this dissertation) Na+ -Nucleoside transporter A/B (this dissertation) A (Hosoya etal., 1998) Organic cation transporter OCT-type NP-apical (this dissertation) A (Ueda etal., 2000) OCTN-type P-glycoprotein efflux pump1 A (Kawazu etal., 1999) A (Saha etal., 1998) Taurine transporter A (Shioda etal., 2002) N.D. Ion Transporter C a2 + channel P (Huff and Reinach, 1985; Rich and Rae, 1995) N.D. L-type gene (Shepard and Rae, 1998) N.D. Cl‘ channel A (Marshall and Hanrahan, 1991) A (Kompella etal., 1993) CIC-3 gene (Shepard and Rae, 1998) N.D. K+ channel B (Rae, 1990) B (Kompella etal, 1992) IRK1 gene (Shepard and Rae, 1998) N.D. Na+ channel amiloride-sensitive A (Midelfart, 1987; this dissertation) NPb TTX NP (Shepard and Rae, 1998) N.D. Na+ -K+ -CI' cotransporter B (Huang etal., 1983) B (Kompella etal., 1993) NKCC1 & NKCC2 g en es (Bildin et al., 2001) N.D. Na+ -HC0 3 cotransporter G (Shepard and Rae, 1998) N.D. CI7 HC0 3 ' exchanger P (Bonanno and G iasson, 1992) N.D. G (Shepard and Rae, 1998) N.D. K+ /H+ exchanger P (Bonanno, 1991) N.D. G (Shepard and Rae, 1998) N.D. Na+ /C a2+ exchanger P (Rich and Rae, 1995) N.D. Na+ /H+ exchanger B (Bonanno and Machen, 1989) N.D. C a2 + -ATPase P (Reinach etal., 1991) N.D. Na+ /K+ -ATPase B (Klyce and Wong, 1977) B (Kompella etal., 1993) A = apical, B = basolateral, P = present (detected but not identified), NP = not present, G = gen e detected N.D. = not determ ined anot observed in this Ph.D. dissertation bKompella, Kim and Lee, unpublished observation 246 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VII. REFERENCES 247 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Adler, K.B., Cheng, P.W. and Kim, K.C. (1990). Characterization of guinea pig tracheal epithelial cell maintained in biphasic organotypic culture: cellular composition and biochemical analysis of released glycoconjugates. Am. J. Respir. Cell Mol. Biol. 2, 145-154. Ahmed, I. Gokhale, R.D., Shah, M.V. and Patton, T.F. (1987). Physicochemical determinants of drug diffusion across the conjunctiva, sclera, and cornea. J. Pharm. Sci., 76, 583-587. Ahmed, I., and Patton, T.F. (1987). Disposition of timolol and inulin in the rabbit eye following corneal versus non-comeal absorption. Int. J. Pharm., 38, 9. Amasheh, S., Meiri, N., Gitter, A.H., Schoneberg, T., Mankertz, J., Schulzke, J.D., and Fromm, M. (2002). Claudin-2 expression induces cation-selective channels in tight junctions of epithelial cells. J. Cell Sci. 115(24), 4969-4976. Anand, B.S. and Mitra, A.K. (2002). Mechanism of comeal permeation of L-valyl ester of acyclovir: Targeting the oligopeptide transporter on the rabbit cornea. Pharm. Res. 19(8), 1194-1202. Anderson, J.A. (1980). Systemic absorption of topical ocularly applied epinephrine and dipivefrin. Arch. Ophthalmol., 98, 350. Araie, M., and Maurice, D.M. (1987). The rate of diffusion of fluorophores through the comeal epithelium and stroma. Exp. Eye Res. 44,73-87. Artursson, P. (1990). Epithelial transport of drugs in cell culture. I: A model for studying the passive diffusion of drugs over intestinal absorptive (Caco-2) cells. J. Pharm. Sci. 79, 476- 482. Artursson, P. (1991). Cell cultures as models for dmg absorption across the intestinal mucosa. Cri. Rev. Ther. Drug Carrier Syst. 8(4), 305-330. Artursson, P. and Karlsson, J. (1991). Correlation between oral dmg absorption in humans and apparent drug permeability coefficients in human intestinal epithelial (Caco-2) cells. Biochem. Biophys. Res. Commun. 1753), 880-885. Ashton, P., Wang, W. and Lee, V. H. L. (1991). Location of penetration and metabolic barriers to levobunolol in the pigmented rabbit. J. Pharmacol. Exp. Therap. 259, 719-724. Aukunum, J.V., Sunkara, G., and Kompella, U. (2000). Invest. Ophthalmol. Vis. Sci. 41, S772. Babizhayev, M.A. (1996). Failure to withstand oxidative stress induced by phopholipid hydroperoxides as a possible cause of the lens opacities in systemic diseases and ageing. Biochim. Biophys. Acta. 1315, 87-99. Babizhayev, M.A., Sequin, M.C., Gueyne, J., Evstigneeva, R.P., Ageyeva, E.A. and Zheltukhina, G.A. (1994). L-camosine (beta-alanyl-L-histidine) and carcinine (beta- 248 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. alanylhistamine) act as natural antioxidants with hydroxyl-radical-scavenging and lipid- peroxidase activities. Biochem. J. 304, 509-516. Balakrishnan, A., Jain-Vakkalagadda, B., Yang, C., Pal, D. and Mitra, A.K. (2002). Carrier mediated uptake of L-tyrosine and its competitive inhibition by model tyrosine linked compounds in a rabbit corneal cell line (SIRC)—strategy for the design of transporter/receptor targeted prodrugs. International Journal of Pharmaceutics. 247, 115- 125. Barker, P.M., Boucher, R.C., and Yankaskas, J.R. (1995). Bioelectric properties of cultured monolayers from epithelium of distal human fetal lung. Am. J. Physiol 268, 270-277. Basu, S.K. (1999). Biochemical, molecular biological and computer modeling studies of the conjunctival dipeptide transporter isoforms. Ph.D. Dissertation. University of Southern California, Los Angeles. Basu, S.K., Haworth, I.S., Bolger, M.B. and Lee, V.H. (1998). Proton-driven dipeptide uptake in primary cultured rabbit conjunctival epithelial cells. Invest. Ophthalmol. Vis. Sci. 39(12), 2365-2373. Bell, T.A.G., and Hunnisett, A.G. (1986). Cyclosporin A: tissue levels following topical and systemic administration to rabbits. Br. J. Ophthalmol. 70, 852-855. Belt, J.A., Marina, N.M., Phelps, D.A. and Crawford, C.R. (1993). Nucleoside transport in normal and neoplastic cells. Adv. Enzyme Regul. 33, 235-252. BenEzra, D., and Maftzir, G. (1990). Ocular penetration of cyclosporin A: the rabbit eye. Invest. Ophthalmol. Vis. Sci. 31, 1362-1366. Bildin, Y.N., Iserovich, P., Fischbarg, J., and Reinach, P.S. (2001). Differential expression of Na:K:2Cl Cotransporter, glucose transporter 1, and aquaporin 1 in freshly isolated and cultured bovine corneal tissues. Exp. Biol. Med. 226(10), 919-926. Blein-Sella, O., and Adolphe, M. (1995). SIRC cytotoxicity test. Methods Mol. Biol. 43, 161-167. Boldyrev, A.A., Bulygina, E.R., Volynskaia, E.A., Kurella, E.G. and Tiulina, O.V. (1995). The effect of hydrogen peroxide and hypochlorite on brain Na, K-ATPase activity. Biokhimiia. 10, 1688-1696. Boll, M., Herget, M., Wanger, M., Weber, W.M., Markovich, D., Biber, J., Clauss, W., Murer, H. and Daniel, H. (1996). Expression cloning and functional characterization of the kidney cortex high-affinity proton-coupled peptide transporter. Proc. Natl. Acad. Sci. 93, 284-289. Bonanno, JA. And Machen, T.E. (1989). Intracellular pH regulation in basal corneal epithelial cells measured in corneal explants: characterization of Na/H exchange. Exp. Eye. Res. 49, 129-142. 249 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bonanno, J.A. (1990). Lactate-proton cotransport in rabbit corneal epithelium. Curr. Eye Res. 9, 707-712. Bonanno, J.A., and Giasson, C. (1992). Intracellular pH regulation in fresh and cultured bovine corneal endothelium. I. Na+ /H+ exchange in the absence and presence of HC03 \ Invest. Ophthalmol. Vis. 33(11), 3058-3067. Bonanno, J.A., Klyce, S.D. and Cragoe E.J. Jr. (1989). Mechanism of chloride uptake in rabbit corneal epithelium. Am. J. Physiol. 257(2.1), C290-C296. Bonanno, J.A. and Poise, K.A. (1987). Corneal acidosis during contact lens wear: effects of hypoxia and C02. Invest. Ophthalmol. Vis. Sci. 28(9), 1514-1520. Bonanno, J.A. and Poise, K.A. (1987). Effect of rigid contact lens oxygen transmissibility on stromal pH in the living human eye. Ophthalmology. 94, 1305-1309. Borderie, V.M., Kantelip, B.M., Delbosc, B.Y., Oppermann, M.T., and Laroche, L. (1995). Morphology, histology and ultrastructure of human C31 organ-cultured corneas. Cornea. 14(3), 300-310. Borok, Z., Hami, A., Danto, S.I., Zabski, S.M., and Crandall, E.D. (1995). Rat serum inhibits progression of alveolar epithelial cells toward the type I cell phenotype in vitro. Am. J. Respir. Cell Mol. Biol. 12(1), 50-55. Borst, P., Evers, R., Kool, M. and Wijnholds, J. (2000). A family of drug transporters: the multidrug resistance-associated proteins. J. Natl. Cancer Inst. 92, 1295-1302. Bradford, M.M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248- 254. Brown, R.M. and Middleton, C.A. (1981). Contact-induced spreading in cultures of comeal epithelial cells. J. Cell. Sci. 51, 143-152. Brown, R.M. and Middleton, C.A. (1985). Morphology and locomotion of individual epithelial cells in culture. J. Cell. Sci. 78, 105-115. Broxterman, H.J., Giaccone, G. and Lankelma, J. (1995). Multidrug resistance proteins and other dmg transport-related resistance to natural product agents. Curr. Opin. Oncol. 7, 532- 540. Burckhardt, G. and Wolff, N.A. (2000). Structure of renal organic anion and cation transporters. Am. J. Physiol. Renal. Physiol. 278, F853-F866. Burstein, N.L., Leopold, I.H., and Bemacchi, D.B. (1985). Trans-scleral iontophoresis of gentamicin. J. Ocul. Pharmacol., 1, 363. 250 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Camber, O., Rehbinder, C., Nikkila, T. and Edman, P. (1987). Morphology of the pig cornea in normal conditions and after incubation in a perfusion apparatus. Acta. Vet. Scand. 28, 127- 134. Canadia O.A. (1972). Ouabain and sodium effects on chloride fluxes across the isolated bullfrog cornea. Am. J. Physiol. 223(5), 1053-1057. Carpenter, C.P. and Smyth, H.F. (1946). Chemical bums of the rabbit cornea. Am. J. Ophthalmol. 29, 1363-1372. Castro-Munozledo, F. (1994). Development of a spontaneous permanent cell line of rabbit comeal epithelial cells that undergoes sequential stages of differentiation in cell culture. J. Cell Sci. 107,2343-2351. Castro-Munozledo, F., Valencia-Garcia, C., and Kuri-Harcuch, W. (1997). Cultivation of rabbit comeal epithelial cells in semm-free medium. Invest. Ophthalmol. Vis. Sci. 38(11), 2234-2244. Chan, W.K., Decker, E.A., Chow, C.K. and Boissonneault, G.A. (1994). Effect of dietary camosine on plasma and tissue antioxidant concentrations and on lipid oxidation in rat skeletal muscle. Lipids. 29,461-466. Chandrasena, G., Giltay, R., Patil, S.D., Bakken, A. and Unadkat, J.D. (1997). Functional expression of human intestinal Na+-dependent and Na+-independent nucleoside transporters in Xenopus laevis oocytes. Biochem. Pharmacol. 53(12), 1909-1918. Chang, J.-E., Basu, S.K., and Lee, V.H.L. (2000). Air-interface condition promotes the formation of tight comeal epithelial cell layers for dmg transport studies. Pharm. Res. 17(6), 670-676. Chang, S.C., Bundgaard, H., Buur, A. and Lee, V.H.L. (1987). Improved comeal penetration of timolol by prodrugs as a means to reduce systemic dmg load. Investigative Ophthalmology and Visual Science, 28, 487-491. Chang, S.C., Chien, D.S., Bundgaard, H. and Lee, V.H.L. (1988). Relative effectiveness of prodrug and viscous solution approaches in maximizing the ratio of ocular to systemic absorption of topically applied timolol. Exp. Eye Res. 46, 59-69. Chang, S.-C., Bundgaard, H., Buur, A., and Lee, V. (1987). Improved comeal penetration of timolol by prodrugs as a means to reduce systemic dmg load. Invest. Ophthal. Vis. Sci., 28, 487-491. Chang, S.C. and Lee, V.H.L. (1987). Nasal and conjunctival contributions to the systemic absorption of topical timolol in the pigmented rabbit: implications in the design of strategies to maximize the ratio of ocular to systemic absorption. J. Ocular Pharmacol. 3, 159-169. 251 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chang, C.W., L. Ye, C.W., Defoe, D.M. and Caldwell, R.B. (1997). Serum inhibits tight junction formation in cultured pigment epithelial cells. Invest. Ophthal. Vis. Sci. 38, 1082- 1093. Chantret, I., Rodolosse, A., Barbat, A., Dussaulx, E., Brot-Laroche, E., Zweibaum, A. and Rousset, M. (1994). Differential expression of sucrase-isomaltase in clones isolated from early and late passages of the cell line Caco-2: evidence for glucose-dependent negative regulation. J. Cell. Sci. 107, 213-225. Chapman, S.J., Walsh, A., Beckett, E. and Vickers, C.F. (1989). A fully differentiating epidermal model with extended viability: development and partial characterization. J. Invest. Dermatol. 93(6), 762-768. Chen, F.S. and Maurice, D.M. (1990). The pH in the precorneal tear film and under a contact lens measured with a fluorescent probe. Exp. Eye. Res. 50, 251-259. Chiang, C.H. and Schoenwald, R.D. (1986). Ocular pharmacokinetic models of clonidine-3H hydrochloride. J. Pharmacokinet. Biopharm. 14(2), 175-211. Chiang, C.H., Schoenwald, R.D., and Huang, H.S. (1986). Comeal permeability of adrenergic agents potentially useful in glaucoma. J. Taiwan Pharmaceut. Assoc., 38, 67-84. Chien, S.D. and Schoenwald, R.D. (1986). Improving the ocular absorption of phenylephrine. Biopharm. Drug Disp. 7, 453-462. Chiou, W.L., Chung, S.M., Wu, T.C. (2000). Apparent lack of effect of P-glycoprotein on the gastrointestinal absorption of a substrate, tacrolimus, in normal mice. Pharm. Res. 17(2), 205-208. Chong, S., Dando, S.A., Soucek, K.M. and Morrison, R.A. (1996). In vitro permeability through Caco-2 cells is not quantitatively predictive of in vivo absorption for peptide-like drugs absorbed via the dipeptide transporter system. Pharm. Res. 13, 120-123. Chrai, S.S., and Robinson, J.R. (1974). J. Pharm. Sci. 63, 1218. Cintron, C., Hassinger, L., Kublin, C.L. and Friend, J. (1979). A simple method for the removal of rabbit comeal epithelium utilizing n-heptanol. Ophthalmic. Res. 11, 90-96. Cogbum, J.N., Donovan, M.G. and Schasteen, C.S. (1990). Correlation of Caco-2 transport with human oral bioavailability. J. Controlled. Release. 13, 314-315. Cole, S.P. and Deeley, R.G. (1998). Multidrug resistance mediated by the ATP-binding cassette transporter protein MRP. BioAssays. 20, 934-940. Colegio, O.R., Van Itallie, C.M., McCrea, H.J., Rahner, C., and Anderson, J.M. (2002). Claudins create charge-selective channels in the paracellular pathway between epithelial cells. Am. J. Physiol. Cell Physiol. 283(1), C142-7. 252 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Coles, W.H. and Jaros, P.A. (1984). Dynamics of ocular surface pH. Br. J. Ophthalmol. 68(8), 549-552. Comerford, K.M., Wallace, TJ. and Karhausen, J., Louis, N.A., Montalto, M.C. and Colgan, S.P. (2002). Hypoxia-inducible factor-1-dependent regulation of the multidrug resistance (MDR1) gene. J. Cancer Res. 62, 3387-3394. Conroy, C. and Buck, R. (1992). Influence of ion pairing salts on the transcomeal permeability of ionized sulfonamides. J. Ocular Pharmacol. 8, 233-240. Cooper, E.R. and Kasting, G. (1987). Transport across epithelial membranes. J. Controlled Release. 6, 23-35. Crawford, C.R., Cass, C.E. and Young, J.D. (1998). Stable expression of a recombinant sodium-dependent, pyrimidine-selective nucleoside transporter (CNT1) in a transport- deficient mouse leukemia cell line. Biochem. Cell Biol. 76(5), 843-851. Crosson, C.E. and Petrovich, M. (1999). Contributions of adenosine receptor activation to the ocular actions of epinephrine. Investigative Ophthalmology & Visual Science. 40, 2054- 2061. Dantzig, A.H. and Bergin, L. (1990) Uptake of the cephalosporin, cephalexin, by a dipeptide transport carrier in the human intestinal cell line, Caco-2. Biochim. Biophys. Act. 1027(3), 211-217. Dantzig, A.H., Hoskins, J., Tabas, L.B., Bright, S., Shepard, R.L., Jenkins, I.L., Duckworth, D.C., Sportsman, J.R., Mackensen, D., Rostek, P.R. and Skatrud. P.L. (1994). Association of intestinal peptide transport with a protein related to the cadherin superfamily. Science. 264, 430-433. Delie, F., Mah, S., Scherer, B.J., and Rubas, W. (1997). Temperature affects TEER and mannitol permeability across Caco-2 monolayers and rabbit jejunum. Proceed. Int’l. Symp. Control Rel. Bioact. Mater., 24 (5053), 325-326. Desai, S.D., Blanchard, J. (1994). In Encyclopedia of Pharmaceutical Technology; Swarbick, J., Boylan, J.C., Eds.; Marcel Dekker: New York, 11,43-75. Deutsch, C. (1990). K+ channels and mitogenesis. Prog. Clin. Biol. Res. 334, 251-271. Diebold, Y„ Herreras, J.M., Callejo, S., Argueso., P. and Calonge, M. (1998). Carbomer- versus cellulose-based artificial-tear formulations: Morphologic and toxicological effects on a comeal cell line. Cornea. 17, 433-440. Dikstein, S. and Maurice, D.M. (1972). The active control of comeal hydration. I sr. J. Med. Sci. 8(8), 1523-1528. Doane, M.G., Jensen, A.D., and Dohlman, C.H. (1978). Penetration routes of topically applied eye medications. Am. J. Ophthalmol. 85, 383-386. 253 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Dodoo, A.N., Bansal, S., Barlow, D.J., Bennet, F.C., Hider, R.C., Lansley, A.B., Lawrence, M.J., and Marriott, C. (2000). Systematic investigations of the influence of molecular structure on the transport of peptides across cultured alveolar cell monolayers. Pharm. Res. 17(1), 7-14. Donn, A., Maurice, D. and Mills, N. (1959). Studies in the living cornea in vitro. II. The active transport of sodium across the epithelium. Arch. Ophthalmol. 62, 748-757. Doughty, M.J. (1997). Scanning electron microscopy study of the tarsal and orbital conjunctival surfaces compared to peripheral comeal epithelium in pigmented rabbits. Documenta Ophthalmologica 93, 345-371. Dresser, M.J., Leabman, M.K. and Giacomini, K.M. (2001). Transporters involved in the elimination of drugs in the kidney: organic anion transporters and organic cation transporters. J. Pharm. Sci. 90(4), 397-421. Duffel, M.W., Ing, I.S., Segarra, T.M., Dixson, J.A., Barfnecht, C.F. and Schoenwald, R.D. (1986). N-Substituted sulfonamide carbonic anhydrase inhibitors with topical effects on intraocular pressure. J. Med. Chem., 29, 1488-1494. Edelhauser, H.F. and Maren, T.H. (1988). Permeability of human cornea and sclera to sulfonamide carbonic anhydrase inhibitors. Arch. Ophthalmol. 106, 1110-1115. Edward, A. and Prausnitz, M.R. (2001). Predicted permeability of the cornea to topical drugs. Pharm. Res. 18(11), 1497-1508. Eggli, P., Boulton, M. and Marshall, J. (1989). Growth characteristics of central and peripheral bovine comeal epithelial cells in vitro. Graefe’ s Arch. Clin. Exp. Ophthalmol. 227, 263-270. Egorova, E.V., Babizhaev, M.A., Travkin, A.G., Tolchinskaia, A.I. and Ioshin, I.E. (1989). Oxidation of pyridine nucleotides in injuries of the cornea. Vestn. Oftalmol. 105,43-46. Ehlers, N. and Ehlers, D. (1968). Effects of hydrostatic and colloid-osmotic pressure on electrical potential and short-circuit current across the explanted rabbit cornea. Acta. Ophthalmol. (Copenh). 46(4), 767-778. Elferink, R.P.J.O., Meijer, D.K.F., Kuipers, F., Jansen, P.L.M., Groen, A.K. and Groothuis, G.M.M. (1995). Hepatobiliary secretion of organic compounds; molecular mechanisms of membrane transport. Biochim. Biophys. Acta. 1241, 215-268. Eller, M.G., Schoenwald, R.D., Dixson, J.A., Segarra, T., and Barfkneckt, C.F. (1985). Topical carbonic anhydrase inhibitors III: Opimization model for comeal penetration of ethoxzolamide analogues. J. Pharm. Sci., 74, 155-160. Ellis, P.P., Wu, P.Y., and Pfoff, D.S. (1992). Effect of nasolacrimal occlusion on timolol concentrations in the aqueous humor of the human eye. J. Pharm. Sci. 81, 219-220. 254 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Esser, P., Tervooren, D., Heimann, K., Kociok, N., Bartz-Schmidt, K.U., Walter, P. and Weller, M. (1998). Intravitreal daunomycin induces multidrug resistance in proliferative vitreoretinopathy. Invest. Ophthalmol. Vis. Sci. 39, 164-170. Faulds, D., Goa, K.L., and Benfield, P. (1993). Cyclosporin A. A review of its pharmacodynamic and pharmacokinetics properties, and therapeutic use in imuunoregulatory disorders. Drugs. 45, 954-1040. Fei, Y.-J., Kanai, Y. Nussberger, S., Ganapathy, V. Leibach, F.H., Romero, M.F., Singh, S.K., Boron, W.F. and Hediger, M.A. (1994). Expression cloning of a mammalian proton- coupled oligopeptide transporter. Nature. 368. 563-566. Felipe, A., Valdes, R., Santo, B., Lloberas, J., Casado, J. and Pastor-Anglada M. (1998). Na+-dependent nucleoside transport in liver: two different isoforms from the same gene family are expressed in liver cells. Biochem. J. 330, 997-1001. File, R.R., and Patton, T.F. (1980). Topically applied pilocarpine. Human papillary response as a function of drop size. Arch. Ophthalmol. 98, 112-115. Fischer, F., Voigt, G., Liegl, O. and Wiederholt, M. (1974). Effect of pH on potential difference and short circuit current in the isolated human cornea. Pfliigers. Arch. 349, 119- 131. Fischer, F.H., Schmitz, L., Hoff, W., Schartl, S., Liegl, O. and Wiederholt, M. (1978). Sodium and chloride transport in the isolated human cornea. Pfliigers. Arch. 373, 179-188. Flores-Riveros, J.R., McLenithan, J.C., Ezaki, O. and Lane, M.D. (1993). Insulin down- regulate expression of the insulin-responsive glucose transporter (GLUT4) gene: Effects on transcription and mRNA turnover. Proc. Natl. Acad. Sci. USA 90, 512-516. Foets, B., Missotten, L., Vanderveeren, P., and Goossens, W. (1985). Prolonged survival of allogeneic corneal grafts in rabbits treated with topically applied cyclosporin A: systemic absorption and local immunosuppressive effect. Br. J. Ophthalmol. 69(8), 600-603. Fonteles, M.C., Ganapathy, V., Pashley, D.H. and Leibach, F.H. (1983). Dipeptide metabolism in the isolated perfused rat kidney. Life Sci. 33(5), 431-436. Forbes, D.J., Pozos, R.S., Nelson, J.D. and Lima, P.H. (1984). Characterization of rat corneal epithelium maintained in tissue culture. Current Eye Research. 3, 1471-1479. Friend, J., Kinoshita, S., Thoft, R.A. and Eliason, JA. (1982). Corneal epithelial cell cultures on stromal carriers. Invest. Ophthalmol. Vis. Sci. 23, 41-49. Fu, X.C. and Liang, W.Q. (2001). Prediction of comeal permeability using polar molecular surface areas. Pharmazie. 56(8), 667. Fu, X.C. and Liang, W.Q. (2002). A simple model for the prediction of comeal permeability. Int. J. Pharm. 232, 193-197. 255 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fuller, S.D. and Simons, K. (1986). Transferrin receptor polarity and recycling accuracy in “tight” and “leaky” strains of Madin-Darby canine kidney cells. J. Cell Biol. 103(5), 1767- 1779. Gan, L.S., Niederer, T., Eads, C. and Thakker, D. (1993). Evidence for predominantly paracellular transport of thyrotropin-releasing hormone across CACO-2 cell monolayers. Biochem. Biophys. Res. Commun. 197(2), 771-777. Ganapathy M.E., Brandsch, M., Prasad, P.D., Ganapathy, V., and Leibach, F.H. (1995). Differential recognition of (3-lactam antibiotics by intestinal and renal peptide transporters, PEPT1 and PEPT2. J. Biol. Chem. 270, 25672-25677. Ganapathy, M.E., Huang, W., Rajan, D.P., Carter, A.L., Sugawara, M., Iseki, K., Leibach, F.H. and Ganapathy, V. (2000). J. Biol. Chem. 275(3), 1699-1707. Geerlin, G., Daniels, J.T., Dart, J.K.G, Cree, I.A. and Khaw, P.T. (2001). Toxicity of natural tear substitutes in a fully defined culture module of human corneal epithelial cells. Investigative Ophthalmology and Visual Science. 42(5), 948-956. Gerstin, K.M., Dresser, M.J., Wang, J. and Giacomini, K.M. (2000) Molecular cloning of a Na+-dependent nucleoside transporter from rabbit intestine. Pharm. Res. 17(8), 906-910. Giasson, CJ. and Bonanno, J.A. (1994). Corneal epithelial and aqueous humor acidification during in vivo contact lens wear in rabbits. Invest Ophthalmol. Vis. Sci. 35, 851-861. Giasson, C. and Bonanno, J. (1994). Facilitated transport of lactate by rabbit comeal endothelium. Exp. Eye Res. 9, 73-81. Gilbard, J.P., Carter, J. B., Sang, D.N., Refojo, M.F., Hanninen, L.A., Kenyon, K.R. (1984). Morphologic effect of hyperosmolarity on rabbit comeal epithelium. Ophthalmology 91, 1205-1212. Gipson, I.K., and Grill, S.M. (1982). A technique for obtaining sheets of intact rabbit comeal epithelium. Invest. Ophthalmol. Vis. Sci. 23, 269-273. Gonzalez-Mariscal, L., Islas, S., Contreras, R.G., Garcia-Villegas, M.R., Betanzos, A., Vega, J., Diaz-Quinonez, A., Martin-Orozco, N., Ortiz-Navarrete, V. Cereijido., M. and Valdes, J. (1999). Molecular characterization of the tight junction protein ZO-1 in MDCK cells. Exp. Cell. Res. 248(1), 97-109. Gorodeski, G.I. and Goldfarb, J. (1997). Extracellular ATP regulates transcervical permeability by modulating two distinct paracellular pathways. Am. J. Physiol. 272(5.1), C1602-C1610. Goskonda, V.R., Hill, R.A., Khan, M.A. and Reddy, I.K. (2000). Permeability of chemical delivery systems across rabbit corneal (SIRC) cell line and isolated corneas: a comparative study. Pharmaceutical Development and Technology. 5(3), 409-416. 256 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Goskonda, V.R., Khan, M.A., Hutak, C.M. and Reddy, I.K. (1999). Permeability characteristics of novel mydriatic agents using an in victor cell culture model that utilizes SIRC rabbit comeal cells. Journal of Pharmaceutical Sciences. 88, 180-184. Grant, R.L. and Acosta, D. (1996). Interactions of intracellular pH and intracellar calcium in primary cultures of rabbit comeal epithelial cells. In Vitro Cell. Dev. Biol. Anim. 32(1), 38- 45. Grant, R.L., Yao, C., Gabaldon, D. and Acosta, D. (1992). Evaluation of surfactant cytotoxicity potential by primary cultures of ocular tissues: I. Characterization of rabbit comeal epithelial cells and initial injury and delayed toxicity studies. Toxicology. 76, 153- 176. Grass, G.M. and Robinson, J.R. (1988). Mechanisms of comeal drug penetration I: In vivo and in vitro kinetics. J. Pharm. Sci. 77, 3-14. Green, K. (1968) Relation of epithelial ion transport to comeal thickness and hydration. Nature. 217, 1074-1075. Green, K. (1965). Ion transport in isolated cornea of the rabbit. Am. J. Physiol. 209(6), 1311- 1316. Green, P.G., Hadgraft, J., and Ridout, G. (1989). Enhanced in vitro skin permeation of cationic drugs. Pharm. Res. 6(7), 628-32. Gregory, D.S. (1990). Timolol reduces IOP in normal NZW rabbits during the dark only. Invest. Ophthalmol. Vis. Sci. 4, 715-721. Gres, M.-C., Julian, B., Bourrie, M., Meunier, V., Roques, C., Berger, M., Boulenc, X., Berger, Y. and Fabre, G. (1998). Correlation between oral drug absorption in humans, and apparent drug permeability in TC-7 cells, a human epithelial intestinal cell line: comparison with the parental Caco-2 cell line. Pharmaceutical Research. 15, 726-733. Groneberg, D.A., Nickolaus, M., Springer, J., Doring, F., Daniel, H. and Fischer, A. (2001). Localization of the peptide transporter PEPT2 in the lung: implications for pulmonary oligopeptide uptake. Am. J. Pathol. 158(2), 707-714. Haaskjold, E., Refsum, H., Refsum, S.B. and Bjerknes, R. (1992). Cell kinetics of the rat comeal epithelium. APM1S. 100(12), 1123-1128. Haaskjold, E., Refsum, S.B. and Bjerknes, R. (1990). Circadian variations in the DNA synthesis of the rat comeal epithelium. A study using double labeling with tritiated thymidine. Virchows. Arch. B. Cell. Pathol. Incl. Mol. Pathol. 58(3), 229-234. Hackworth, L.A., Faraji-Shadan, F., Schuschereba, S.T., and Bowman, P.D. (1990). Serum- free culture of porcine and rabbit comeal epithelial cells. Curr. Eye Res. 9(9), 919-923. 257 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Haddad A. (2000). Renewal of the rabbit comeal epithelium as investigated by autoradiography after intravitreal injection of 3H-thymidine. Cornea. 19(3), 378-383. Hakvoort, A., Haselbach, M., Wegener, J., Hoheisel, D., and Galla, J.H. (1998). The polarity of choroid plexus epithelial cells in vitro is improved in serum-free medium. J. Neurochem. 71, 1141-1150. Hamalainen, K.M., Kananen, K., Auriola, S., Kontturi, K. and Urtti, A. (1997). Characterization of paracellular and aqueous penetration routes in cornea, conjunctiva, and sclera. Invest. Ophthalmol. Vis. Sci. 38(3), 627-634. Hamalainen, K.M., Kontturi, K., Auriola, S., Murtomaki, L., and Urtti, A. (1997). Estimation of pore size and pore density of biomembranes from permeability measurements of polyethylene glycols using an effusion-like approach. J. Controlled Release, 49, 97-104. Hamann, S. (2002). Molecular mechanisms of water transport in the eye. Int. Rev. Cytol. 215,395-431. Hamano, H., Hori, M., Hamano, T., Kawabe, H., Mikami, M., Mitsunaga, S. and Hamano, T. (1983). Effects of contact lens wear on mitosis of comeal epithelium and lactate content in aqueous humor of rabbit. Jpn. J. Ophthalmol. 27(3), 451-458. Han, Y.-H., Sweet, D.H., Hu, D.-N., and Pritchard, J.B. (2001). Characterization of a novel cationic drug transporter in human retinal pigmented epithelial cells. J. Pharmacol. Exp. Ther. 296(2), 450-457. Hara, T. and Maurice, D.M. (1972). Changes in the swelling pressure of the comeal stroma with time, hydration and temperature, determined by a new method. Exp. Eye Res. 14(1), 40- 48. Harvitt, D.M. and Bonanno, J.A. (1998). pH dependence of comeal oxygen consumption. Invest. Ophthalmol. Vis. Sci. 39, 2778-2781. Hatanaka, T., Kamon, T., Uozumi, C., Morigaki, S., Aiba, T., Katayama, K., and Koizumi, T. (1996). Influence of pH on skin permeation of amino acids. J. Pharm. Pharmacol. 48 (7), 675-9. Hay, R., Caputo, J., Chen, T.R., Macy, M., McClintock, P. and Reid, Y. (1992). American type culture collection catalogue of cell lines and hybridomas. American Type Culture Collection (7th Ed.). Hazlett, L.D., Spann, B., Wells, P. and Berk, R.S. (1980). Desquamation of the comeal epithelium in the immature mouse: a scanning and transmission microscopy study. Exp. Eye Res. 31,21-30. Hidalgo, I.J., Raub, TJ. and Borchardt, R.T. (1989). Characterization of the human colon carcinoma cell line (Caco-2) as a model system for intestinal epithelial permeability. Gastroenterology. 96(3), 736-749. 258 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hodson, S., Kaila, D., Hammond, S., Rebello, G. and al-Omari, Y. (1992). Transient chloride binding as a contributory factor to corneal stromal swelling in the ox. J. Physiol. 450, 89-103. Hodson, S., Wigham, C., Williams, L., Mayes, K.R. and Graham, M.V. (1981). Observation on the human cornea in vitro. Exp. Eye Res. 32(3), 353-360. Holash, J.A., and Stewart, P.A. (1993). The relationship of astrocyte-like cells to the vessels that contribute to the blood-ocular barriers. Brain Res. 629, 218-224. Holden, B., Sweeney, D., and Sanderson, G. (1984). The minimal precorneal oxygen tension to avoid corneal edema. Invest. Ophthalmol., 25,476. Holland, E.J. and Mannis, MJ. (2002). Ocular surface disease Tsubota, S.C.G., Tseng, K. and Nordlund, M.L. Anatomy and physiology of the ocular surface. Springer-Verlag, New York, Inc. p. 11. Horibe, Y., Hosoya, K., Kim, KJ. and Lee, V.H. (1998). Carrier-mediated transport of monocarboxylate drugs in the pigmented rabbit conjunctiva. Invest. Ophthalmol. Vis. Sci. 39(8), 1436-1443. Hosoya, K., Horibe, Y., Kim, KJ. and Lee, V.H. (1997). Na(+)-dependent L-arginine transport in the pigmented rabbit conjunctiva. Ex. Eye Res. 65(4), 547-553. Hosoya, K., Horibe, Y., Kim, KJ. and Lee, V.H. (1998). Nucleoside transport mechanisms in the pigmented rabbit conjunctiva. Invest. Ophthalmol. Vis. Sci. 39(2), 372-377. Hosoya, K.-I., Kim, K.-J., and Lee, V.H.L. (1996). Age-dependent expression of p- glycoprotein pgl70 in caco-2 cell monolayers. Pharm. Res. 13(6), 885-890. Huang, A.J., Tseng, S.C. and Kenyon, K.R. (1989). Paracellular permeability of comeal and conjunctival epithelia. Invest. Ophthalmol. Vis. Sci. 30(4), 684-689. Huang, A.J., Tseng, S.C. and Kenyon, K.R. (1990). Alteration of epithelial paracellular permeability during comeal epithelial wound healing. Invest. Ophthalmol. Vis. Sci. 31(3), 429-435. Huang, A.J., Tseng, S.C. and Kenyon, K.R. (1991). Change of paracellular permeability of ocular surface epithelium by vitamin A deficiency. Invest. Ophthalmol. Vis. Sci. 32(3), 633- 639. Huang, H.-S., Schoenwald, R.D., and Lach, J. L. (1983). Comeal penetration behavior of 3- blocking agents II: assessment of barrier contributions. J. Pharm. Sci. 72, 1272-1281. Huang, H.-S., Schoenwald, R.D., and Lach, J. L. (1983). Comeal penetration behavior of 3- blocking agents II: In Vitro-In Vivo Correlations. J. Pharm. Sci. 72(11), 1279-1281. 259 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Huff, J.W. and Reinach, P.S. (1985). Mechanism of inhibition of net ion transport across frog comeal epithelium by calcium channel antagonists. J. Membr. Biol. 85(3), 215-223. Hui, H.W., Zeleznick, L. and Robinson, J.R. (1984). Ocular disposition of topically applied histamine, cimetidine, and pyrilamine in the albino rabbit. Curr. Eye Res. 3(2), 321-230. Hutak, C.M., Jacaruso, R.B., Carson, S. and Barletta, M.A. (1991). The use of cell lysis as an index of ocular irritation potential. J. Toxicol. Cutan. Ocular Toxicol. 5(2), 143. Hutak, C.M., Kavanagh and M.E., Reddy, I.K. (2002). Comparative development of SIRC rabbit comeal cells grown on polycarbonate- and polyester-based filters. Skin Pharmacol. Appl. Skin Physiol. 15, 133-138. Hutak, C.M., Kavanagh, M.E., Reddy, I.K. and Barletta, M.A. (1997). Growth pattern of SIRC rabbit comeal cells in microwell inserts. J. Toxicol. 16(3), 145-156. Idriss, H.T., Hannun, Y.A., Boulpaep, E. and Basavappa, S. (2000). Regulation of volume- activated chloride channels by P-glycoprotein: phosphorylation has the final say! J. Physiol. 534, 629-636. Igarashi, H., Sato, Y., Hamada, S., and Kawasaki, T. (1984). Studies on rabbit comeal permeability of local anesthetics. Jpn. J. Pharmacol., 34, 429-434. Inui, K., Yamamoto, M., and Saito, H. (1992). Transepithelial transport of oral cephalosporins by monolayers of intestinal epithelial cell line Caco-2: Specific transport systems in apical and basolateral membranes. J. Pharmacol. Exp. Ther. 261, 195-201. Irie, M., Terada, T., Sawada, K., Saito, H., and Inui, K.-I. (2001). Recognition and transport characteristics of nonpeptidic compounds by basolateral peptide transporter in Caco-2 cells. J. Pharmacol. Exp. Ther. 298(2), 711-717. Iseki, K., Sugawara, M., Saitoh, N. and Miyazaki, K. (1993). The transport mechanisms of organic cations and their zwitterionic derivatives across rat intestinal brush-border membrane. II. Comparison of the membrane potential effect on the uptake by membrane vesicles. Biochim. Biophys. Acta. 1152, 9-14 Jacaruso, R.B., Barletta, M.A., Carson, S. and Hardig, W. (1985). An in vitro method for assessing comeal opacification potential using a rabbit comeal cell line. J. Toxicol. Cutan. Ocular Toxicol. 4(1), 49. Jankowska, L., Bar-Ilan, A. and Maren, T. (1986). The relations between ionic and nonionic diffusion of sulfonamides across the rabbit cornea. Invest. Ophthal. Vis. Sci. 27, 29-37. Jelamskii, S., Sun, X.C., Herse, P., and Bonanno, J.A. (2000). Basolateral Na+ -K+ -2Cf cotransport in cultured and fresh bovine comeal endothelium. Invest. Ophthalmol. Vis. Sci. 41,488-495. 260 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Johnson, L.G., Dickman, K.G., Moore, K.L., Mandel, L.J., and Boucher, R.C. (1993). Enhanced Na+ transport in an air-liquid interface culture system. Am. J. Physiol 264, 560- 565. Johansen, S., Rask-Pedersen, E., and Prause, J.U. (1996). A bioavailability comparison in rabbits after a single topical ocular application of prednisolone acetate formulated as a high- viscosity gel and as an aqueous suspension. Arch. Ophthalmol. Scand. 74, 253-258. Joshi, A.J. (1994). Microparticulates for ophthalmic drug delivery. J. Ocul. Pharmacol. 10, 29-45. Jumblatt, M.M. and Neufeld, A.H. (1983). P-adrenergic and serotonergic responsiveness of rabbit corneal epithelial cells in culture. Invest. Ophthalmol. Vis. Sci. 24, 1139-1143. Kajikawa, T., Mishima, H.K., Murakami, T., and Takano, M. (1999). Role of P-glycoprotein in distribution of rhodamine 123 into aqueous humor in rabbits. Curr. Eye Res. 18(3), 240- 246. Karlsson, J., Kuo, S.-M., Ziemniak, J., and Artursson, P. (1993). Transport of celiprolol across human intestinal epithelial (Caco-2) cells: mediation of secretion of multiple transporters including p-glycoprotein. Br. J. Pharmacol. 110, 1009-1016. Karlsson, J., Ungell, A., Grasjo, J., and Artursson, P. (1999). Paracellular drug transport across intestinal epithelia: influence of charge and induced water flux. Eur. J. Pharm. Sci. 9(1), 47-56. Kaswan, R.L. (1988). Intraocular penetration of topically applied cyclosporine. Transpl. Proc. 20(suppl. 2), 650-655. Kawazu, K., Shiono, H., Tanioka, H., Ota, A., Ikuse, T., Takashina, H., and Kawashima, Y. (1998). Beta adrenergic antagonist permeation across cultured rabbit corneal epithelial cells grown on permeable supports. Curr. Eye Res. 17, 125-131. Kawazu, K., Yamada, K., Nakamura, M., and Ota, A. (1999). Characterization of cyclosporin A transport in cultured rabbit epithelial cells: P-glycoprotein transport activity and binding to cyclophilin. Invest. Ophthalmol. Vis. Sci. 40, 1738-1744. Kennedy, A.H., Golden, G.M., Gay, C.L., Guy, R.H., Francoeur, M.L., and Mak, V.H. (1996). Stratum comeum lipids of human epidermal keratinocyte air-liquid cultures: implications for barrier function. Pharm. Res. 13, 1162-1167. Kennedy, B.G. and Mangini, N.J. (2002). P-glycoprotein expression in human retinal pigment epithelium. Mol. Vis. 8, 422-430. Khodadoust, A.A. (1967). Techniques for the preparation of sheets of pure corneal epithelium. Am. J. Ophthalmol. 63, 942-945. 261 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Kishida, K., and Otori, T. (1980). A quantitative study on the relationship between transcorneal permeability of drugs and their hydrophobicity. Jpn. J. Ophthalmol. 24, 251- 259. Kleinberg, J., Dea, F.J., Anderson, J.A. and Leopold, I.H. (1979). Intraocular penetration of topically applied lincomycin hydrochloride in rabbits. Arch. Ophthalmol. 97(5), 933-936. Klyce, S.D. (1973). Relationship of epithelial membrane potentials to corneal potential. Exp. Eye Res. 15, 567-575. Klyce, S.D. (1972). Electrical profiles in the corneal epithelium. J. Physiol. 226, 407-429. Klyce, S.D. (1981). Stromal lactate accumulation can account for corneal oedema osmotically following epithelial hypoxia in the rabbit. J. Physiol. 321, 49-64. Klyce S.D., Neufeld, A.H. and Zadunaisky, J.A. (1973). The activation of chloride transport by epinephrine and Deb cyclic-AMP in the cornea of the rabbit. Invest. Ophthalmol. 12(2), 127-139. Klyce, S.D., and Wong, R.K.S. (1977). Site and mode of adrenaline action on chloride transport across the rabbit comeal epithelium. J. Physiol. 266,777-799. Kim, S.W., Yeum, C.H., Kim, S., Oh, Y., Choi, K.C. and Lee, J. (2001). Amphotericin B decreases adenylyl cyclase activity and aquaporin-2 expression in rat kidney. J. Lab. Clin. Med. 138, 243-249. Knipp, G.T., Ho, N.F.H., Barsuhn, C.L., and Borchardt, R.T. (1997). Paracellular diffusion in Caco-2 cell monolayers: Effect of perturbation on the transport of hydrophilic compounds that vary in charge and size. J. Pharm. Sci. 86(10), 1105-1110. Kobayashi, S., Zimmermann, H. and Millhom, D.E. (2000) Chronic hypoxia enhances adenosine release in rat PC 12 cells by altering adenosine metabolism and membrane transport. J. Neurochem. 74, 621-632. Koepsell, H. (1988). Organic cation transporters in intestine, kidney, liver, and brain. Annu. Rev. Physio. 60, 243-266. Koepsell, H., Gorboulev, V. and Amdt, P. (1999). Molecular pharmacology of organic cation transporters in kidney. J. Membr. Biol. 167(2), 103-117. Kohen, R., Yamamoto, Y., Cundy, K.C. and Ames, B.N. (1988). Antioxidant activity of camosine, homocamosine, and anserine present in muscle and brain. Proc. Natl. Acad. Sci. USA. 85,3175-3179. Komatsu, A., Ohashi, K., Oba, H., Kakehashi, T., Mizushima, Y., Shirasawa, E., and Horiuchi, M. (1988). Application of lipid microsphere drag delivery system to steroidal ophthalmic preparation. Jpn. J. Ophthalmol. 32,41-43. 262 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Kondo, M., Finkbeiner, W.E. and Widdicombe, J.H. (1991). A simple technique for culture of highly differentiated cells from dog tracheal epithelium. Am. J. Physiol. 261, L106-L117. Kondo, M., Tamaoki, J., Sakai, A., Kameyama, S., Kanoh, S. and Konno, K. (1997). Increased oxidative metabolism in cow tracheal epithelial cells cultured at air-liquid interface. Am. J. Respir. Cell Mol. Biol. 16, 62-68. Kovbasnjuk, O.N., Szmulowicz, U. and Spring, K.R. (1998). Regulation of the MDCK cell tight junction. J. Membr. Biol. 161(1), 96-104. Kruse, F.E. and Tseng, S.C.G. (1991). A serum-free clonal growth assay for limbal, peripheral, and central corneal epithelium. Invest. Ophthalmol. Vis. Sci. 32, 2086-2095. Kurpakus, M.A., Stock, E.L., Jones, J.C.R. (1990). Expression of the 55/64 kDa keratins in ocular surface epithelium. Invest. Ophthalmol. Vis. Sci. 31,448-456. Kwon, Y., Kamath, A.V., and Morris, M.E. (1996). Inhibitors of P-glycoprotein-mediated daunomycin transport in rat liver canalicular membrane vesicles. J. Pharm. Sci. 85, 935-939. Kyyronen, K., and Urtti, A. (1990). Improved ocular:systemic absorption ration of timolol by viscous vehicle and phenylephrine. Invest. Ophthalmol. Vis. Sci. 31, 1827-1833. Lai, Y.K. (1989). Adverse effect of intraoperative phenylephrine 10%: case report. Br. J. Ophthalmol. 73(6), 468-469. Lauweryns, B., Van Den Oord, J.J. and Missotten, L. (1993). The transitional zone between limbus and peripheral cornea. An immunohistochemical study. Invest. Ophthalmol. Vis. Sci. 34(6), 1991-1999. LeBourlais, C., Acar, L., Zia, H., Sado, P.A., Needham, T., and Leverge, R. (1998). Ophthalmic drug delivery systems-recent advances. Progr. Retin. Eye Res. 17, 33-58. Lee, J., Yoo, K.S., Kang, Kim, S.W. and Choi, K.C. (2001). Gentamicin decreases the abundance of aquaporin water channels in rat kidney. Jpn. J. Pharmacol. 85, 391-398. Lee, V.H.L. (1992). Minimizing the systemic absorption of topically applied ophthalmic drugs in rabbits. S.T.P. Pharma. Sciences. 2(1), 5-12. Lee, V.H.L. (1993). In Ophthalmic Drug Delivery Systems; Mitra, A.K., Ed.; Marcel Dekker: New York, 59-82. Lee, V.H.L. (1993). In Biopharmaceutics of Ocular Drug Delivery; Edman, P., Ed.; CRC press: Boca Raton, FL, 121-143. Lee, V.H.L., Chien, D.-S., and Sasaki, H. (1988). Ocular ketone reductase distribution and its role in the metabolism of ocularly applied levobunolol in the pigmented rabbit. J. Pharmacol. Exp Ther. 246(3), 871-878. 263 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Lee, V.H.L., Hui, H.W., and Robinson, J.R. (1980). Comeal metabolism of pilocarpine in pigmented rabbits. Invest. Ophthalmol. Vis. Sci., 9, 210. Lee, V.H., and Robinson, J.R. (1986). Topical ocular drug delivery: recent developments and future challenges. J. Ocul. Pharmacol. 2, 67-108. Lee, V.H.L., Luo, A.M., Li, S., Podder, S.K., Chang, S.C., Ohdo, S., and Grass, G.M. (1991). Pharmacokinetic basis for nonadditivity of IOP lowering in timolol combinations. Invest. Ophthalmol. Vis. Sci., 32, 2948. Leier, I., Jedlitschky, G., Buchholz, U., Cole, S.P., Deeley, R.G. and Keppler, D. (1994). The MRP gene encodes an ATP-dependent export pump for leukotriene C4 and structurally related conjugates. J. Biol. Chem. 269, 27807-27810. Leibowitz, H.M., Berrospi, A.R., Kupferman, A., Restropo, G.V., Galvis, V. and Alvarez, J.A. (1977). Penetration of topically administered prednisolone acetate into the human aqueous humor. Am. J. Ophthalmol. 83(3), 402-406. Lennermas, H., Palm, K., Fagerholm, U. and Artursson., P. (1996). Comparison between active and passive drug transport in human intestinal epithelial (Caco-2) cells in vitro and human jejunum in vivo. Int. J. Pharm. 127, 103-107. Leung, G.P., Cheung, K.H., Tse, C.M. and Wong, P.Y. (2001). Biol. Reprod. Na+ reabsorption in cultured rat epididymal epithelium via the Na+/nucleoside cotransporter. Biol. Reprod. 64(3), ISA-169. Liang, R., Fei, Y.J., Prasad, P.D., Ramamoorthy, S., Han, H., Yang-Feng, T.L., Hediger, M.A., Ganapathy, V. and Leibach, F.H. (1995). Human intestinal HVpeptide cotransporter. Cloning, functional expression, chromosomal localization. J. Biol. Chem. 270. 6456-6463. Liaw, J., and Robinson, J.R. (1992). The effect of polyethylene glycol molecular weight on comeal transport and the related influence of penetration enhancers. Int. J. Pharm. 88 (1-3), 125-140. Liaw, J., Rojanasakul, Y. and Robinson, J.R. (1991). The effect of charge type and charge density on comeal transport of lysine and glutamic acid-Evidence for active transport. Pharm. Res. 8(Suppl), S-130. Liminga, M. and Oliw, E.H. (2000). Studies of lipoxygenases in the epithelium of cultured bovine cornea using an air-interface model. Exp. Eye Res. 71(1), 57-67. Lindberg, K., Brown, M.E., Chaves, H.V., Kenyon, K.R. and Rheinwald, J.G. (1993). In vitro propagation of human ocular surface epithelial cells for transportation. Investigative Ophthalmology & Visual Science. 34, 2672-2679. Liu, W., Liang, R., Ramamoorthy, S., Fei, Y.J., Ganapathy, V., Hediger, M.A., Ganapathy, M.E. and Leibach, F.H. (1995). Molecular cloning of PepT2, a new member of the 264 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. H+ /peptide cotransporter family, from human kidney. Biochim. Biophys. Acta. 1235,461- 466. Lowry, M., Hall, D.E. and Brosnan, J.T. (1985). Increased activity of renal glycine- cleavage-enzyme complex in metabolic acidosis. Biochem. J. 231(2), 477-480. Lowry, M., Hall, D.E. and Brosnan, J.T. (1985). Hydroxyproline metabolism by the rat kidney: distribution of renal enzymes of hydroxyproline catabolism and renal conversion of hydroxyproline to glycine and serine. Metabolism. 34(10), 955-961. Lowry, M., Hall, D.E. and Brosnan, J.T. (1985). Metabolism of glycine- and hydroxyproline-containing peptides by the isolated perfused rat kidney. Biochem. J. 229(2), 545-549. Maichuk, I.F., Formaziuk, V.E. and Sergienko, V.I. (1997). Development of camosine eyedrops and assessing their efficacy in comeal diseases. Vestn. Oftalmol. 113, 27-31. Mairbaurl, H., Wodopia, R., Eches, S., Schulz, S. and Bartsch, P. (1997). Impairment of cation transport in A549 cells and rat alveolar epithelial cells by hypoxia. Am. J. Physiol. 273 (17), L797-L806. Mandel, L.J., Doctor, R.B. and Bacallao, R. (1994). ATP depletion: a novel method to study junctional properties in epithelial tissues. II. Internalization of Na+, K(+)-ATPase and E- cadherin. J. Cell. Sci. (.12), 3315-3324. Marchis, S.D., Melcangi, R.C., Modena, C., Cavaretta, I., Peretto, P., Agresti, C. and Fasolo, A. (1997). Identification of the glial cell types containing camosine-related peptides in the rat brain. Neuroscience Lett. 237, 37-40. Maren, T.H., Jankowska, L., Sanyal, G. and Edelhauser, H.F. (1983). The transcomeal permeability of sulfanomide carbonic anhydrase inhibitors and their effect on aqueous humor secretion. Exp. Eye Res., 36,457-480. Margolis, F.L. and Grillo, M. (1984). Camosine homocamosine and anserine in vertebrate retinas. Neurochem. Int. 6, 207-209. Margolis, F.L. The role of peptides in neuronal function. J.L. Barker: T.G. Smith (eds). (1980). Camosine: an olfactory neuropeptide, Dekker. 545-572. Marshall, W.S. and Klyce, S.D. (1983). Cellular and paracellular pathway resistances in the “tight” Cl- -secreting epithelium of rabbit cornea. J. Membr. Biol. 73(3), 275-282. Mathias, N.R. (1997). An air-interface rabbit tracheal epithelial cell culture model for the characterization of nucleoside and nucleoside drug transport. Thesis. Mathias, N.R., Kim, K.-J. and Lee, V.H.L. (1996). Targeted drug delivery to the respiratory tract: solute permeability of air-interface cultured rabbit tracheal epithelial cell monolayers. J. Drug Targeting. 4, 79-86. 265 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mathias, N.R., Robison, T.W., Kim, KJ., and Lee, V.H.L. (1995). Development and characterization of rabbit tracheal epithelial cell culture models for drug transport studies. Pharm. Res. 12, 1499-1505. Matsumoto, S., Saito, H., and Inui, K. (1994). Transcellular transport of oral cephalosporins in human intestinal epithelial cells, Caco-2: Interaction with dipeptide transport systems in apical and basolateral membranes. J. Pharmacol. Exp. Ther. 270, 498-504. Matsuo, T. (1998). Expression of amiloride-sensitive sodium channel in rat eye. Acta. Med. Okayama. 52(5), 279-283. Matsuzaki, J., Yamamoto, C., Miyama, T., Takanaga, H., Matsuo, H., Ishizuka, H., Kawahara, Y., Kuwano, M., Naito, M., Tsuruo, T., and Sawada, Y. (1999). Contribution of P-glycoprotein to bunitrolol efflux across blood-brain barrier. Biopharm. Drug Dispos. 20(2), 85-90. Maurice, D.M. (1967). Epithelial potential of the cornea. Exp. Eye Res. 6(2), 138-140. Maurice, D.M. (1967). The use of fluorescein in ophthalmological research. Invest. Ophthalmol., 6, 464. Maurice, D.M. (1997). Characterization of paracellular penetration routes. Invest. Ophthalmol. Vis. Sci. 38(11), 2177-2180. Maurice, D.M., and Mishima, S. (1984). Ocular Pharmacokinetics. In: Sears M.L., ed. Handbook of Experimental Pharmacology. Berlin: Springer-Verlag. 16-119. Mauger, T.F. and Hill, R.M. (1992). Comeal epithelial healing under contact lenses. Quantitative analysis in the rabbit. Acta Ophthalmol. (Copenh). 70(3), 361-365. McNamara, M., Clynes, M., Dunne, B., NicAmhlaoibh, R., Lee, W.R., Bames, C. and Kennedy, S.M. (1996). Br. J. Ophthalmol. 80(11), 1009-1012. McPherson, S.D., Draheim, J.W.J., Evans, V.J., and Earle, W.R. (1956). The viability of fresh and frozen corneas as determined in tissue culture. Am. J. Ophthalmol. 41, 513-521. Meijer, D.K.F., Smit, J.W. and Muller, M. (1997). Hepatobiliary elimination of cationic drugs: the role of P-glycoproteins and other ATP-dependent transporters. Adv. Drug Delivery Rev. 25, 159-200. Meller, D. and Tseng, S.C. (1999). Conjunctival epithelial cell differentiation on amniotic membrane. Invest. Ophthalmol. Vis. Sci. 40(5), 878-886. Mercader, J., Gomez-Angelats, M., del Santo, B., Casado, F.J., Felipe, A. and Pastor- Anglada, M. (1996). Nucleoside uptake in rat liver parenchymal cells. Biochem. J. 317, 835- 842. 266 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Meredith, D. and Boyd, C.A.R. (1995). Dipeptide transport characteristics of the apical membrane of rat lung type II pneumocytes. Am. J. Physiol. 269, L137-L143. Merriman-Smith, B.R., Young, M.A., Jacobs, M.D., Kistler, J. and Donaldson, PJ. (2002). Molecular identification of P-glycoprotein: a role in lens circulation? Invest. Ophthalmol. Vis. Sci. 43(9), 3008-3015. Mesonero, J., Mahraoui, L., Matosin, M., Rodolosse, A., Rousset, M. and Brot-Laroche, E. (1994). Expression of the hexose transporters GLUT1-GLUT5 and SGLT1 in clones of Caco-2 cells. Biochem. Soc. Trans. 22(3), 681-684. Midelfart, A. (1987). The effect of amiloride on Na, K and water in bovine corneal epithelium. Exp. Eye Res. 45(6), 751-762. Mikkelson, T.J., Chrai, S.S. and Robinson, J.R. (1973). Altered bioavailability of drugs in the eye due to drug-protein interaction. J. Pharm. Sci. 62(10), 1648-1653. Mirshahi, M., Mirshahi, S., Golestaneh, N., Nicolas, C., Mishal, Z., Lounes, K.C., Hecquet, C., Dagonet, F., Pouliquen, Y. and Agarwal, M.K. (2001). Mineralocorticoid hormone signaling regulates the ‘epithelial sodium channel’ in fibroblasts from human cornea. Ophthalmic Res. 33(1), 7-19. Mirshahi, M., Nicolas, C., Mirshahi, S., Golestaneh, N., d’Hermies F. and Agarwal, M.K. (1999). Immunochemical analysis of the sodium channel in rodent and human eye. Exp. Eye Res. 69(1), 21-23. Mishima, S., Gasset, A., Klyce, S.D., and Baum, J.L. (1966). Determination of tear volume and tear flow. Invest. Ophthalmol., 5, 264. Mishima, S. and Hedbys. B.O. (1967). The permeability of the comeal epithelium and endothelium to water. Exp. Eye Res. 6(1), 10-32. Morimoto, K., Nakai, T., and Morisaka, K. (1987). Evaluation of permeability enhancement of hydrophilic compounds and macromolecular compounds by bile salts through rabbit corneas in-vitro. J. Pharm. Pharmacol. 39, 124-126. Morimoto, K., Yamahara, H., Lee, V.H. and Kim, KJ. (1994). Transport of thyrotropin- releasing hormone across rat alveolar epithelial cell monolayers. Life Sci. 54(26), 2083-2092. Mortell, K.H., Marmorstein, A.D., and Cramer, E.B. (1993). Fetal bovine serum and other sera used in tissue culture increase epithelial permeability. In Vitro Cell. Dev. Biol. 29A, 235-238. Muchtar, S. Abdulrazik, M. Frucht-Pery, J. and Benita, S. (1997). Ex-vivo permeation study of indomethacin from a submicron emulsion through albino rabbit cornea. J. Controlled Release, 44, 55-64. 267 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mueckler, M. (1990). Family of glucose transporter genes: Implications for glucose homeostasis and diabetes. Diabetes. 39, 6-11. Musson, D.G., Bidgood, A.M., and Olejnik, O. (1992). An in vitro comparison of the permeability of prednisolone, prednisolone sodium phosphate, and prednisolone acetate across the NZW rabbit cornea. J. Ocular Pharmacol., 8, 139-149. Nakanishi, T., Hatanaka, T., Huang, W., Prasad, P.D., Leibach, F.H., Ganapathy, M.E., and Ganapathy, V. (2001). Na+ -and Cl'-coupled active transport of carnitine by the amino acid transporter ATB0 + from mouse colon expressed in HRPE cells and Xenopus oocytes. J. Physiol. 532(2), 297-304. Narawane, M., Podder, S.K., Bundgaard, H., and Lee, V.H.L. (1992). Segmental differences in drug permeability, esterase activity and ketone reductase activity in the albino rabbit intestine. J. Drug Targeting 1, 7-17. Nelson, W.L., fraunfelder, F.T., Sills, J.M., Arrowsmith, J.B., and Kruitdky, J.N. (1986). Adverse respiratory and cardiovascular events attributed to timolol ophthalmic solution. Am. J. Ophthalmol., 102, 606. Nguyen, T.D., Koh, D.S., Moddy, M.W., Fox, N.R., Savard, C.E., Kuver. R., Hille, B. and Lee, S.P. (1997). Characterization of two distinct chloride channels in cultured dog pancreatic duct epithelial cells. Am. J. Physiol. 272(1.1), G172-180. Niederkom, J.Y., Meyer, D. R., Ubelaker, J. E., and Martin, J. H. (1990). Ultrastructural and immunohistological characterization of the SIRC corneal cell line. In Vitro Cell. Dev. Biol. 26,923-930. Nielsen, H.M., and Rassing, M.R. (2000). TR146 cells grown on filters as a model of human buccal epithelium: IV. Permeability of water, mannitol, testosterone and P-adrenoceptor antagonists. Comparison to human, monkey and porcine buccal mucosa. Int. J. Pharm., 194, 155-167. North-Root, H.M., Yackovich, F., Demetrulias, J., Gacula, M.(Jr.) and Heinze, J.E. (1982). Evaluation of an in vitro cell toxicity test using rabbit comeal cells to predict the eye irritation potential of surfactants. Toxicol. Lett. 14, 207. O’Brien, W.J. and Edelhauswer, H.F. (1977). The comeal penetration of trifluorothymidine, adenine arabinoside, and idoxuridine: a comparative study. Invest. Ophthalmol. Vis. Sci. 16(12), 1093-1103. Oh, C., Saville, B.A., Cheng, Y.L., et al. (1995). A compartmental model for the ocular pharmacokinetics of cyclosporine in rabbits. Pharm. Res. 12, 433-437. Ohdo, S., Grass, G.M. and Lee, V.H.L. (1991). Improving the ocular/systemic ratio of topical timolol by varying the dosing time. Invest. Ophthalmol. Vis. Sci. 32, 2790-2798. 268 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ouiddir, A., Planes, C., Fernandes, I., VanHesse, A. and Clerici, C. (1999). Hypoxia upregulates activity and expression of the glucose transporter GLUT1 in alveolar epithelial cells. Am. J. Respir. Cell Mol. Biol. 21, 710-718. Panzanelli, P., Cantino, D. and Sassoe-Pognetto, M. (1997). Co-localization of camosine and glutamate in photoreceptors and bipolar cells of the grog retina. Brain Res. 758, 143-152. Pastor-Anglada, M., Casada, F.J., Valdes, R., Mata, J., Garcia-Manteiga, J., and Molina, M. (2001). Complex regulation of nucleoside transporter expression in epithelial and immune system cells. Mol. Membr. Biol. 18(1), 81-85. Paterson, A.R., Kolassa, N. and Cass, C.E. (1981). Transport of nucleoside drugs in animal cells. Pharmacol. Ther. 12(3), 515-536. Patton, T.F. and Robinson, J.R. (1976). Quantitative precomeal disposition of topically applied pilocarpine nitrate in rabbit eyes. J. Pharm. Sci. 65(9), 1295-1301. Patton, T.F. and Robinson, J.R. (1975). Ocular evaluation of polyvinyl alcohol vehicle in rabbits. J. Pharm. Sci. 64(8), 1312-1316. Paul, S., Breuninger, L.M., Tew, K.D., Shen, H. and Kruh, G.D. (1996). ATP-dependent uptake of natural product cytotoxic drugs by membrane vesicles establishes MRP as a broad specificity transporter. Proc. Natl. Acad. Sci. USA. 93, 6929-6934. Pessotto, P., Valeri, P. and Arrigoni-Martelli, E. (1994). The presence of L-camitine in ocular tissues of the rabbit. J. Ocular Pharm. 10(4), 643-651. Philp, N., Chu, P., Pan, T., Zhang, R.Z., Chu, M., Stark, K., Boettiger, D., Yoon, H. and Kieber-Emmons, T. (1995) Developmental expression and molecular cloning of REMP, a novel retinal epithelial membrane protein. Exp. Cell Res. 219, 64-73. Pintor, J., Peral, A., Hoyle, C.H.V., Redick, C., Douglass, J., Sims, I. and Yerxa, B. (2002). Effects of diadenosine polyphosphates on tear secretion in New Zealand white rabbits. The Journal of Pharmacology and Experimental Therapeutics. 300,291-297. Pintor, J., Peral, A., Pelaez, T., Martin, S. and Hoyle C.H.V. (2003). Presence of diadenosine polyphosphates in the aqueous humor: their effect on intraocular pressure. The Journal of Pharmacology and Experimental Therapeutics. 304, 342-348. Plagemann, P.G. and Aran, J.M. (1990). Characterization of Na(+)-dependent, active nucleoside transport in rat and mouse peritoneal macrophages, a mouse macrophage cell line and normal rat kidney cells. Biochim. Biophys. Act. 1028(3), 289-298. Prasad, P.D., Leibach, F.H., Mahesh, V.B. and Ganapathy, V. (1992). Specific interaction of 5-(N-methyl-N-isobutyl) amiloride with the organic cation-proton antiporter in human placental brush-border membrane vesicles. Transport and binding. J. Biol Chem. 267(33) 23632-23639. 269 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Prausnitz, M.R., and Noonan, J.S. (1998). Permeability of cornea, sclera, and conjunctiva: a literature analysis for drug delivery to the eye. J. Pharm. Sci. 87, 1479-1488. Preisser, L., Teillet, L., Aliotti, S., Gobin, R., Berthonaud, V., Chevalier, J., Corman, B. and Verbavatz, J.-M. (2000). Downregulation of aquaporin-2 and -3 in aging kidney is independent of V2 vasopressin receptor. Am. J. Physiol. Renal. Physiol. 279, F144-F152. Pritchard, J.B. and Miller, D.S. (1993). Mechanisms mediating renal secretion of organic anions and cations. Physiol. Rev. 73(4), 765-796. Pritchard, J.B. and Tang-Liu, D.D. (1990). A corneal perfusion device for estimating ocular bioavailability in vitro. J. Pharm. Sci. 79(2), 153-157. Raber, I. And Breslin, C.W. (1978). Toleration of artificial tears - the effect of pH. Can. J. Ophthalmol. 13(4), 247-249. Rae, J.L., Dewey, J., Rae, J.S., Nesler, M. and Cooper, K. (1990). Single potassium channels in comeal epithelium. Invest. Ophthalmol. Vis.Sci. 31(9), 1799-1809. Rebouche, CJ. and Seim, H. (1998). Carnitine metabolism and its regulation in microorganisms and mammals. Annu. Rev. Nutr. 18, 39-61. Reinach, P. and Nagel, W. (1985). Implications of an anomalous intracellular electrical response in bull frog comeal epithelium. J. Membr. Biol. 87(3), 201-209. Reinach, P.S., Tarvin, J.T. and Hirsch, M. (1991). Changes in cellular membrane and paracellular conductances by amphotericin B in the epithelium of the bullfrog cornea. Biochim. Biophys. Acta. 1066(2), 115-123. Reinsprecht, M., Rohn, M.H., Spadinger, R.J., Pecht, I., Schindler, H. and Romanin, C. (1995). Blockade of capacitive Ca2+ influx by Cl- channel blockers inhibits secretion from rat mucosal-type mast cells. Mol. Pharmacol. 47(5), 1014-1020. Rich, A. and Rae, J.L. (1995). Calcium entry in rabbit comeal epithelial cells: evidence for a nonvoltage dependent pathway. J. Membrane Biol. 144, 177-184. Richard, N.R., Anderson, J.A., Weiss, J.L. and Binder, P.S. (1991). Air/liquid comeal organ culture: a light microscopic study. Curr. Eye Res. 10(8), 739-749. Riley, M.V. (1969). Aerobic glycolysis in the ox cornea. Exp. Eye Res. 8(2), 201-204. Robison, T.W., and Kim, K. J. (1994). Air-interface cultures of guinea pig airway epithelial cells: effects of active sodium and chloride transport inhibitors on bioelectric properties. Exp. Lung Res. 20, 101-117. Roch-Ramel, F., Besseghir, K. and Murer, H. (1992). Renal excretion and tubular transport of organic anions and cations. Handbook of Physiology (a Critical, Comprehensive Presentation of Physiological Knowledge and Concepts). 48, 2189-2262. 270 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rodrigues, M., Ben-Zvi, A., Krachmer, J., Schermer, A., and Sun, T.T. (1987). Suprabasal expression of a 64-kilodalton keratin (no. 3) in developing human comeal epithelium. Differentiation. 34, 60-67. Rojanasakul, Y., Wang, L.Y., Bhat, M., Glover, D.D., Malanga, C.J., and Ma, J.K. (1992). The transport barrier of epithelia: a comparative study on membrane permeability and charge selectivity in the rabbit. Pharm. Res. 9(8), 1029-34. Rong, B.L., Dunkel, E.C., and Pavan-Langston, D. (1988). Molecular analysis of acyclovir- induced suppression of HSV-1 replication in rabbit comeal cells. Invest. Ophthal. Vis. Sci. 29, 928-932. Rubas, W., Cromwell, M.E.M., Shahrokh, Z., Villagran, J., Nguyen, T.-N., Wellton, M., Nguyen, T.-H. and Mrsny, RJ. (1996). Flux measurements across Caco-2 monolayers may predict transport in human large intestinal tissue. Journal of Pharmaceutical Sciences. 85, 165-169. Rubas, W., Jezyk, N. and Grass, G.M. (1993). Comparison of the permeability characteristics of a human colonic epithelial (Caco-2) cell line to colon of rabbit, monkey, and dog intestine and human drug absorption. Pharm. Res. 10, 113-118. Ruifrok, P.G. (1981). Uptake of quaternary ammonium compounds into rat intestinal brush border membrane vesicles. Biochem. Pharmacol. 30, 2637-2641. Saettone, M.F., Chetoni, P., Torracca, M.T., Burgalassi, S., and Giannaccini, B. (1989). Evaluation of muco-adhesive properties and in vivo activity of ophthalmic vehicles based on hyaluronic acid. Int. J. Pharm. 51, 203-212. Saettone, M.F., Giannaccini, B., Savigni, P. and Wirth, A. (1980). The effect of different ophthalmic vehicles on the activity of tropicamide in man. J. Pharm. Pharmacol. 32(7), 519- 521. Saettone, F., Giannaccini, B., Teneggi, A., Savigni, P., and Tellini, N. (1982). Vehicle effects on ophthalmic bioavailability: the influence of different polymers on the activity of pilocarpine in rabbit and man. J. Pharm. Pharmacol. 34, 464-466. Saha, P., Kim, K.-J., and Lee, V.H.L. (1996). A primary culture model of rabbit conjunctival epithelial cells exhibiting tight barrier properties. Curr. Eye Res. 15, 1163-1169. Saha, P., Kim, K.-J., Yamahara, H., Crandall, E.D. and Lee, V.H. (1994). Influence of lipophilicity on P-blocker permeation across rat alveolar epithelial cell monolayers. J. Contr. Rel. 32, 191-200. Saha, P., Uchiyama, T., Kim, K.-J. and Lee, V.H.L. (1996). Permeability characteristics of primary cultured rabbit conjunctival epithelial cells to low molecular weight drugs. Curr. Eye Res. 15, 1170-1174. 271 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Saha, P., Yang, J J., and Lee, V.H.L. (1998). Existence of a p-glycoprotein drug efflux pump in cultured rabbit conjunctival epithelial cells. Invest. Ophthalmol. Vis. Sci. 39, 1221-1226. Saito, H., and Inui, K. (1993). Dipeptide transporters in apical and basolateral membranes of the human intestinal cell line Caco-2. Am. J. Physiol. Gastrointest. Liver Physiol. 265,G289- G294. Salminen, L., and Huupponen, R. (1989). Systemic effects of ocular drugs. Adv. Drug React. Acute Poison Rev. 8, 89-96. Shah, D. and Shen, W.C. (1994). The establishment of polarity and enhanced transcytosis of transferring receptors in enterocyte-like Caco-2 cells. J. Drug Target. 2(2), 93-99. Sasaki, H., Igarashi, Y., Nagano, T., Nishida, K., and Nakamura, J. (1995). Different effects of absorption promoters on corneal and conjunctival penetration of ophthalmic beta- blockers. J. Pharm. Res. 12, 1146-1150. Schermer, A., Galvin, S., Sun, T-T. (1986). Differentiation-related expression of a major 64K corneal keratin in vivo and in culture suggests limbal location of corneal epithelial stem cells. J. Cell Biol. 103, 49-62. Schlingemann, R.O., Hofman, P., Klooster, J., Blaauwgeers, H.G., Van der, G.R. and Vrensen, G.F. (1998). Ciliary muscle capillaries have blood-tissue barrier characteristics. Exp. Eye Res. 66, 747-754. Schoenwald, R.D. (1985). Controlled Drug Bioavailability; Smolen, Ball, Eds.; Wiley & Sons: New York, pp 257-306. Schoenwald, R.D. (1990). Ocular drug delivery: Pharmacokinetic considerations. Clin. Pharmacokinet., 18, 255-269. Schoenwald, R.D., and Houseman, J.A. (1982). Disposition of cyclosphosphamide in the rabbit and human cornea. Biopharmaceut. Drug Disposition. 3, 231-241. Schoenwald, R.D., and Huang, H.-S. (1983). Corneal penetration behavior of P-blocking agents I: Physicochemical factors. J. Pharm. Sci. 72(11), 1266-1272. Schoenwald, R.D. and Ward, R.L. (1978). Relationship between steroid permeability across excised rabbit cornea and octanol-water partition coefficients. J. Pharm. Sci. 67, 786-788. Shen, J., Elbert, KJ., Yamashita, F., Lehr, C.-M., Kim, K.-J., and Lee, V.H.L. (1999). Organic cation transporter in rabbit alveolar epithelial cell monolayers. Pharm. Res. 16, 1279-1286. Shepard, A.R. and Rae, J.L. (1998). Ion transporters and receptors in cDNA libraries from lens and cornea epithelia. Curr. Eye. Res. 17(7), 708-719. Shu, Y., Bello, C.L., Mangravite, L.M., Feng, B. and Giacomini, K.M. (2001). Functional characteristics and steroid hormone-mediated regulation of an organic cation transporter in 272 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Madin-Darby canine Kidney cells. Journal of Pharmacology and Experimental Therapeutics. 299, 392-398. Siefert, B., and Keipert, S. (1997). Influence of alpha-cyclodextrin and hydroxyalkylated beta-cyclodextrin derivatives on the in vitro comeal uptake and permeation of aqueous pilocarpine-HCl solutions. J. Pharm. Sci., 86, 716-720. Sieg, J.W., and Robinson, J.R. (1975). Vehicle effects on ocular drug bioavailability i: evaluation of fluorometholone. J. Pharm. Sci. 64, 931-936. Sloan, J.L. and Mager, S. (1999). Cloning and functional expression of a human Na+ -and Cl' -dependent neutral and cationic amino acid transporter. J. Biol. Chem. 274, 23740-23745. Smet, M.D., and Nussenblatt, R.B. (1993). Clinical use of cyclosporine in ocular disease. Int. Ophthalmol. Clin. 33, 31-45. Smulders, A.P. and Wright, E.M. (1971). Galactose transport across the hamster small intestine: the effect of sodium electrochemical potential gradients. J. Physiol. 212(1), 277- 286. Sonoda, T. and Tatibana, M. (1978). Metabolic fate of pyrimidines and purines in dietary nucleic acids ingested by mice. Biochim. Biophys. Acta. 521, 55-56. Sorensen, T., and Jensen, F.T. (1979). Tear flow in normal human eyes: Determination by means of radioisotope and y camera. Acta Ophthalmol., 57, 564. Spahn-Langguth, H., Baktir, G., Radschuweit, A., Okyar, A., Terhaag, B., Ader, P., Hanafy, A. and Langguth, P. (1998). P-glycoprotein transporters and the gastrointestinal tract: evaluation of the potential in vivo relevance of in vitro data employing talinolol as model compound. Int. J. Clin. Pharmacol. Ther. 36(1), 16-24. Steele, R.E., Preston, A. S., Johnson, J. P., and Handler, J. S. (1986). Porous-bottom dishes for culture of polarized cells. Am. J. Physiol. 251, 136-139. Stewart, B.H., Chan, O.H., Lu, R.H., Reyner, E.L., Schmid, H.L., Hamilton, H.W., Steinbaugh, B.A. and Taylor, M.D. (1995). Comparison of intestinal permeabilities determined in multiple in vitro and in situ models: relationship to absorption in humans. Pharm. Res. 12, 693-699. Stocker, F.W., Eiring, A., Georgiade, R., and Beorgiade, N. (1958). A tissue culture technique for growing comeal epithelial stromal, and endothelial tissues separately. Am. J. Ophthalmol. 46, 294-298. Suhonen, P., Jarvinen, T., Peura, P., and Urtti, A. (1991). Permeability of pilocarpic acid diesters across albino rabbit cornea in vitro. Int. J. Pharm., 74, 221-228. Sun, T.T., and Green, H. (1977). Cultured epithelial cells of cornea, conjunctiva and skin: absence of marked intrinsic divergence of their differentiated states. Nature, 269, 489-492. 273 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tak, R.V., Pal, D. Gao, H., Dey, S., Mitra, A.K. (2001) Transport of acyclovir ester prodrugs through rabbit corea and SIRC-rabbit corneal epithelial cell line. Journal of Pharmaceutical Sciences. 90, 1505-1515. Takahashi, H., Kaminski, A.E., and Zieske, J.D. (1996). Glucose transporter 1 expression is enhanced during corneal epithelial wound repair. Exp. Eye. Res. 63, 649-659. Tamai, I., Ohashi, R., Nezu, J., Yabuuchi, H., Oku, A., Shimane, M., Sai, Y., and Tsuji, A. (1998). Molecular and functional identification of sodium-dependent, high affinity human carnitine transporter OCTN2. J. Biol. Chem. 273, 20378-20382. Tamai, I., Yabuuchi, H., Nezu, J.-I., Sai, Y., Oku, A., Shimane, M. and Tsuji, A. (1997). Cloning and characterization of a novel human pH-dependent organic cation transporter, OCTN1. FEBSLett. 419, 107-111. Tanaka, K., Fujita, T., Yamamoto, Y., Murakami, M., Yamamoto, A and Muranishi, S. (1996). Enhancement of intestinal transport of thyrotropin-releasing hormone via a carrier- mediated transport system by chemical modification with lauric acid. Biochim. Biophys. Act. 1283(1), 119-126. Tang-Liu, D.D.-S., Liu, S.S., and Weinkam, R.J. (1984). Ocular and systemic bioavailability of ophthalmic flurbiprofen. J. Pharmacokinet. Biopharm., 12, 611. Tang-Liu, D.D.-S., Liu, S., Richman, J., and Weinkam, R. (1986). HPLC quantitation of levobunolol and its metabolite, dihydrolevobunolol, in biological fluids. J. Liquid Chromatography. 9(10), 2237-2252. Tang-Liu, D.D.-S., Richman, J.B., Weinkam, R.J., and Tarkruri, H. (1994). Effect of four penetration enhancers on corneal permeability of drugs in vitro. J. Pharm. Sci. 83, 85-90. Tang-Liu, D.D.-S., Shackleton, M., and Richman, J.B. (1988). Ocular metabolism of levobunolol. J. Ocular Pharm. 4(3), 269-278. Tao, W., Liou, G.I., Wu, X. and Reinach, P.S. (1995). ETb and epidermal growth factor receptor stimulation of wound closure in bovine corneal epithelial cells. Investigative Ophthalmology and Visual Science. 36(13) 2614-2622. Tavakoli-Saberi, M.R., and Audus, K.L. (1989). Physicochemical factors affecting beta- adrenergic antagonist permeation across cultured hamster pouch buccal epithelium. Int. J. Pharm. 56, 135-142. Taylor, D.C., Pownall, R. and Burke, W. (1985). The absorption of beta-adrenoceptor antagonists in rat in-situ small intestine; the effect of lipophilicity. J. Pharm. Pharmacol. 37(4), 280-283. Terada, T., Sawada, K., Saito, H., Hashimoto, Y., and Inui, K. (1999). Functional characteristics of basolateral peptide transporter in the human intestinal cell line Caco-2. Am. J. Physiol. Gastrointest. Liver Physiol. 276, G1435-G1441. 274 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Terada, T., Sawada, K., Ito, T., Saito, H., Hashimoto, Y., and Inui, K.-I. (2000). Functional expression of novel peptide transporter in renal basolateral membranes. Am. J. Physiol. Renal Physiol. 279, F851-F857. Terao, T., Hisanaga, E., Sai, Y., Tamai, I., and Tsuiji, A. (1996). Active secretion of drugs from the small intestinal epithelium in rats by p-glycoprotein functioning as an absorption barrier. J. Pharm. Pharmacol. 48, 1083-1089. Thiel, M.A., Morlet, N., Schulz, D., Edelhauser, H.F., Dart, J.K., Coster, D.J., Williams, K.A. (2001). A simple comeal perfusion chamber for drag penetration and toxicity studies. Br. F. Ophthalmol. 85,450-453. Thwaites, D.T., Hirst, B.H. and Simmons, H.L. (1993). Passive transepithelial absorption of thyrotropin-releasing hormone (TRH) via a paracellular route in cultured intestinal and renal epithelial cell lines. Pharm. Res. 10(5), 674-681. Tielsch, J.M., Sommer, A., Witt, K., Katz, J., and Royall, R.M. (1990). Blindness and visual impairment in an American urban population: The Baltimore Eye Survey. Arch. Ophthalmol. 108, 286-290. Tribulova, N., Sosner, I., Varon, D., and Manoach, M. (1999). Attenuation of Ca paradox injury in guinea pig heart by K+ channel blocker, d-sotalol. J. Submicrosc. Cytol. Pathol. 31(1), 137-147. Trinkaus-Randall, V., Newton, A. and Franzblau, C. (1988). Influence of substratum on comeal epithelial cell growth and protein synthesis. Investigative Ophthalmology and Visual Science. 29, 1800-1809. Tumheim, K. and Lauterbach, K.O. (1977). Absorption and secretion of monoquatemary ammonium comounds by the isolated intestinal mucosa. Biochem. Pharmacol. 26, 99-108. Tumheim, K. and Lauterbach, K.O. (1980). Interaction between intestinal absorption and secretion of monoquatemary ammonium compounds in guinea pigs-a concept for the absorption kinetics of organic cations. J. Pharmacol. 212,418-424. Ueda, H., Horibe, Y., Kim, K.-J. and Lee, V.H.L. (2000). Functional characterization of organic cation drag transport in the pigmented rabbit conjunctiva. Invest. Ophthalmol. Vis. Sci. 41(3), 870-876. Ullrich, KJ. (1994). Specificity of transporters for ‘organic ions’ and ‘organic cations’ in the kidney. Biochim. Biophys. Acta. 1197, 45-62. Urtti, A. (1994). Delivery of antiglaucoma drugs: ocular vs. systemic absorption. Journal of Ocular Pharmacology. 10, 349-357. Urtti, A. and Salminen, L. (1993). Minimizing systemic absorption of topicallhy administered opthalmic drugs. Surv. Ophthalmol. 37(6), 435-456. 275 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ussing, H.H., and Zehran, K. (1950). Active transport of sodium as the source of electric current in the short-circuited isolated frog skin. Acta Physiol. Scand. 23, 110-126. Vaidyanathan, G., Jay, M., Bera, R.K., Mayer, P.R., and Brazzell, R.K. (1990). Scintigraphic evaluation of the ocular disposition of 18F-imirestat in rabbits. Pharm. Res., 7, 1198. Van Bree, J.B., de Boer, A.G., Danhof, M., Ginsel, L.A., and Breimer, D.D. (1988). Characterization of an “in vitro” blood-brain barrier: Effects of molecular size and lipophilicity on the cerebrovascular endothelial transport rates of drugs. J. Pharmacol. Exp. Ther. 247, 1233-1239. Van der Heyden, C., Weekers, J.F. and Schoffeniels, E. (1975). Sodium and chloride transport across the isolated rabbit cornea. Exp. Eye. Res. 20(1), 89-96. Van Meer, G. and Simons, K. (1986). The function of tight junctions in maintaining differences in lipid composition between the apical and the basolateral cell surface domains of MDCK cells. EMBO. J. 5(7), 1455-1464. Walter, E. and Kissel, T. (1994). Transepithelial transport and metabolism of thyrotropin- releasing hormone (TRH) in monolayers of a human intestinal cell line (Caco-2): evidence for an active transport component? Pharm. Res. 11(11), 1575-1580. Wang, L., Chen, L., Walker, V. and Jacob, T.J. (1998). Antisense to MDR1 mRNA reduces P-glycoprotein expression, swelling-activated Cl- current and volume regulation in bovine ciliary epithelial cells. J. Physiol. 511(1), 33-44. Wang, X., Fedorko, L., Marunaka, Y. and O’Brodovich, H. (1993). Whole-cell Cl- currents in a human peripheral airway epithelial cell line. Can. J. Physiol. Pharmacol. 71(9) 662-670. Wang, W., Sasaki, H., Chien, D.-S. and Lee, V.H.L. (1991). Lipophilicity influence on conjunctival drug penetration in the pigmented rabbit: a comparison with corneal penetration. Curr. Eye Res., 10, 571-579. Whikehart, D.R., Montgomery, B., Soma, D.H., and Wells, J.D. (1991). Beta-blocking agents inhibit Na+K+ATPase in cultured comeal endothelial and epithelial cells. J. Ocul. Pharmacol. 7(3), 195-200. Wiederholt, M., Kossendrup, D., Schulz, W., et al. (1986). Pharmacokinetic of topical cyclosporin A in the rabbit eye. Invest Ophthalmol Vis Sci. 27, 519-524. Wilcox, G.E. (1969). Preparation of primary bovine comeal epithelia cell cultures for use in virological investigations. Applied Microbiology. 18(2), 268-269. Woodcook, S., Williamson, I., Hassan, I. And Mackay, M. (1991). Isolation and characterisation of clones from the Caco-2 cell line displaying increased taurocholic acid transport. J. Cell. Sci. 98, 323-332. 276 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Wu, J., Zhang, J.J., Koppel, H. and Jacob, T.J. (1996). P-glycoprotein regulates a volume- activated chloride current in bovine non-pigmented ciliary epithelial cells. J. Physiol. 491(3), 743-755. Wu, N., Chiang, C., and Lee, A. (1993). Studies of carbonic anhydrase inhibitors: Physicochemical properties and bioactivities of new thiadiazole derivatives. J. Ocular Pharmacol., 9, 97-108. Wu, X., Huang, W., Prasad, P.D., Seth, P., Rajan, D.P., Leibach, F.H., Chen, J., Conway, S J. and Ganapathy, V. (1999). Functional characteristics and tissue distribution pattern of organic cation transporter 2 (OCTN2), an organic cation/carnitine transporter. J. Pharmacol. Exp. Ther. 290, 1482-1492. Wu., X., Prasad, P.D., Leibach, F.H. and Ganapathy, V. (1998). cDNA sequence , transport function, and genomic organization of human OCTN2, a new member of the organic cation transporter family. Biochem. Biophys. Res. Commun. 246, 589-595. Yamashita, K., Kim, KJ. and Lee, V.H.L. (1996). Transport characteristics of L-camosine in rabbit tracheal epithelial cell monolayers cultured at an air-interface. Pharm. Res. 13, S249. Yamaya, W., Finkbeiner, W.E., Chun, S.Y., and Widdicombe, J.H. (1992). Differentiated structure and function of cultures from human tracheal epithelium. Am. J. Physiol. 262, L713-L724. Yang, J.J. and Lee, V.H.L. (2000). Drug efflux pumps in rabbit conjunctival epithelial cells. Ph.D. Dissertation, USC, Dept, of Pharm. Sciencs, Los Angeles. Yang, J.J., Ueda, H., Kim, K. J., and Lee, V. H. L. (2000). Meeting future challenges in topical ocular drug delivery: Development of an air-interfaced primary culture of rabbit conjunctival epithelial cells on a permeable support for drug transport studies. J. Contr. Rel. Yang, J.J., Kim, KJ., and Lee, V.H.L. (2000). Role of P-glycoprotein in restricting propranolol transport in cultured rabbit conjunctival epithelial cell layers. Pharm. Res. 17(5), 533-538. Yang, W. and Acosta D. (1994). Cytotoxicity potential of surfactant mixtures evaluated by primary cultures of rabbit comeal epithelial cells. Toxicology Letters. 70, 309-318. Yanoff, M. and Cameron, J.D. (1977). Human cornea organ cultures: epithelial -endothelial interactions. Invest. Ophthalmol. Vis. Sci. 16(4), 269-273. Yee, Shiyin. (1997). In vitro permeability across Caco-2 cells (colonic) can predict in vivo (small intestinal) absorption in man-fact or myth. Pharmaceutical Research. 14, 763-766. Yokogawa, K., Miya, K., Tamai, I., Higashi, Y., Nomura, M., Miyamoto, K. and Tsuji, A. (1999). Characteristics of L-camitine transport in cultured human hepatoma HLF cells. J. Pharm. Pharmacol. 51(8), 935-940. 277 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Yokogawa, K., Yonekawa, M., Tamai, I., Ohashi, R., Tatsumi, Y., Higashi, Y., Nomura, M., Hashimoto, N., Nikaido, H., Hayakawa, J., Nezu, J., Oku, A., Shimane, M., Miyamoto, K. and Tsuji, A. (1999). Loss of wild-type carrier-mediated L-camitine transport activity in hepatocytes of juvenile visceral steatosis mice. Hepatology. 30(4), 997-1001. Yokohama, S., Yamashita, K., Toguchi, H., Takeuchi, J., and Kitamori, N. (1984). Absorption of thyrotropin-releasing hormone after oral administration of TRH tartrate monohydrate in the rat, dog and human. J. Pharmacobio-Dyn. 7, 101-111. Yokohama, S., Yoshioka, T., Yamashita, K., and Kitamori, N. (1984). Intestinal absorption mechanism of thyrotropin-releasing hormone. J. Pharmacobio-Dyn. 7,445-451. Yorio, T. and Bentley, PJ. (1978). Stimulation of the short-circuit current (sodium transport) across the skin of the frog (Rana pipiens) by corticosteroids: structure-activity relationships. J. Endocrinol. 79(3), 283-290. Yoshida, F. and Topliss, J.G. (1996). Unified model for the corneal permeability of related and diverse compounds with respect to their physicochemical properties. J. Pharm. Sci. 85(8), 819-823. Yoshizumi, M.O., Cohen, D., Verbukh, I., Leinwand, M., Kim, J., and Lee, D.A. (1987). Experimental transscleral iontophoresis of ciprofloxacin. J. Ocul. Pharmacol., 7, 163. Zadunaisky, J.A. (1966). Active transport of chloride in frog cornea. Am. J. Physiol. 211(2), 506-512. Zadunaisky, J.A. andLande, M.A. (1971). Active chloride transport and control of corneal transparency. Am . J. Physiol. 221(6), 1837-1844. Zamora, J.M., Pearce, H.L., and Beck, W.T. (1988). Physical-chemical properties shared by compounds that modulate multidrug resistance in human leukemic cells. Mol. Pharmacol. 33,454-462. Zhang, L., Schaner, M.E. and Giacomini, K.M. (1998). Functional characterization of an organic cation transporter (hOCTl) in a transiently transfected human cell line (HeLa). J. Pharmacol. Exp. Ther. 286(1), 354-361. Zimmerman, T.J., Kooner, K.S., Kandarakis, A.S. and Ziegler, L.P. (1984). Improving the therapeutic index of topically applied ocular drugs. Arch. Ophthalmol. 102(4), 551-553. Zollner, N. (1982). Purine and pyrimidine metabolism. Proc. Nutr. Soc. 41, 329-342. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Drug efflux pumps in rabbit conjunctival epithelial cells
PDF
Conjunctival epithelial uptake of biodegradable nanoparticles: Mechanism, intracellular distribution, and absorption enhancement
PDF
Alteration of the in vitro and in vivo processing of a polypeptide, BBI, through conjugation with palmitic acid
PDF
Cytochrome P450 3A metabolism in the rabbit lacrimal gland and conjunctiva
PDF
Drug delivery to the posterior region of the eye: Characterization of organic cation transporters
PDF
Insulin-transferrin conjugate as an oral hypoglycemic agent
PDF
Characterization of dynamin II and clathrin adaptor complex in the regulation of gastric hydrogen,potassium-ATPase recycling in oxyntic cells
PDF
Arginine deiminase-mediated modulation of argininosuccinate and nitric oxide synthesis in cultured cell lines
PDF
Basolateral transport of nucleosides in rabbit tracheal epithelial cells: An equilibrative nucleoside transporter outcome
PDF
Design, synthesis, biological testing and QSAR analysis of new Schiff bases of N-hydroxysemicarbazide as inhibitors of tumor cells
PDF
Effect of lipidization on transport and uptake of peptides across rat alveolar epithelial cell monolayers
PDF
A dynamic apical actin cytoskeleton facilitates exocytosis of tear proteins in rabbit lacrimal acinar epithelial cells
PDF
An air-interface rabbit tracheal epithelial cell culture model for the characterization of nucleoside and nucleoside drug transport
PDF
High mobility group A2 (HMGA2): Molecular dissections of its functions and regulation
PDF
DNA structures associated with the Fragile X triplet repeat sequences
PDF
Characterization of microtubule motors in the apical trafficking of hydrogen,potassium-ATPase in oxyntic cells
PDF
How cells cry: The cytoskeleton as a facilitator of regulated secretion and membrane trafficking in lacrimal gland acinar cells
PDF
Characterization of lacrimal gland acinar cell cytoskeletal structure and its participation in stimulated secretion
PDF
Computer prediction of peptide binding to class I MHC molecules
PDF
Filter-grown Madin-Darby canine kidney cell monolayers: as an in-vitro model for the study of pentration enhancers
Asset Metadata
Creator
Chang-Lin, Joan-En
(author)
Core Title
Drug transport mechanisms in the rabbit corneal epithelium: In vitro-in vivo correlation
School
Graduate School
Degree
Doctor of Philosophy
Degree Program
Pharmaceutical Sciences
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
health Sciences, ophthalmology,Health Sciences, Pharmacology,OAI-PMH Harvest
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Lee, Vincent H.L. (
committee chair
), Haworth, Ian S. (
committee member
), Koda, Robert T. (
committee member
), Okamoto, Curtis T. (
committee member
), Shen, Wei-Chiang (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-630144
Unique identifier
UC11334994
Identifier
3116676.pdf (filename),usctheses-c16-630144 (legacy record id)
Legacy Identifier
3116676.pdf
Dmrecord
630144
Document Type
Dissertation
Rights
Chang-Lin, Joan-En
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
health Sciences, ophthalmology
Health Sciences, Pharmacology