University of Southern California Dissertations and Theses

An air-interface rabbit tracheal epithelial cell culture model for the characterization of nucleoside and nucleoside drug transport

(USC Thesis Other)

An air-interface rabbit tracheal epithelial cell culture model for the characterization of nucleoside and nucleoside drug transport

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text direct^ fix > m the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter 6ce, tWule others may be fix > m any type o f computer printer. 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 bleedthrough, substandard margins, I and improper alignment can adversety afifect reproduction. Si In the unlikdy event that the author did not send UMI 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. Oversize materials (e g., maps, drawings, charts) are reproduced by sectioning the original, b^inning at the upper Idt-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back o f the book. Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available ft> r any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information C om paiQ r 300 North Zeeb Road, Ann Arbor MI 48106-1346 USA 313/76M700 800/521-0600 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. AN AIR-INTERFACE RABBIT TRACHEAL EPITHELIAL CELL CULTURE MODEL FOR THE CHARACTERIZATION OF NUCLEOSIDE AND NUCLEOSIDE DRUG TRANSPORT by Neil R. Mathias 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, 1997 Copyright 1997 Neil R. Mathias Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. X J M X Number: 9816102 UMI Microform 9816102 Copyright 1998, by UMI Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. UMI 300 North Zeeb Road Ann Arbor, MI 48103 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. L ' n iv e r s it y o f s o u t h e r n C a l i f o r n i a THE g r a d u a t e s c h o o l UNrvHRsrrv park LOS ANGELES. CALIFORNIA 9000? This dissertaticn, xcritten by R. _ jto t l ^ a s ........................................ under the direction of h.ds Dissertation Committee, and approved by aU its menvsers, has been presented to and accepted by Tne Graduate School in partial fulfillment of re quirements for the degree of .DOCTOR OF PHILOSOPHY D ean cr Graduate Stuaies Date ....... DISSERTA TIO N C O M M ITTEE •••••••••••••••••••••••••••**•*»* ••••••••••••• Chairperson * * * * * * * * L < . Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Dedicated to my parents and my late sister Lynette. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENT I would like to express m y sincere appreciation to Dr. Vincent Lee for being an outstanding graduate advisor. H is excellent guidance and p erspicui^ in the pursuit o f science has been inspirational. The encouragement, high standards and scientific fieedom have made my graduate education scientifically stim ulating, challenging and enjoyable. His untiring efforts to be the best, have left invaluable and indelible lessons in the strife for excellence both professionally and personally. I would like to sincerely thank my com m ittee. Dr. Kwang-Jin Kim, Dr. Ian Haworth, Dr. Herman von G rafenstein, and Dr. W ei-Chiang Shen for their trem endous encouragement and insightful suggestions through all the long and frequent com m ittee meetings. Their sound guidance and experimental expertise has helped to shape the research in this dissertation fo r w hich I am very grateful. I would also like to acknowledge Dr. M elvin Trousdale for his enthusiastic assistance and support in all the virus-related w ork in this dissertation. Lastly, I would like to thank all my laboratory colleagues whom I had the opportunity to w ork w ith in the past and the present. Their support and help m ade my graduate education a m em orable and enjoyable one. Ill Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. i ACKNOWLEDGEMENT_______________________________________________ üi LIST OF TABLES:_____________________________________________________ix LIST OF SCHEMES:___________________________________________________xi LIST OF FIGURES:___________________________________________________ xü LIST OF ABBREVIATIONS___________________________________________ xiv I. INTRODUCTION____________________________________________________ 1 1. GENERAL INTRODUCTION_________________________________________ 2 2. THE AIRWAY EPITHELIUM-------------------------------------------------------------- 4 2.1. B io lo g y o f th e A ir w a y s...........................................................................................................4 2.2. So lu te a n d D r u g T ran sport in t h e A ir w a y: P a u c it y o f In f o r m a t io n............. 7 2.3. N eed f o r L o ca l D eliv er y.........................................................................................................9 2.4. M o d els fo r R espir a to r y D ru g D e l iv e r y ......................................................................12 2.4.L In vivo M odels:............................................................................................................ 12 2.3.2. In vitro M odels:........................................................................................................... 13 2.4. A irw a y E pith elia l C ell Cu l t u r e s.................................................................................... 16 3. EPITHELIAL BARRIERS TO DRUG ABSORPTION------------------------------23 3.1. P en etra tio n B a r r ie r .............................................................................................................. 23 3.1.1. M ucociliary E scalator................................................................................................23 3.1.2. The A pical M em brane................................................................................................24 3.1.3. The Tight Junction...................................................................................................... 25 3.2. En zy m a tic B a r r ie r ..................................................................................................................27 4. RESPIRATORY VIRAL INFECTIONS------------------------------------------------- 27 4.1. Stra teg ies to c o u n ter viral in f e c t io n s........................................................................28 4.2.CURRENT R espir a t o r y An tiv ira l T r e a t m e n t................................................................29 4.3. D eliv er y St r a t e g ie s................................................................................................................30 4.3.1. Prodrug Approach.......................................................................................................31 4.3.2. D rug D elivery System s...............................................................................................32 4.3.3. C arrier-m ediated Transport...................................................................................... 32 4.4. In fec tio n of A ir w a y E pithelial C e l l s ........................................................................... 34 4.4.1. Adenovirus Infection................................................................................................... 34 4.4.2. Virus-induced C ellular C hanges.............................................................................. 35 IV Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5. NUCLEOSIDE TRANSPORTERS____________________________________ 38 5.1. H is t o r y o f N u c l e o s id e T r a n s p o r t e r s..................................................................................39 5.2. C l a s s if ic a t io n o f N u c l e o s id e T r a n s p o r t e r s.................................................................. 40 5.3. E q u il ib r a t iv e v s. C o n c e n t r a t iv e T r a n s p o r t e r s .........................................................43 5.4. N a ^-d e p e n d e n c e ................................................................................................................................. 43 5.5. Kin e t ic P a ra m eters a n d S e l e c t iv it y.............................................................................. 44 5.6. N u c l e o s id e T r a n s p o r t e r s in O t h e r T is s u e s .....................................................................45 5.6.1. L ung................................................................................................................................45 5.6.2. Intestinal E pithelial C ells...........................................................................................46 5.6.3. K idney E pithelial C ells............................................................................................... 48 5.6.4. B rain Tissues................................................................................................................ 49 5.6.5. O ther C ell Types.......................................................................................................... 50 5.7. Ex p r e s s io n a n d C l o n in g ...............................................................................................................51 n . STATEMENT OF THE PROBLEM___________________________________ 57 G oals a n d O b jec tiv e s.......................................................................................................................58 m . MATERIALS AND METHODS_____________________________________ 65 1. A n im a l M o d e l................................................................................................................................. 66 2. R a b b it T r a c h ea l E p it h e l ia l C u l t u r e s.............................................................................. 66 3. B io elec tr ic M e a s u r e m e n t s..................................................................................................... 71 4. E lec t r o n M icro sc o py o f T ra c h ea l Epith elia l C u l t u r e s........................................71 5. P er io d ic A c id Sch iff (P A S ) Sta in in g o f T r a c h e o c y t e s..............................................72 6. UssiNG C h a m b er St u d ie s ........................................................................................................... 73 7. TRANSEPITHELIAL DRUG TRANSPORT STUDIES....................................................................... 74 8. M ea su r em en t of N u c l e o sid e U ptak e:.................................................................................80 9. M e t a b o l ism o f N u c l e o sid e s: ...................................................................................................82 10. In fe c t io n o f E pit h e lia l C ell C u ltu res............................................................................ 82 (a) Infectivity............................................................................................................................ 82 (b) M easurem ent o f B ioelectric Properties in A denovirus Type 5 Infected C ells 83 (c) M easurem ent ofT ransepithelial M annitol Transport in A denovirus Type 5 Infected C ells........................................................................................................................... 84 (d) M easurem ent o f N ucleoside Transport Function in Adenovirus Type 5 Infected C ells...........................................................................................................................................84 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IV. RESULTS________________________________________________________ 95 A. DEVELOPMENT OF AN AIRWAY EPITHELIAL CELL CULTURE M ODEL_____________________________________________________________ 96 1. P r im a r y C u ltured M onolayers o f T ra ch ea l E pith elia l C e l l s............................96 2. M o r ph o l o g y of A ir w a y E pithelial C u l t u r e s................................................................ 96 2.1 Electron M icroscopy........................................................................................................96 2.2 P eriodic A cid S c h iff Staining..........................................................................................97 3. B io el ec tr ic pro perties o f A irw a y E pith elia l C ell C u ltu r es.................................99 3.1 Selectivity fo r Substratum (Extracellular M atrix).....................................................102 3.2 E ffect o f Fetal Bovine Serum (FBS).............................................................................104 B. ELECTROPHYSIOLOGICAL AND BIOCHEMICAL CHARACETRIZATION______________________________________________ 106 1. Ef f e c t of O u a b a in .......................................................................................................................106 2. E f f e c t o f A m ilo rid e a n d F u r o sem id e................................................................................ 108 3. E f f e c t of B e n z a m il.................................................................................................................... 112 4. E f f e c t o f N -P h en y la n th ra n ilic A c i d ............................................................................... 114 5. E f f e c t o f 4,4’-D iiso th io cy a n o to stilb en e-2,2’d isu lfo n ic acid (D ID S )..............116 6. B io c h e m ic a l C h a r a c t e r i z a t i o n ........................................................................................... 119 C. PERMEABILITY CHARACTERIZATION____________________________122 1. T r a n sp o r t C h a ra cteristics of M o d e l So lutes in A ir- in terfa ced (A IC ) a nd L iq u id-c o v er ed C u ltu r es (L C C )...............................................................................................122 2. In f l u e n c e o f So l u te M o lec u la r S i z e ............................................................................... 125 FTTC-Dextran Transport..................................................................................................... 125 3. In f l u e n c e of So l u te L ipo ph ilic ity...................................................................................... 132 fi-A drenergic Com pound Transport...................................................................................132 D. EVIDENCE FOR CARRIER-MEDIATED NUCLEOSIDE UPTAKE---------136 1. M et a b o l ism O F N u c l e o sid e s...................................................................................................136 2. U r id in e In flu x in to a irw a y e pith elia l c e l l s.................................................................138 3. NA^-DEPENDENCE OF URIDINE UPTAKE.................................................................................... 138 4. S e n sit iv ity TO In h ib it o r s......................................................................................................... 141 5. C o n c en tr a tio n D e pe n d e n c e.................................................................................................. 144 6. In h ib it io n St u d ie s........................................................................................................................146 7. S u b st r a t e Selec tiv ity.............................................................................................................. 149 8. U p t a k e of N u cleo sid e A n a lo g s: 2 ’-D e o x y a d en o sin e an d H P M P C ....................150 E. FUNCTIONAL IDENTIFICATION OF TRANSPORTERS: INHIBITION PATTERN WITH PHYSIOLOGICAL NUCLEOSIDES------------------------------153 F. TRANSEPITHELIAL TRANSPORT AND LOCALIZATION OF NUCLEOSIDE TRANSPORTERS______________________________________ 156 vi Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. G. ADENOVIRUS INFECTION OF AIRWAY EPITHELIAL CELLS----------161 1. In f e c t io n o f A ir w a y E pit h e l ia l C e l l s ..................................................................................161 2. E f f e c t o f A d e n o v ir u s T ype 5 In f e c t io n o n E pit h e l ia l S t r u c t u r e........................164 2.1 Alteration o f E pithelial Bioelectric Properties..........................................................164 2.2 Infection and Paracellular Perm eability................................................................... 165 3. In f e c t io n a n d C a r r ie r - m e d ia t e d N u c l e o s id e T r a n s p o r t F u n c t io n ....................165 V. DISCUSSION____________________________________________________ 169 A. DEVELOPMENT AND CHARACTERIZATION OF AIRWAY EPITHELIAL CELL CULTURE MODEL____________________________________________ 171 1. P r im a r y C u l t u r e o f A irw a y E p i t h e l i a l C e l l s .............................................................171 1.1 E ffect o f M edia Supplem ents........................................................................................171 1.2 E ffect o f S u b stra ta ........................................................................................................./ 73 1. A irw a y E p i th e li a l C e l l C u l t u r e M o d e ls , AIC v s. L C C ..........................................174 2.1.1. M orphology a n d Cellular D ifferentiation.............................................................174 2.1.2 Comparison w ith Native Tissue................................................................................177 2.2. B io e le c tr ic P r o p e r t i e s .........................................................................................................178 2.3. Io n T r a n s p o r t P r o p e r t i e s ................................................................................................... 180 2.3.1. Electrogenic N a^-current in A irw ay Epithelial Cells.......................................... 181 2.3.1. Electrogenic CF current in A irw ay Epithelial Cells.............................................182 2.4. C o m p a riso n w ith N a tiv e T issu e a n d O t h e r A irw a y C u l t u r e M o d e ls ..........184 3. B io c h e m ic a l C o m p a ris o n ........................................................................................................189 4. E f f e c t o f C u l t u r e C o n d itio n o n D r u g T r a n s p o r t .................................................... 192 B. PERMEABILITY CHARACTERISTICS OF AIRWAY EPITHELIUM 196 Comparison o f A irw ay and Alveolar E pithelial Cells Culture M odels....................... 200 C. NUCLEOSIDE TRANSPORT STUDIES______________________________206 1 C a r r ie r - m e d ia te d U rid in e U p ta k e in A irw a y E p ith e lia l C e l l s ...........................206 1.1 Uptake Rate, Directionality, Na* and Temperature dependence......................... 206 1.2. Inhibition Studies..........................................................................................................208 1.3. Substrate Selectivity o f the N ucleoside Transport P rocess....................................209 1.4. Uptake o f N ucleoside Analogs ^H-2 ’ -Deoiyadenosine a n d ^H-HPMPC............ 210 2. F u n c tio n a l I d e n tif i c a tio n o f t h e N u c le o s id e T r a n s p o r t e r s .............................. 211 3. B a s o l a t e r a l L o c a liz a tio n ....................................................................................................213 4. N u c le o s id e T r a n s p o r t e r S ite s .............................................................................................215 5. N u c le o s id e T r a n s p o r t e r R e g u l a tio n ..............................................................................216 6. P o s s ib le P h y s io lo g ic a l R o le o f N u c le o s id e T r a n s p o r t e r s ..................................218 7. R a t i o n a l D e sig n o f A n t i v i r a l A n a l o g s ......................................................................... 220 8. F e a s ib ility o f A ir w a y A n t i v i r a l N u c le o s id e D ru g D e l i v e r y ............................. 222 VII Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. D. Vm US INFECTION STUDIES OF THE AIRWAY EPITHELIUM-----------224 1. In fec tio n o f A irw a y E pith elia l C e l l s ............................................................................ 224 2. B a r r ie r Pr o perties a n d A d e n o v ir u s T ype 5 In fec tio n............................................ 226 3. In fe c t io n a n d C a rr ier-m ed ia ted N ucleoside T r a n spo r t F u n c t io n..................228 E. IN V ITR O -m VIVO CORRELATION-------------------- 231 VI. CONCLUSION__________________________________________________ 237 A. SUMMARY OF FINDINGS_________________________________________ 238 B. SIGNIFICANCE OF THE FINDINGS________________________________ 243 C. FUTURE CONSIDERATIONS______________________________________ 246 V n. REFERENCES__________________________________________________ 252 VIII Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES: Page Table 1-1 : Summary o f airw ay epithelial cell culture m odels 20 Table 1-2: N ucleoside analog interaction w ith nucleoside or nucleobase transporters 52 Table 1-3: N ucleoside transporters and their kinetic properties in various tissues 54 Table 3-1 : List o f all chem ical suppliers and their location 86 Table 3-2: List o f H PLC conditions for B-adrenergic analysis 79 Table 4-1 : B ioelectric properties o f rabbit tracheal epithelial cells 101 Table 4-2: Effect o f substratum on airway epithelial bioelectric properties 103 Table 4-3 : Effect o f FBS on airw ay epithelial bioelectric properties 105 Table 4-4: Half-tim es for m axim al response and R t changes w ith ion transport inhibitors 118 Table 4-5: Lactate release rate from AIC and LCC 121 Table 4-6: Perm eability coefScients o f model solutes across AIC and LCC 124 Table 4-7: M olecular w eight, size, difRisivity, and perm eability coefficients ofFIT C -dextrans 127 Table 4-8: Perm eability o f cultured airw ay and alveolar epithelial cell m onolayers to B-adrenergic agents 135 Table 4-9: Uptake rate o f uridine from the apical or basolateral m em brane, or in the presence o r absence o f Na^ and nitrobenzythioinosine, NBTI 142 Table 4-10: Effect o f ion transport inhibitors on uridine uptake 143 IX Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T able 4-11: Inhibition o f ^H-uridine, H -thym idine and H -inosine w ith physiological nucleosides and other physiological substrates. 147 T able 4-12: Na^-dependence and kinetic param eters o f uridine, thym idine and inosine uptake 155 T able 4-13: Transepithelial perm eability o f uridine, thym idine and inosine across airway epithelial cell m onolayers 158 T able 5-1 : Com parison o f polar solute perm eability in nasal, tracheal and alveolar epithelial barriers 195 T able 5-2: Param eters o f sigm oidal relationship between B-adrenergic agent perm eability in airw ay and alveolar epithelial cell cultures 204 Table 5-3: Com parison o f pulm onary absorption rate o f hydrophilic solutes in various experim ental m odels 205 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF SCHEMES: Page Schem e 1-1 : Classificatioa o f nucleoside transporters 42 Schem e 2-1: Flowchart o f specific aim # 1 61 Schem e 2-2: Flowchart o f specific aim # 2 64 Schem e 3-1 : Flowchart o f steps in th e culturing o f rabbit tracheal epithelial cells 69 XI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES: Page Fig. 1-1: Biology o f the airways 6 Fig. 3-1 : Chem ical structures o f inhibitors 91 Fig. 3-2: Chem ical structures o f m odel solutes 92 Fig. 3-3: Chemical structures o f nucleosides 93 Fig. 3-4: Chemical structures o f nucleoside analogs 94 Fig. 4-1 : Electron m icroscopy o f airway epithelial cell cultures 98 Fig. 4-2: Bioelectric properties o f airway epithelial cell cultures 100 Fig. 4-3: Effect o f ouabain on airw ay epithelial cell cultures 107 Fig. 4-4: Effect o f am iloride and furosem ide on airw ay epithelial cell cultures 110 Fig. 4-5: Effect o f furosem ide and am iloride on airw ay epithelial cell cultures 111 Fig. 4-6: Effect o f benzam il on airw ay epithelial cell cultures 113 Fig. 4-7: Effect o f NPAA on airway epithelial cell cultures 115 Fig. 4-8: Effect o f DIDS on airway epithelial cell cultures 117 Fig. 4-9: Lactate release and localization 120 Fig. 4-10: Transepithelial transport o f model drugs across AIC and LCC 123 Fig. 4-11 : FITC-dextran transport in airway epithelial cells 128 Fig. 4-12: Size exclusion chrom atography o f FITC-dextran transport sam ples 129 Fig. 4-13: Relationship between FITC-dextran m olecular weight and perm eability 130 Fig: 4-14: Equivalent pore size estim ation 131 Fig: 4-15: Tim e course o f B-adrenergic agent transport 133 Fig: 4-16: R elationship between drug lipophlicity and perm eability 134 Fig. 4-17: U ridine m etabolism 137 Fig. 4-18: N a \ tem perature and direction dependence o f uridine uptake 139 X ll Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig: 4-19: Na^:uridm e coupling stoichiom etry 140 Fig. 4-20: Concentration dependence o f uridine uptake 145 Fig. 4-21 : Dixon Plot: Inhibition o f uridine uptake w ith adenosine 148 Fig. 4-22: N ucleoside analog inhibition o f uridine uptake 151 Fig: 4-23: 2 -D eoxyadenosine and HPMPC uptake 152 Fig. 4-24: N ucleoside transport across airway epithelial cell m onolayers 157 Fig. 4-25: Basolateral uptake o f uridine 159 Fig. 4-26: Basolateral efflux o f uridine 160 Fig. 4-27: Ad5 infection o f airw ay epithelial cells 162 Fig. 4-28: Im m unofluoresecence staining o f Ad5 in airw ay epithelial cells 163 Fig: 4-29: Ad5 antigen release from infected airway epithelial cells to the bathing fluid 164 Fig. 4-30: Ad5 effect o n the bioelectric properties o f airway epithelial cell m onolayers 166 Fig. 4-31 : Ad5 effect on m annitol perm eability 167 Fig: 4-32: Ad5 effect on uridine uptake 168 Fig. 5-1: Schematic representation o f ion transport processes in the airw ay epithelium 188 Fig. 5-2: Schematic representation o f cellular respiration and lactate release 191 Fig. 5-3: C orrelation betw een B-adrenergic agent perm eability in airway and alveolar epithelial cell m onolayers 203 Fig. 5-4: Schematic representation o f overall ion and drug transport processes in airw ay epithelial cells 236 Fig. 6-1 : Schematic representation o f nucleoside transport in airw ay epithelial cell m onolayers 242 Xlll Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF ABBREVIATIONS 2’-dA - 2 ’-deo}Qradenosme 2 -dl - 2’-deoxyinosine 2 -dU - 2 ’-deoxyuridine 2-CA - 2-chloroadenosm e 3 ’-dA - 3’-deoxyadenosine 3TC - thiacytidine 5’-M T - S’-m ethylthioadenosine 5-FU - S-fluoronracil 5-FUrd - S-fluorouridine 6-M OR - 6 m ercaptopurine A549 - lung carcinom a cell line AB - apical to basolateral direction ACV - acyclovir Ad5 - adenovirus type 5 AIC - air-interfaced culture AIDS - acquired im m une-deficiency syndrom e Ara-A - adenosine arabinoside Ara-C - cytosine A rabinoside AZA - 8-azidoadenosine AZT - 3 -azidothym idine BA - basolateral to apical direction BBM V - brush-border membrane vesicles BRS - bicarbonated Ringer's solution BSA - bovine serum album in Caco-2 - colon carcinom a cell line CMV - cytom egalovirus COPD - congestive obstructuve pulm onary disease CPE - cytopathic effects d4T - stavudine ddA - dideoxyadenosine ddC - dideoxycytosine ddG - didexyguanosine ddl - dideoxyinosine ddT-dideoxythym idine dFdC - gem citabine DIDS - 4,4'-diisothiocyanotostilbene-2,2'- disulfonic acid DM E - Dulbecco's m odified essential m edium ECM - extracellular m atrix EDTA - ethylenediam ine tetraacetic acid ei - equilibrative NBTI-insensitive transporter EIA - enzym e immunoassay es - equilibrative NBTI-sensitive transporter F I2 - Ham 's F-12 nutrient m ixture FBS - fetal bovine serum FD - FITC-labeled dextrans FDIO - FITC-dextran 10,000 daltons FD20 - FITC-dextran 20,000 daltons FD4 - FITC-dextran 4,000 daltons FD40 - FITC-dextran 40,000 daltons FD70 - FITC-dextran 70,000 daltons FITC - fiuoresceine isothiocyanate GCV - ganciclovir HBSS - Hank's balanced salt solution HTV - hum an immunodeficiency virus HPLC - high performance liquid chrom atography HRP - horseradish peroxidase HSV - herpes simplex virus H U rd - 5-hydroxyuridine IDU - 5-iodo-2’-deoxyuridine lEC - intestinal epithelial cell line Isc - short-circuit current lU rd - 5-iodouridine Ki - inhibition constant Km - M ichaelis-M enton constant LCC- liquid-covered cultures L LC -PK l - kidney epithelial cell line Log P - log octanoI/pH 7.4 buffer partition coefficient M CT - m onocarboxylic acid transporter M DCK - M adin Darby canine kidney cell line M I - m ock-infected M OI - m ultiplicity o f infection M W - m olecular weight XIV Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. N I - Na-nucleoside transporter (purine selective) N2 - N a-nucleoside transporter (pyrim idine selective) N3 - N a-nucleoside transporter (broad selective) N4 - Na-nucleoside transporter (broad selective) NBTI - nitroben^lthioinosine NPAA - N -phenylanthranilic acid Papp - apparent perm eability coefBcient PAS - periodic acid schiff stain PBS - posphate buffered saline PD - potential difference pfu - plaque form ing units PI - post-infection PKA - protein kinase A PKC - protein kinase C PM A - phobol m yristyl acetate RBC - red blood cells RSV - respiratory syncytial virus R t - transepithelial electrical resistance SFR - sodium -free Ringer’ s solution S-M EM - Eagles minimum essential medium (suspension culture) TLC - thin layer chrom atography Vmax - maximum velocity XV 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. 1. GENERAL INTRODUCTION Recent advances in biotechnology and pharm aceutical research have led to the developm ent o f genetically engineered peptides, proteins, genes, and the discovery o f new chemical entities w ith im proved pharm acological and pharm aceutical properties. Chemical and enzym atic instability o f drugs w hen adm inistered by the oral o r parenteral routes has resulted in the em ergence o f alternate strategies in drug delivery such as inhalation delivery. Inhalation route for drug delivery offers: (1) good patient com pliance due to its convenience and noninvasive nature, (2) im proved enzym atic stability o f drugs compared to the intestine (M a e t a i, 1996; L itterst et a i, 1975), (3) the avoidance o f hepatic first-pass m etabolism , (4) excellent bioavailability - (alm ost 100%) based on the fraction o f drug deposited to the airways (A djei and Garren, 1990), and (5) site-specific local delivery elim inating m ultiorgan toxicity (M ontgom ery et a l, 1987). Features that m ake the respiratory tract attractive as a portal o f drug delivery include: its enorm ous surface area o f ~ 140 m^ for drug absorption, a thin alveolar epithelial perm eability barrier o f <1 pm (Rennard e t al., 1991), and blood supply involving the entire cardiac output (Chediak and W anner, 1990). Drugs delivered locally to the respiratory tract via inhalation m ay be absorbed throughout the conducting airw ay down to the term inal bronchioles and ultim ately the distal alveoli. The prim ary barrier to the transport o f drugs is the surface epithelial Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lining. Therefore, a detailed study o f the barrier properties is essential for a better understanding o f the airw ay drug absorption process. A convenient tool for investigating the airway epithelial structure and function is a model system o f isolated epithelial cells grown in a controlled environm ent that m im ics the native epithelium m orphologically, electrically, biochem ically and in barrier function. Such a model may serve not only in the overall understanding o f epithelial characteristics but play a pivotal role in understanding and dissecting absorption m echanism s, m etabolism , toxicity and delivery o f drugs intended for inhalation. Respiratory pathogens such as viruses firequently affect the upper airway epithelium spreading dow n the respiratory tract in the latter stages o f infection. Local delivery o f antiviral nucleoside analogs to the respiratory tract offers all the advantages mentioned above in the fight against respiratory viral infections (D ebs et al., 1988). In addition, delivery at the onset o f infection would prevent spread o f infection to the alveolar region and other parts o f the body. Antiviral nucleoside analogs are inherently toxic and disrupt a num ber o f cellular processes w ith constant exposure for prolonged tim es (M orris, 1994). Therefore, the two most significant benefits o f targeting nucleoside analogs to the respiratory epithelium w ould be the avoidance o f m ultiorgan toxicity generally associated w ith conventional oral or parenteral routes o f adm inistration, and the selective delivery to the target site o f action. Unfortunately, alternate routes o f antiviral drug delivery have not been explored and consequently, very little is known about nucleoside analog absorption, m etabolism and delivery in the airway epithelium (Debs e t a l, 1988; Knight and G ilbert, 1988). H ence, this project Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. involved the developm ent o f an airway epithelial cell culture model for understanding the m echanism o f nucleoside and antiviral nucleoside analog transport in the airways. This research proposes a strategy to facilitate the delivery o f these im perm eable nucleoside analogs to epithelial cells. 2. THE AIRWAY EPITHELIUM 2.1. Biology o f the Airways The airways o f the lung provide a pathw ay for the bulk flow o f inhaled air in and out o f the peripheral alveoli. The overall airw ays are divided into 2 general com partm ents, as depicted in Fig. 1-1: the conducting zone and the respiratory zone. The conducting zone comprises o f the first 16 generation o f the airways from the trachea (generation 0) that bifurcates to the m ain stem bronchi (generation 1), that further subdivide into bronchi that enter tw o left and three right lung lobes. The conducting zone ends w ith the term inal bronchioles. Thus, is involved in the bulk m ovem ent o f air to the respiratory zone. All structures in the respiratory zone com prising o f generation 17 to 23, participate in gas exchange beginning from the respiratory bronchioles and eventually subdividing into the alveolar sacs (A ltiere and Thompson, 1996). The w all o f conducting airw ays consists o f three m ajor components: (1) a mucosa that is com prised o f an airway epithelium and connective tissue lam ina, (2) a smooth m uscle sleeve, and (3) an enveloping connective tissue tube, supported by cartilage rings. The epithelial Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lining th at covers th e entire lum inal area o f the conducting airw ay is continuous but com posed o f nonuniform ceil types that exhibit region selectivity. The epithelial lin ing serves two prim ary functions: protection against inhaled particles and invading pathogens, separating the external lum inal environm ent from the internal environm ent, and fluid and electrolyte balance o f the lum inal m icroenvironm ent m aintaining airw ay hum idity and surface liquid com position and fluidity. The first line o f defense against inhaled particles and pathogens is the m ucus layer supplied constantly by the subm ucosal glands and the secretory cells in the epithelium . This m ucus layer serves to: (a) protect the epithelium from dehydration, and (b) provide a protective barrier against inhaled particles, m icrobes and viruses by physical entrapm ent and m etabolic inactivation (M arriott, 1990). The second line o f defense is the specialized tight junctions that restricts the penetration o f inhaled pathogens and particles through the intercellular regions to the underlying tissue. The epithelium consists o f three m ajor cell types: ciliated, secretory, and basal cells in a pseudostratifred colum nar epithelial arrangem ent (Harkema e t a l, 1991). C iliated cell com prise about 43% o f the total cell population (Plopper et al., 1983). They are tall colum nar cells w ith cilia o f the apical surface. A bout 19% o f the cells are secretory in nature (C lara and m ucus cells). Their distinguishing feature are the granular appearance o f the cytoplasm . B asal cells are generally sm all spindle shaped cells and com prise o f 28% o f the cell population (Plopper et a l, 1983). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. •aefiM broncftr tmnctiiolts \ lamtinal bmnchidtas nupintoiy brwetüol^ aivmolarducts atv90lërsacs Fig. 1-1. M odel o f airw ay structure. Conducting (trachea to term inal bronchioles) and respiratory (respiratory bronchioles to alveolar sacs) zones o f the airw ays. From (A ltiere and Thompson, 1996). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.2. Solute and Drug Transport in die Airway: Paucity o f information A lthough the respiratory epithelial ion transport processes and its relation to diseased conditions such as cystic fibrosis and asthma have been the subject o f extensive study (W iddicombe, 1986; Knowles et a i, 1991), the same caimot be said for epithelial solute and drug transport processes. Solute transport process play a very im portant role in nutrient and fluid absorption (Kim and Crandall, 1995). Physiologically they may function in the regulation o f epithelial fluid lining, in the salvage o f essential nutrients, or in the excretion o f endogenous m etabolites (Joris and Quinton, 1989; Kim and Crandall, 1995). Evidence for solute transport processes in the conducting airway include: Na^- glucose cotransporter (Joris and (Quinton, 1989; Basset et a i, 1988), the organic anion transporter (C loutier and Lesniak, 1985; Enna and Schanker, 1973), and the dipeptide transporter (Yam ashita e t al., 1996). It has been postulated that the depth o f the aqueous fluid lining the airways m ay be m aintained in part by solute transport processes with an inward Na^ gradient such as the Na^-glucose cotransport (Joris and Quinton, 1989; Basset et a i, 1988). In other epithelia, oftentim es these solute transport processes may be involved in the transport o f therapeutically active drugs that are structurally sim ilar to the endogenous o r nutrient substrates. A case in point is L-a-m ethyldopa, an amino acid analog; stavudine (d4T), a nucleoside analog; p-am ino hippurate an organic anion; or procainam ide, an organic cation. Taking lessons from the intestinal epithelia (o f the same em bryonic origin), targeting drugs to the respiratory solute transporters on the luminal cell Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. membrane would facilitate the delivery o f these generally impermeable hydrophilic drugs to the cell interior either by 6 cilitated difhision or ion-coupled (Na^ or K*) concentrative carrier-m ediated mechanisms. The reasons for the p a u c i^ o f drug transport studies in the airway epithelium is not known but m ay include: (1) the lack o f adequate physiological information on biosynthetic pathways, nutrient salvage pathways, and other metabolic pathways, (2) the lack o f recognition o f the airways as a potential site for drug absorption, in comparison to the alveolar region, (3) the presum ption that airway epithelial drug absorption may contribute very little to the overall pulm onary absorption o f inhaled drugs, and (4) the lack o f convenient and relevant m odels for detailed m echanistic studies o f drug absorption, metabolism and toxicity in the airway epithelium. There are no known reports o f solute or drug transport via specialized membrane- protein carriers in cultured airw ay epithelial cells. Therefore, this study is the first to apply an airway epithelial cell culture model to characterize a nutrient solute transport process in the airway lumen. Moreover, the utilization o f these solute transporters to deliver normally impermeable exogenous nutrient-like drugs to cells was demonstrated. Recently, prelim inary evidence for a dipeptide transporter was reported in air-interfaced rabbit tracheal epithelial cells based on flux and electrophysiological studies (Yamashita et a l, 1996). Two studies using excised tissue have reported evidence for the Na -glucose transporter (Joris and (Quinton, 1989; Basset e t a i, 1988), and others demonstrated evidence for the organic anion transporter (Enna and Schanker, 1973; Cloutier and Lesniak, 1985). In equine tracheal strips, — 20% o f total Isc was attributed to an Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. electrogenic Na^-coupled glucose transporter (Joris and Quinton, 1989). In the rat airways, Enna and Schanker (Enna and Schanker, 1973) demonstrated that the absorption o f the organic anion phenol red 6om the airway lumen is saturable and inhibited by other organic anions such as benzylpenicUlin, sulfonic acid dyes, and cephalothin. In the canine trachea, under open circuit conditions, net PAH absorption was found w ith a Km o f 754 pM (Cloutier and Lesniak, 1985). Besides these, no other active solute transport processes known to be present in other epithelia such as the nucleoside transporters, the nucieobase transporters, the am ino acid transporters, and the phosphate transporter have been reported in the airway epithelium . In the intestinal and renal brush-border, equilibrative (facilitated diffusion) and concentrative nucleoside transporters have been identified to mediate the rapid and highly selective influx o f nucleosides to the epithelial cell interior. Moreover, selective antiviral nucleoside analogs were reported to interact with these transporters resulting in high intracellular concentrations. Evidence for nucleoside analog interaction w ith nucleoside transporters has been reported in rat alveolar macrophages (Plagemann, 1991), erythrocytes (M ahony et a i, 1991; Dom in et a i, 1993), choroid plexus (Huang et a i, 1994), intestinal (G utierrez and Giacom ini, 1993) and kidney epithelium (B rett et a i, 1993). By targeting nucleoside drugs to the nucleoside transporter absorption was significantly im proved leading to higher bioavailabili^ and enhanced potency o f antiviral drug (W aclawski and Sinko, 1996). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.3. Need for Local Delivery A facet o f inhalation delivery th at presents trem endous potential yet to be exploited, is the local delivery to the epithelial lining o r the underlying tissue in the lung. Areas w here this approach has met w ith great success are in the treatm ent o f hyperactive o r obstructive airw ay diseases such as asthm a, congestive obstructive pulm onary disease (COPD), cystic fibrosis, and infectious diseases o f the respiratory tract evoked by bacteria (especially gram -negative bacteria). Research has focused on understanding the etiology o f these diseased states, the nature o f the dysfunction, and the identification o f the receptors, carriers o r other m em brane proteins im plicated in the disease. In turn, this has translated to the developm ent o f drugs (substrates, agonists or antagonists) delivered by aerosol that correct the dysfunction, alleviating the disease sym ptom s and possibly the cause (Ilow ite and Niederm an, 1990; Seale, 1990). Despite being a w idespread affliction in healthy and imm unocom prom ised patients, the treatm ent o f viral infections o f the respiratory tract has received little attention. A lthough virus in fectiv i^ and replication has been studied w ith various species (rotavirus, influenza, adenovirus, respiratory syncytial virus) in airw ay epithelial cells (W inther et a l, 1990; R eiss e t a l, 1991; C loutier et a l, 1989), interest has m ostly centered around studying virus-host cell interactions (D efer et a l, 1990), assessing the virus-evoked alteration o f epithelial structure and function (C loutier e t a l, 1989; D efer e t a l, 1990; R eiss e t a l, 1991). E xcept for two studies that investigate the feasibility^ o f aerosol treatm ent (w ith ganciclovir, foscam et, and ribavirin) o f respiratory epithelium 10 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (D ebs et al., 1988; G ilbert et al., 1992), very little is know n about antiviral drug delivery to the airway epithelium . Consequently, know ledge about: the feasibility o f local delivery to epithelial cells, antiviral drugs efGcacy and potency in the airway epithelial cells, sensitivity and toxicity o f nucleoside analogs to the epithelial cells, and th e targeting o f nucleoside drugs to specific sites on the cells surface rem ain to be explored. Such inform ation would lead to the developm ent o f strategies to facilitate delivery o f nucleoside analogs to epithelial cells thereby optim izing therapeutic benefit w hile reducing undesirable nonspecific toxic effects inherent w ith antiviral drugs. Treatm ent o f respiratory infection via oral o r parenteral routes often result in h ig h risk o f toxic effects. It has been dem onstrate that certain respiratory infections do n o t respond well to system ic treatm ent (M axwell, 1983). There are a num ber o f possible explanations, albeit, poor biodistribution in the lung accounts for a large degree o f the failure (Stout and D erendorf, 1987). It is generally accepted that in order to achieve adequate antiviral inhibition, the concentrations a t the site o f infection should exceed th at required to inhibit virus growth. Since infection generally occur by inhalation at the surface epithelium , it m ay be directly accessible to locally delivered drugs. This may provide a greater therapeutic advantage over system ic delivery while reducing o f the extrapulm onary and extra-airw ay toxicity associated w ith w hole body loading. Using aerosol antibiotics as a guide, it is generally assum ed that adequate concentrations o f the drug can be achieved at the airw ay epithelial surface in the airway secretions in order to treat respiratory infections. For antibiotics such as pentam idine, gentam ycin, and tobram ycin (all currently under clinical trials) (Ilowite and Niederm an, 1 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1990), intratracheal or aerosol adm inistration yielded high surface liquid concentrations that resulted in low er incidence o f pneiunonia or a decrease in colonization rate compared to control groups (M ontgomery et al., 1987). A erosol delivery produced higher surface liquid concentrations than systemic delivery, w hile no serum accum ulation w as observed for respiratory delivered drugs (Stout and Derendorf, 1987; Ilow ite and N iederm an, 1990). For virus infections o f the respiratory tract ribavirin (trade nam e V irazole, ICN Pharm aceuticals) a nucleoside analog is the only approved drug for the treatm ent o f respiratory syncytial virus (RSV) in infants. For adults, at this tim e no m edication is available for the local treatm ent o f any respiratory viral infection. 2.4. Models for Respiratory Drug Delivery The m odels used to evaluate drug absorption and m etabolism can be grouped into two: in vivo m odels involving the whole anim al, and in vitro models involving isolated tissue and epithelial cells. 2.4.1. In vivo Models: Developed by E nna and Schanker (Enna and Schanker, 1972) the in situ rat lung model was used for the evaluation o f drug absorption and m etabolism . Following intratracheal instillation the appearance o f drug and m etabolites in the blood was estim ated (Schanker, 1978). D elivery o f aerosolized drugs utilizing nebulizers or dry-powder generators have been used to directly deliver drugs to the airw ays o f anesthetized 12 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. anim als or in exposure cham bers (plythesm ographic cham bers) (Raeburn et al., 1992). The advantage o f in vivo studies is that it explores the feasibility o f respiratory drug delivery and provides an overall assessm ent o f drug absorption in the lung. In addition, this technique provides useful inferm ation on drug deposition, distribution, and lung function. How ever, these m odels require expensive equipm ent and furnish no inform ation on the m echanistic aspects o f drug absorption, due to their com plexities and m ultiple variables (such as the presence o f non-epithelial cell ^rpes and basem ent mem brane, and com plications w ith the presence o f preexisting inflam m ation or infection conditions, and lim ited viability). 2.3.2. In vitro Models: (a) Isolated perfused lung models: Lungs isolated m odels from rats, guinea pigs, or rabbits have been com m only used. The lung can either be directly perfused w ith the drug solution and be recycled continuously or blood concentrations can be estim ated w ith a single-pass perfusion technique (Brazzell et ai., 1982). The advantage o f this m odel in that: it provides a good estim ate o f lung absorption and m etabolism where the lung structure and cellular organization is intact, w hile excluding other organs. The lim itations o f this m odel are that like all excised tissue m odels they have lim ited v ia b ili^ generally for a few hours (2-4 hr), and require skillful surgical technique. Drugs evaluated utilizing this m odel include: cardiac glycosides, antibiotics, organic anions, organic cations and steroids (Enna and Schanker, 1972; Schanker, 1978; L anm anera/., 1973). 13 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (b) Lung slices and tissue hom ogenates Lung slices generally 0.6 - 1 m m are placed in dishes containing dosing solution o f drugs is used to evaluate the extent o f tissue associated drug (Sm ith and W yatt, 1981). This m odel is useful for evaluating the extent o f drug uptake into the lung parenchym a, as w ell as for providing a crude estim ate o f drug stability against lung enzym es. The disadvantage o f these m ethods is that the supply o f drug substrates and the rem oval o f m etabolite products are unnatural. Slicing o f the lung parenchyma results in the release o f intracellular enzym es, cofactors, ions, DN A other intracellular m etabolites that com plicate data interpretation to a large extent. Exam ples o f drugs studied using this model include: putresceine, sperm idine and paraquat (Smith and W yatt, 1981). (c) Isolated cell system s A lveolar m acrophages and epithelial cells isolated by protease (elastase, collagenase or dispase o r bacterial protease) treatm ent have been used to estim ate drug transport (M agnani et a i, 1996; Plagem ann, 1991). Evaluation o f drug transport in isolated cells is perform ed by standard techniques such as rapid filtration where isolated cells are introduced into a tube containing the drug solution, and uptake term inated by rapid filtration w here drug solution is washed o ff cells retained on filter paper (generally done for brush-border m em brane vesicle BBMV) (O liver and Paterson, 1971; Plagem ann, 1991). A nother m ethod com m only used is the oil-stop method, where uptake is term inated by draw ing isolated cells through an oil layer by centrifugation (W ohlheuter 14 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. et al., 1978). These m odels are useful to identify uptake characteristics o f particular cell types. They generally yield a large num ber o f cells that are am enable for experimental m anipulation. The drawbacks w ith isolated cells are the presence o f contaminating cell types (such as RBC’s) that d istort data interpretation. Further disadvantages include: the cleavage o f cell surface proteins o f interest (transporters and receptors) by the protease action during cell isolation, and th e inabilify to ascertain the localization or directionalify o f transport. N ucleoside analogs such as AZT, and dideoxynucleoisdes have been studied in isolated cells (O liver and Paterson, 1971; Plagem ann, 1991) (d) Epithelial cell culture The culturing o f isolated epithelial cells on permeable filter supports has emerged as a convenient tool for the dissection o f drug absorption m echanism and the screening o f perm eability, m etabolism , toxicify and efficacy o f drugs intended for inhalation delivery. C ultured on a physiological substratum , w ith grow th factor supplemented media, epithelial cells form confluent, polarized and w ell differentiated sheet o f cells that m im ics the native epithelia. B y introducing the drug solution on either the apical or basolateral surface, the am ount o f drug taken up or the am ount o f drug transported across the epithelial cell layer can be determ ined. The advantages and disadvantages are discussed in the next session. D rugs evaluated in respiratory epithelial cell cultures include: fi-adrenergic agents, steroids, dextrans, di- and tripeptides, and proteins such as insulin, horseradish peroxidase and album in (M athias et al., 1995; M athias et al., 1996; Yam ashita et al., 1996; M orim oto et al., 1994; M atsukawa et al., 1996). 15 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.4. Airway Epitheliai Cell Cultures The four m ajor culture types o f airway epithelial cells are organ cultures, tissue explants, prim ary cultures, and transform ed cells o r carcinom a cell lines (V an Scott et al., 1991). The earliest airw ay epithelial culture m odels were organ cultures and tissue explants representing the sim plest type o f culture designed for short-term studies such as m ucociliary differentiation (Lane et al., 1976) and phenotypic characterization (Trump et al., 1980), as the morphological characteristics are generally well preserved. Organ cultures involve the m aintenance o f a confluent epithelial layer on an intact lam ina or basem ent m em brane, whereas explant cultures refer to the proliferation o f epithelial cells aw ay from the excised explant onto a supporting substrate. The drawbacks to these system s are that they are take a long tim e to develop, have poor reproducibili^, are often contam inated with the grow th o f non-epithelial cell types, lose differentiation w ith distance from the explant, and often show central necrotic centers (Speirs and C utz, 1993). Prim ary cultures o f respiratory epithelial cells are particularly useful in probing the structure and function o f the epithelium in healthy and diseased states (W iddicombe, 1986; Stutts et al., 1985), and in elucidation o f their drug transport characteristics (M athias et a i, 1995; M athias et al., 1996). The advantages o f airw ay epithelial cell cultures are; (1) they provide a controlled environm ent for the study o f epithelial diseased states such as cystic fibrosis (W iddicombe, 1986), inflam m ation (Paradiso et 16 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a l, 1991; Crom w ell ef a/., 1992), and înfectioa (Reiss era/., 1991; Becker e /a /., 1993) and ion and drug transport phenom ena under these conditions (Stutts et al., 1985); (2) the apical and basolateral m em branes are directly accessible for evaluation o f perm eation pathw ays and drug transport m echanism s (M athias et a l, 1996); (3) the exclusion o f non-epithelial cell types, the basem ent membrane and other com plexities associated w ith tissue, sim plifies d ata analysis and interpretation; (4) they present a rapid and convenient means o f assessing drug perm eability, enzym atic stability, perm eation enhancem ent, and targeting to a particular cellular site; and (5) they have extended viability (up to two days) com pared to other in vitro and in vivo m odels (3-6 hr) under defined experim ental conditions (Jam igan et al., 1983). Som e o f the disadvantages o f culture epithelial cells include: (a) differences in cellular structure and barrier properties com pared to native tissue that m ay im pact drug transport; (b) the lack o f in vrvo-like conditions in drug delivery and m ucociliary clearance; (c) the lack o f physiological conditions (smooth m uscle tone, respiratory circulation, subm ucosal gland secretion, and neurohumoral influence) and diseased states such as inflam m ation, and im m une response that may influence drug absorption, and (d) a lim ited number o f cells th at need to be fieshly isolated from the anim al for each culture procedure. The fourth type o f airway epithelial cell cultures are the im m ortalized cultures transform ed in vitro by oncogenes, carcinogens or viral vectors (Gruenert et al., 1995). These m odels are useful for studying the biochem ical aspects o f disease conditions such as cystic fibrosis (Kunzelm ann et al., 1993; Shen et al, 1994), neoplastic progression 17 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (N ettesheim and B arrett, 1984), inflam m ation (Bigby et al., 1989) and infection (Reiss et a i, 1991). W hile transform ed airw ay epithelial cell cultures provides a continuous source o f epithelial cells due to abundant proliferation, they exhibit dedifferentiated ultrastructural characteristics and lose the ability to establish tight junctions (G ruenert et al., 1995). Thus, m ay not be appropriate for evaluating drug transport in the airw ay epithelium . In the prim ary culturing o f airway epithelial cells, the optim um culture conditions for adequate differentiation revolve around selecting the right substratum , the right m ix o f grow th factors as m edia additives, and creating an air-interface. In the early developm ental studies, airway epithelial cultures from a variety o f species w ere cultured on denuded tracheal grafts (Terzaghi et al., 1978) or fibroblast feeder cells (Lechner et al., 1981), supplem ented w ith serum in the grow th m edium. W ith the availability o f physiologically relevant substrata such as collagen and fibronectin, growth factors (epiderm al grow th factor), hormones (insulin, transferrin and hydrocortisone), and other inducers o f differentiation (cholera toxin and retinoic acid), epithelial cells grew to confluence quicker and exhibited m arked increase in differentiation (studies listed in Table 1-1). This is indicated by the presence o f cilia, secretory granules or keratin-like squam ous differentiation, and m ucin-like glycoprotein secretion. Taking lessons from the culturing o f epiderm al and com eal cells, exposure o f the airw ay epithelial cell m onolayers to an air-interface has been found to vastly im prove the level o f differentiation resulting in a m orphology that closely resem bles that o f the native epithelium . The bioelectric properties are elevated (Robison and Kim, 1994; Johnson et 1 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a i, 1993; Yam aya et al., 1992), and ciliogenesis stim ulated (de long et al., 1994; C lark et al., 1995) w hen cultured on a collagen substratum , in a horm one-supplemented defined m edium , at an air-interface. A sum m ary o f airw ay epithelial cultures procedures: the dissociating enzyme, the substratum for attachm ent, the m edia com position and grow th and differentiation factors used and the m ajor focus o f the study is provided in Table 1-1. 19 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 (D T 3 O Q . C S Q . T 3 (D W W O 3 O (D 8 T 3 V C 3. ( Q 3 (D (D (D T 3 O Q . C g . O 3 CD Q . O C T 3 CD (ft C / J O 3 Table 1-1 : Summary of Airway Epithelial Cell Culture Models Cell Type Enzyme Substratum Media Composition Rabbit Tracheal epithelium Tracheal epithelium Tracheal epithelium G u I dm P ig Tracheal epithelium Tracheal epithelium Tracheal epithelium Hamster Tracheal epithelium Tracheal epithelium Study Objective Reference Protease type XIV Tissue culture F 12 + (0.01-10%) PBS, EOF (10 ng/ml) INS Growth and difTerentiation (Wu and Smith, 1 982) treated dishes (5 pg/ml), Tf (5 pg/ml) Protease type XIV collagen Protease type XIV collagen Protease type XIV collagen Protease type XIV collagen Protease type XIV collagen Protease digestion collagen Protease type XIV collagen F-12 + INS ( I Opg/ml), Tf (5 pg/ml), EOF Mucin secretion (12.5 pg/ml), BPE (0.05%), RA (0.1 pM) (Kim, 1985) F-12 + INS (5 pg/ml), Tf (5pg/ml), EOF (25 Electrical properties and (Zietlin et al., 1988) ng/ml), CT (20 ng/ml), HC (0,1 pM), BPE ion transport (15 p^m ) PC-1 Medium PC-1 Medium PC-1 Medium Electrical properties and (Robison er a/., 1993) differentiation Mucin secretion Release of reactive oxidant species (Adler era/., 1990) (Kinnula era/., 1992) F-12 + INS (5 pg/ml), Tf (5 pg/ml), EOF (25 Morphology, (Whitcutt et al, 1988) ng/ml), HC (I pM), CT (40 pg/m), BPE (15 differentiation and mucin pg/ml), RA (0.1 pM) secretion DME + 2% FBS or INS (10 pg/ml), Tf (5 Growth and differentiation (Lee era/., 1984) pg/ml), HC (I pM), BPE (0.4%), EGF (12.5 ng/ml)) 7 3 (D T 3 O Q . C S Q . T 3 (D W W O 3 O (D 8 T 3 V C 3. ( Q 3 (D Table 1-1 (continued) Canine Tracheal epithelium Tracheal epithelium Collagenase collagen Protease type XIV collagen DME:F-I2 + FBS(5%) Electrical properties (Coleman era/., 1984) F-12 + FBS (3%) or serum free INS (10 Growth and difTerentiation (Van Scott era/., 1988) pg/ml), Tf (5 pg/ml), ECGF (7.5 pg/ml),EGF (25 ng/ml), CT (0.01 pg/ml), HC (0.36 pg/ml), T3 (20 ng/ml) CD CD T 3 O Q . c S - . o 3 Bronchial epithelium Bat Tracheal epithelium Protease collagen Protease type XIV collagen F-12 + INS (10 pg/ml), Tf (5 pg/ml), EGF Electrical properties, (Johnson er a/., 1993) (25 ng/ml), HC (I pM), CT (10 ng/ml), 13 30 oxygen consumption and nM) lactate release DME:F-I2 + INS (10 pg ml), Tf (5 pg/ml), Ciliogenesis in an air- (Clark er a/., 1995) EGF (25 ng/ml), HC (0.1 pg/ml), CT (0.1 interface pg/ml), PEA (50 pM) CD Q . T 3 CD C / J C / J o' 3 B ffix io a Tracheal epithelium Human Tracheal epithelium Tracheal epithelium Protease type XIV collagen Collagenase collagen Protease type XIV collagen H2I:F-I2 + FBS or INS (10 pg/ml), Tf (5 Growth and differentiation (Kondo era/., 1993) pg/ml), EGF (25 ng/ml), HC (0.36 pg/ml), CT (0.01 pg/ml), T3 (20 ng/ml) DME;F-I2 + FBS(5%) DME:F-I2 + FBS (5%) or serum-free INS (10 pg/ml), Tf (5 pg/ml), EGF (25 ng/ml), ECGF (7.5 pg/ml), HC (0.4 pg/ml), T3 (20 ng/ml) Electrical properties and (Widdicombe er a/., differentiation 1985) Electrical properties and (Yamaya era/., 1992) differentiation N ) 7 3 (D T 3 O Q . C S Q . T 3 (D W W O 3 O (D 8 3. ( Q 3 (D Table 1-1 (continued) Tracheobronch Protease type XIV collagen ial epithelium Bronchial Trypsin • EDTA 3T3-Feeder epithelium layer DME;F-12 + growth factors (not mentioned) Growth and (Wu et al„ 1990) differentiation and mucin secretion DME:F-12 + FBS(5%) Growth and differentiation (de Jong era/,, 1993) c 3. 3" CD CD T 3 O Q . C S - . o 3 T 3 O Abbreviations; BPE - bovine pituitary extract; CT - Cholera toxin; DME - Dulbecco’s minimum essential medium; EGF • Epidermal growth factor; ECGF • Endothelium cell growth factor; F-12 • Ham’s nutrient mixture; FBS - Fetal bovine serum; HC - Hydrocortisone; INS - insulin, PEA - Phosphaethanolamine; RA - Retinoic acid; Tf - Transferrin, T3 - Triiodotyrosine. CD Q . O c " D ( D (/) o' D 3. EPITHELIAL BARRIERS TO DRUG ABSORPTION 3.1. Penetration Barrier 3.1.1. Mucociliary Escaiator The first barrier encountered by an inhaled drug is the m ucus layer. It is composed o f a com plex m ixture o f mucus glycoproteins (3-4% ), im m unoglobulins, and numerous other substances in 90% w ater (A ltiere and Thom pson, 1996; M arriott, 1990). Biphasic in nature, it consists o f a low er sol-layer o f low viscosity in contact w ith the ciliated and secretory cell surface epithelium . The upper gel-layer is m ore viscous trapping inhaled particles, bacteria, viruses and drugs. The building blocks o f m ucin is a 500 kO glycopeptide that cross-link to form branched glycoproteins (M arriott, 1990). C ontrolled by the ciliary beat firequency, the m ucus layer is gradually moved upw ard at a rate o f about 5 mm/min from the distal airw ays to the pharynx w here it is sw allowed o r expectorated. In diseased states such as cystic fibrosis, infection and inflamm ation, the m ucociliary escalator function is altered due to hypersecretion o f thick viscous m ucus, o r ciliary dysfunction, or a change in fluid and electrolyte secretion, or airw ay constriction (Jiang et al., 1993). U nder these conditions the delivery o f drugs can be affected reducing the therapeutic efGcacy o f inhaled drugs. In general, the m ucus layer (and unstirred w ater layer) does not offer a com plete barrier fo r sm all molecule transport. Clearly, this would play a larger role for lipophilic drugs th at do not freely partition into this layer. However, its im pact on the perm eability 23 S Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o f hydrophilic drugs such as nucleoside analogs is considerably dim inished. For instance, uridine fluxes were n o t found to be aflected by the unstirred w ater layers (Cabantchick and Ginsburg, 1977). This was concluded flom an infinite cis uridine influx experiment w here the value for unstirred layer effects were not significantly different from zero (infinite cis is the condition when the substrate concentration at the cis side o f the m em brane is high w hile that at the trans side is varied; cis denotes the extracellular side w hile trans denotes the intracellular compartm ent). In cultured rat alveolar epithelial cells and C aco-2 cell monolayers A dson et a i, (Adson et al., 1994) also reached a sim ilar conclusion. U sing m ethylamine as a model drug the perm eability coefficient o f the unstirred aqueous boundary layer w as 1.1x10*^ cm /s in unstirred conditions and 3.2 x 10*^ cm /s under stirred conditions. Thus, the unstirred w ater layer in Transwell culture system accounted for 2-6% o f the overall perm eability. 3.1.2. The Apical Membrane O ther than cilia, the apical m em brane has numerous m icrovilli and a dense glycocalyx (Harkem a et a l. 1991). The apical m em brane o f ciliated cells in the airw ay epithelium is 80 nm thick phospholipid bilayer. The rigidity o f the cell membrane is m aintained by cholesterol, saturated and unsaturated fatty acids. A lthough the apical membrane serves as a barrier for solutes to reach the cytoplasm , lipophilic drugs can diffiise through these regions, while hydrophilic solutes cannot fieely diffiise across the membrane. Specialized membrane bound proteins are involved in the translocation o f the hydrophilic drugs to the cytoplasm ic region. Examples o f such transporters in the 24 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. respiratory epithelium include the Na -glucose transporter (Joris and Quinton, 1989; Basset et aL, 1988), the organic anion transporter, (Enna and Schanker, 1973; C loutier and Lesniak, 1985), the dipeptide transporter (Yam ashita et aL, 1996). 3.1.3. The Tight Junction The airw ay epithelial cells are connected to each other to form a continuous barrier lining the airways (W elsh, 1987). They encircle the perim eter o f cells ju st beneath the apical surface separating the apical 6om the basolateral membrane, thereby providing polarity for the vectorial m ovem ent o f ions and solutes. Freeze-firacture m icroscopic studies reveal that these junctions are m ulti-stranded (Inoue and Hogg, 1977) and not homogenous throughout the airways. The tig h t junctions connecting ciliated cells are hom ogenous com prised o f 6-8 interconnected strands, whereas those connecting ciliated and secretory strands are few er (Schneeberger, 1980). This may explain differences in the tight junction barrier properties betw een different regions in the respiratory tract and species. In the distal airw ays the number o f intercellular fibril strands decreases. The three m ost com m on techniques to estim ate the barrier properties offered by the tight junctions include: freeze fracture m icroscopy, transepithelial electrical resistance, and hydrophilic solute perm eabili^. Electron m icroscopy provides a direct visualization o f tight junctions, electrical resistance (Rt) serves as an index for resistance to the paracellular flow o f ions, and solute flux measurements serves as an index o f the functional restriction to intercellular solute flow based on solute size, shape and charge (G onzalez-M ariscal, 1992). M ilton and Knutson (M ilton and K nutson, 1990) 25 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. i dem onstrated that thin section m icroscopy provides a direct visualization o f junctional structure, w hile R t and solute perm eability is necessitated for evaluation o f tight junctional function.. T he paracellular transport o f hydrophilic drugs across the airw ay epithelium decreases w ith m olecular size (Berg et aL, 1989; W angensteen et aL, 1993; Schneeberger, 1980). It has been speculated th at specific regions w ithin the tight junctions term ed as pores m ay account fo r this transport phenom enon (W angensteen et aL, 1993; M atsukaw a et aL, 1997). These structures, originally derived from studies characterizing the barrier properties o f artificial membranes are hypothetical. N evertheless, they help explain the restriction to the transport o f large hydrophilic drugs across the airw ay epithelium . R at and guinea pig tracheas were reported to have equivalent pores o f 7-8 nm in w idth (W angensteen et aL, 1993). Several equivalent pore models have been suggested (W angensteen et aL, 1993; Adson et aL, 1994; M an et aL, 1986) but in general it appears that tw o m ajor pore populations m ay be present (Taylor, 1990; K im and Crandall, 1983): a sm all pore w ith radii 0.5-1 nm th at m ediates the passage o f ions and sm all electrolyte solutes, and a large pore w ith a radii o f 2-25 nm that m ediates the transport o f drugs (sm all drug solute and m acrom olecules). The pore size varies considerably in different species (M an et aL, 1986; W angensteen et aL, 1993; A dson e ra/., 1994). 26 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3.2. Enzymatic Barrier M etabolizing enzym es in the airways function to detoxify inhaled particles and pathogens. A lm ost all m etabolizing enzym es found in the liver are found in the respiratory tract, but to a lesser extent (M a et aL, 1996). There is lim ited inform ation on m etabolizing enzym es in the lung and even less in the trachea. In general, cytochrom e P450 emqrmes have been the m ost studied o f all en^rm e system s in the respiratory tract. They w ork tow ards converting lipophilic drugs into more polar w ater soluble m etabolites. Phase I enzymes m odify reactive functional groups by oxidation, reduction, and hydrolysis, whereas phase H enzym es conjugate phase I m etabolites with endogenous m olecules to yield highly polar excretable conjugates. Flavin m onooxygenase enzymes are absent in the larynx and trachea, but present in the carina, the conducting airway, and pulm onary region (Sabourin et aL. 1988). In the bronchiolar region, C lara cells have high concentrations o f cytochrom e P450 en2ym es including NA DPH -cytochrom e P-450, flavin m onooxygenases, glutathione S-transferases, U D P- glucoronyl transferase, and epoxide hydrolase (Jones et aL, 1983; Ma et aL, 1996). 4. RESPIRATORY VIRAL INFECTIONS The airway epithelium is susceptible to viral infections from adenoviruses, influenza viruses, picom aviruses, herpes viruses and cytom egaloviruses (Tucker and Compans, 1992), being the first line to defense against inhaled pathogens. In the present study. 27 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. adenovirus was selected as it is an ubiquitous opportunistic virus that infects the respiratory, gastrointestinal and ocular epithelium. Adenoviral infections in im m unocom prom ised patients are associated w ith 60% fetalMes in patients w ith respiratory infections. Such infections are difGcult to treat, tend to be long term, add to the debilitation o f the patient and sometimes rapidly overwhelm them resulting in death (Hierholzer, 1992). In normal patients these viruses along w ith rhinovirus and other co- infecting microbes are the m ain causative agents for common cold and throat infections. 4.1. Strategies to counter viral infections D epending on the stage o f their intervention with the v iral replication cycle, antiviral drugs could be divided into the follow ing groups (D e C lercq, 1995): (i) adsorption inhibitors (i.e., CD4 constructs, polysulfates, polysulfonates, polycarboxylates, and polyoxom etalates), (ii) fusion inhibitors (i.e., plant lectins, succinylated o r aconitylated album ins, and betulinic acid derivatives), (iii) uncoating inhibitors (i.e., bicyclam s), (iv) reverse transcription inhibitors acting either com petitively w ith the substrate binding site (i.e., AZT, ddl, ddC , ddA, d4T and acyclic nucleoside phosphonates such as HPMPC) or allosterically w ith a nonsubstrate binding site (i.e., non-nucleoside reverse transcriptase inhibitors), (v) integration inhibitors, (vi) DNA replication inhibitors, (vii) transcription inhibitors (i.e., antisense oligodeoxynucleotides and Tat antagonists), (viii) translation inhibitors (i.e., antisense oligodeoxynucleotides and ribozym es), (ix) m aturation inhibitors (i.e., protease 28 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. inhibitors, m yristoylation inhibitors, and giycosylation inhibitors), and finally, (x) budding (assem bly/release) inhibitors. One o f the m ost well studied and effective antiviral agents are those that inhibit reverse transcriptase o r viral DNA polym erase. These agents m ainly consist o f nucleoside analogs, that get phosphorylated by host cellular or viral kinases to their 5'-triphosphate species, thereby causing chain term ination when incorporated into the viral DNA, or are inhibitors o f reverse transcriptase o r viral DNA polymerase at concentrations about 50- 200 tim es lower than that for cellular en^rm es (De Clercq, 1995). These agents are the m ost efficacious in the treatm ent o f AIDS and in combination w ith protease inhibitors have em erged as an effective thersq>y. 4.2.Current Respiratory Antiviral Treatment Treatm ent o f adenoviral infections is alm ost nonexistent and is lim ited to symptomatic and supportive care. To date, for respiratory in&ctions via inhalation Ribavirin is the only US Food and Drug Adm inistration approved drug ribavirin for the treatment o f respiratory syncytial virus in infants. Since the 1970's nucleoside analogs have been firmly established as effective and potent antiviral agents. The major drawback to their use with high doses o f oral or intravenously adm inistered drug was the m anifestation o f m ultiple organ toxicity (myelosuppression, neutropenia, leukopenia, gastric mucosal dystrophy) (M orris, 1994), due to loading o f the plasm a com partm ent To circumvent this disadvantage o f nucleoside drugs direct specific local delivery to the site o f infection was considered. The key advantages o f this approach are that: (1) it w ould reduce the overall 29 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dose being delivered as the entire central compartment and other unin& cted peripheral compartments would be spared fix>m drug loading; (2) systemic and extr^ulm onary to x ici^ would be elim inated or m inim ized to a great extent; (3) hepatic m etabolism generally associated w ith oral delivery would be bypassed; (4) by targeting the site o f infection, the presence o f antiviral drug would prevent the spread o f virus. The therapeutic index o f aerosolized ribavirin against RSV pneum onia is enhanced by aerosol adm inistration com pared to oral and parenteral treatm ent (H all et al., 1983). A erosolized treatm ent was reported to localize the distribution and the activity o f the drug. The activity o f aerosolized ganciclovir and phosphonoform ic acid in a m urine cytom egalovirus m odel was specific for the lung, w ith the level o f antiviral a c tiv i^ being equivalent to system ic drug adm inistration (Debs et al., 1988). Furtherm ore, aerosolized adm inistration appeared to have little affect on virus replication in other organs. A erosol adm inistration o f pentam idine in patients w ith AIDS and Pneumocytis carinii pneum onia was shown to be highly effective therapy w ith a m arked decrease in system ic drug-related toxicity when com pared w ith parenteral adm inistration (M ontgom ery et al., 1987). Thus, antiviral aerosol delivery can be specific, confined to the target location thereby improving the therapeutic index in the treatm ent o f respiratory infections. 4.3. Delivery Strategies Although the antiviral activity o f a num ber o f nucleoside analogs has been successfully tested, low bioavailability due to rapid metabolism and poor membrane perm eability, and 30 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. severe toxic side effects are som e o f the problem s associated w ith nucleoside drug delivery (De Clercq, 1995). AZT shows m oderate but highly variable bioavailability o f -63% due to extensive hepatic m etabolism (Y aichoan et al., 1990). The bioavailability o f dideoxyinosine (ddl) ranges from 16-54% due to low intrinsic perm eability (Sinko et ai, 1994) and acid instability in the gut (W aclawski and Sinko, 1996). Likewise, acyclovir (-22% ) and cidofovir (<5% ) have low bioavailability due to its low intestinal permeability (Drusano er a/., 1992; W achsman eta /., 1996). Strategies adopted to circum vent these inherent difSculties w ith nucleoside analogs w ork towards im proving the delivery o f antiviral nucleoside analogs to target cells, prolong their retention w ithin cells, o r enhance their antiviral activities by enhancing their activation to active drug. 4.3.1. Prodrug Approach Increasing the lipophilicity o f parent nucleoside drugs enhances the ability o f the drug to partition into the phospholipid m em branes o f cells. Ester prodrugs o f 2’,3’- didehydro-3 '-deoxythym idine (d4T), w ere w ell correlated w ith lipophilicity (r^ = 0.85) w herein the log partition coefficient range w as -1.5 to 2.9. These prodrugs increased the retention o f d4T in the body and reduced concentrations in the plasm a due to distribution in erythrocytes and fatty tissues (H asegaw a et al., 1993). Likewise, 2 -0 - valerate prodrug o f 6-m ethoxypurine (Prus et a l, 1990), ether, carbonates and urethane prodrugs o f 3 '-deoxythym idine, 3’-azidothym idine, ddl, ddC and ddA (Hammer et a l. 31 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1996), and adam antane ester prodrugs o f AZT (Tsuzuki et aL, 1994) w ere all shown to have im proved mem brane perm eation properties compared to parent drugs. 4.3.2. Drug Delivery Systems Drug carrier system s have been used to overcome some o f the absorption and pharmacokinetic problems to enhance the activities o f drugs for the treatm ent o f viral infections. Nanoparticles o f polyhexylcyanoacrylate or human serum album in loaded with AZT and ddC targeted to cells o f the mononuclear phagocytic system (Bender et al., 1994) accum ulated in monocytes and macrophages, exhibiting good antiviral activity compared to a nearly inactive aqueous solution o f antiviral drugs (Bender et aL, 1994). Antiviral nucleoside drugs coupled to lactosaminated album in fo r targeting to the liver cells for the treatm ent o f hepatitis infections has been dem onstrated to improve delivery and antiviral efficacy. These conjugates are currently under clinical trials as intravenous injections (Fium e et aL, 1993). Selective targeting to the liver was achieved w ith chylomicrons as carriers for nucleoside analogs (iododeoxyuridine), leading to an effective intracellular concentration o f 700 pM J e n s e n et aL. 1995). In case o f delivery to tissues other than liver, the rq>id clearance o f conventional liposom es and m icorspheres by the liver can be considerably decreased by incorporation o f polyethylene glycol- phosphatidylethanolamine (PEG-PE) into the delivery system thus prolonging antiviral action (Kamps er a/., 1996). 32 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4.3.3. Carrier-mediated Transport An alternate strategy to enhance nucleoside drug perm eation in the cell is via the nucleoside transporters. These transporters serve as efficient means o f internalizing hydrophilic nucleoside nutrients normally inaccessible to the cell due to their inability to permeate the cell membrane. Physiologically, nucleoside transporters are involved in the conservation o f cellular nucleic acids as an efficient salvage m echanism for nucleosides secreted to the surface liquid (Plagem ann et aL, 1988), and pharm acologically, they may play an im portant role in accum ulating nucleoside analogs in the target epithelial cells (Jarvis, 1989; B rett c /a /., 1993; W u eta/., 1994; Plagem ann, 1991; G utierrez and Giacom ini, 1993). In oral delivery, high intrinsic intestinal perm eability o f Stavudine (d4T) and 2’,3 '- dideoxycytidine (ddC) led to high bioavailability ranging fix>m 70-90% (Yarchoan et al., 1990; W aclawski and Sinko, 1996). In a single pass intestinal perfusion experiment and intestinal brush-border membrane vesicles, W aclawski and Sinko (W aclawski and Sinko, 1996) dem onstrated that d4T permeates the rat intestine by apical carrier-mediated transport follow ed by facilitated diffiision via a basolateral transporter, leading to high bioavailability. On the other hand, AZT and acyclovir are not known to be substrates for the nucleoside transporters (Hu, 1993; Gati et al., 1984) that m ay explain their relatively low oral bioavailability (10-20%). A t the cellular level, for interaction w ith the human kidney nucleoside transporter, nucleosides w ith m odification to the ribose ring rendered compounds incapable o f interacting w ith the nucleoside transporter. Substitution o f the 33 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. hydroxy group at the 3 -OH (as in AZT or dideoxynucleosides) abolished interaction. Furthermore, a 2 -OH trans to the 3 -carbon (as in Ara-A and Ara-C) impairs interaction w ith the Na^-nucleoside transporter in the kidney involved in the reabsorption and disposition o f nucleoside analogs (Brett et aL, 1993). W hereas, analogs with modification on the base were effective substrates. Table 1-2 summarizes som e o f the interaction o f nucleoside analogs w ith m am m alian cells. The selectivity^ o f antiviral nucleoside analogs varies betw een different cell types. In general, nucleoside analogs w ith modification on the base such as 2-chloroadenosine, 5-fiuorouridine and 5-iodo-2’-deoxyuridine (Jarvis, 1989; G utierrez and Giacom ini, 1993) appear to interact w ith nucleoside transporters while those w ith m odifications to the ribose ring such as AZT, ddC, and Ara-C show no interaction (Plagem ann, 1991; Fang et al., 1996; Brett et al., 1993). In lymphocytes and m acrophages, m ost o f the dideoxynucleoside analogs (ddA, ddC , ddG, ddT, ddl) cannot difiuse into cells. W hile in erythrocytes, dideoxynucleosides and acyclic nucleosides (ACV, GCV) were found to interact with the nucleobase transporters (Mahony et al., 1991; D om in et al., 1993; Gati et al., 1984). In epithelial cells the nucleobase transporters are not w ell characterized as y e t 4.4. Infection o f Airway Epitheiiai Ceiis 4.4.1. Adenovirus infection Adenovirus infection o f respiratory epithelial cells has not been investigated in m uch detail. In the present study, adenovirus type 5 wild type (A d5) w as used as a model virus to exam ine the virus-induced cellular responses in airway epithelial structure and fimction. 34 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The internalization pathway for adenovirus by cells consists o f the follow ing steps (D efer et aL. 1990): (1) recognition and attachm ent o f virus particles to specific cell surface receptors (Svensson et aL, 1981), (2) clustering o f receptors at the cell surface and their invagination into a coated pit, (3) internalization o f the virus-receptor containing vesicle, (4) dissociation o f the clathrin coat to generate an endocytic vesicle, (5) decrease in intravesicular pH by activation o f m em brane proton pumps, (6) conform ational changes in the viral capsid prom pting the release o f virions into the cytoplasm, and (7) transfer o f the virions to the nucleus During adenovirus infection, host cell DNA synthesis is com pletely blocked after about 12 hr after infection, and cellular RNA and protein synthesis blocked after 18-20 hr after infection (Ginsberg et aL. 1990). O f all the excess viral protein synthesized about 10% is utilized in new virus particles w ith the balance accumulating in the nuclear region. Using tem perature-sensitive mutants o f genes required for DNA synthesis, it was found that the genetic function o f the early viral genes rather than the nuclear accum ulation o f viral structural proteins is responsible fo r cell dam age and death (Ginsberg et aL. 1990). 4.4.2. Virus-induced Cellular Changes Infection o f polarized cells under controlled conditions cultured on perm eable filter supports enables dissection o f the virus alteration o f epithelial barrier function as w ell as ion and solute or nutrient tranqx)rt processes, generally studied in intestinal and kidney epithelial cells (Svensson et aL. 1991; G urevitz et aL. 1987; del Castillo et aL. 1991) or fibroblasts (Carrasco, 1978; Ulug etal.. 1989). 35 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Infection o f a particular virus is specific for a particular cell type. In the airway epithelium, hum an influenza virus infects cultured adult and fetal human tracheal and bronchial cells, but not human tracheal fibroblasts, or the human tracheal transformed cell lines 9H T E o, 3HTEo* or rhesus monkey kidney cells (RMKC) (Reiss et cd., 1991). Besides selectivity for a particular cell, different virus elicit different cellular responses. Semliki Forest virus infected MDCK and Caco-2 only when introduced to the basolateral surface, whereas rotavirus infection was sym m etrical in both cell culture models (Svensson et aL, 1991). Both viruses completely abolished R t 24 hr post-infection which was determ ined to be due to disruption o f intercellular tight junctions. In the airway epithelial cells, RSV infection o f the cotton rat airw ay epithelium decreased Isc by 28% but further tightened tight junctions w ith a 100% increase in R t (Cloutier et ai, 1989). On the other hand, influenza infection o f tracheal gland cells did not affect cells m orphologically (CPE), structurally (trypan blue exclusion, lactate dehydrogenase release) or functionally (^H-thymidine uptake) (Gentry et aL, 1988). Adenovirus penetration into the cell stim ulated cointem alization o f macromolecules (fluorescent labeled-HRP) and their release into the cytoplasm by viral lysis o f endosom al membrane (Defer et aL, 1990). This appears to be a consequence o f perinuclear condensation o f the vimentin (actin filam ent) cytoskeletal network. The increase in cellular permeability to HR? depends o n the integrity o f the viral capsid structure and not on the genetic competency or viral replication. Evidence for this was the ability o f a replication deficient noninfectious tem perature-sensitive m utant that induced the sam e enhancem ent in HR? permeability (D efer et aL, 1990). 36 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In addition to alteration o f barrier in teg ri^ , viral infection has been shown to affect epithelial membrane and membrane-bound protein function and expression. Membrane leakiness was also increased in 3T6 mouse cells infected w ith the EMC-virus. The lytic nature o f the infection increased the perm eability o f impermeable nucleotide drugs. The lack o f specificity in the passage o f low m olecular weight drugs (<750) supports the interpretation that membrane leakiness is caused by holes through which sm all molecules can diffiise through (Carrasco, 1978). Virus infection may affect certain m em brane proteins such as the glucose transporter and the ATP/ADP carrier-protein. It has been proposed that HlV-1 increases the m etabolic demands on the cells, creating a favorable environm ent for viral replication. This was dem onstrated by the increased expression o f Glut3, the facilitated glucose transporter, resulting in enhanced glucose transport activity (Sorbara et cd.. 1996). In the guinea pig heart, the coxsackie B3 virus induced a dysfunction in the ATP/ADP carrier creating an imbalance in the myocardial energy m etabolism (Schulze and Schultheiss, 1995). This was attributed to modulation in the expression o f the ATP/ADP carrier. Therefore, virus can potentially affect m em brane-bound carrier-proteins including the nucleoside transporters. Since nucleoside transporters play an important role in nucleic acid reabsorption as w ell as antiviral nucleoside transport into epithelial cells, it is crucial to exam ine W iether virus infection modulates nucleoside transporter activity thereby influencing the delivery o f nucleoside analogs that are substrates for the transporter. 37 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5. NUCLEOSIDE TRANSPORTERS The first lines o f evidence that nucleoside transporters play a role in nucleoside analog transport cam e fix>m cytotoxicity studies with anticancer nucleoside analogs, where the nucleoside transport inhibitor (nitrobenzylthioinosine, NBTI) was found to protect cultured cells (R P M I6410 cells) against cytotoxici^ o f a num ber o f nucleoside analogs (W am ick et aL. 1972; Paterson et aL, 1979). This finding was supported by another study that reported th at a m utant m urine mouse lymphoma cell line that lacks a fimctional nucleoside transporter was resistant to an array o f cytotoxic nucleoside drugs (Cohen et aL, 1979). These observations indicated that the protective effect was due to reduced accessibility o f the nucleoside drugs to cells. This led to the hypothesis that the delivery o f nucleoside analogs to epithelial cells can be facilitated by targeting the endogenous nucleoside carrier-proteins (Paterson et aL, 1981). This strategy to the conventional delivery or delivery system w ould utilize an endogenous cellular process and cellular energy to accum ulate nucleoside drugs within the cell. Furthermore, this would not necessitate any chem ical m odifications or conjugations that m ay affect the activity or metabolic activation o f the parent drug. However, to adopt this strategy a thorough understanding o f the uptake characteristics o f physiological substrates and the selectivity for the interaction o f structurally sim ilar analogs w ith the membrane carrier-protein would be warranted. Recognition o f carrier-m ediated transport o f nucleosides and their analogs follows the following criteria: (1) Evidence fiom rapid kinetics (generally w ithin seconds) o f drug 38 I Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. perm eation into cells; (2) Evidence 6om the effects o f nucleoside-specific transport inhibitors (nitrobenzylthioinosine, NBTI) on cellular uptake o f nucleosides and their analogs; and (3) Selective inhibition by physiological nucleosides and structurally sim ilar analogs i.e., com petitive inhibition o f radiolabeled fluxes with excess unlabeled nucleoside or nucleoside analog (Paterson et ai, 1981). 5.1. History o f Nucleoside Transporters The earliest reports o f carrier-m ediated nucleoside transport was reported as early as 1971 in hum an erythrocytes (O liver and Paterson, 1971). Since then several studies have characterized the selectivity^ o f the uptake process, the kinetic param eters, sensitivity to nucleoside and other ion and solute inhibitors, exam ined the substrate selectivity, and identified the type (isoform ) o f transporter (Gati et a i, 1984; M ahony et a i, 1991; Dom in et ai. 1993). Specialized nucleoside transporters have been identified in the plasm a m em brane o f a num ber o f cell types including epithelial cells in the intestine (V ijayalakshm i and B elt, 1988; Jakobs and Paterson, 1986; Iseki et a i, 1996; Hu, 1993), kidney (G utierrez et a i, 1992; Doherty and Jarvis, 1993; W illiam s and Jarvis, 1991), liver (M ercarder et a i, 1996) and choroid plexus (Wu et a i, 1992), leukem ia cells (Craw ford e t a i, 1990), lym phocytes (Chan et a i, 1993), erythrocytes (Cabantchick and G insburg, 1977), and m acrophages (Plagemann, 1991) listed in Table 1-3. Physiologically, these transporters play a crucial role in the conservation o f cellular nucleic acids (Plagem ann et a i, 1988), as an efiflcient salvage m echanism , and 39 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. t pharm acologically, they may play an im portant role in targeting o f nucleoside analogs to the affected epithelial cells (Brett et aL, 1993; Gutierrez and Giacom ini, 1993; W u et al. 1994). 5.2. Classification ofNuciaosida Transporters Six m ajor nucleoside transporters have been identified in m am m alian cell types, classified by th eir transport mechanism into: equilibrative (also known as facilitated difhision) and concentrative (Na^-driven cotransport) transporters. The equilibrative transporters are ubiquitous, being the prim ary m echanism o f nucleoside transport in neoplastic cells and cell lines, whereas concentrative transporters have been found predom inantly but not exclusively in normal cells (Belt et aL. 1993). Two subtypes o f the equilibrative transporters have been reported, one exhibiting high sensitivity to inhibition w ith NBTI designated as (ef) and the other insensitive to NBTI designated as (gf). Four isoform s o f the Na^-driven nucleoside cotransport system have been identified, that are insensitive to NBTI, w ith differing but overlapping substrate selectivity, and Na^-coupling stoichiometry. The isoform exhibiting selectivity for all the purines (adenosine, guanosine and inosine) and uridine is N1 (also known as cif, form ycin being a m odel purine), that selective for all pyrim idines (thym idine, cytidine and uridine) and adenosine are term ed N 2 (also known as cit, thym idine being a model pyrim idine). U ridine and adenosine are common substrates fo r all transporters. In the choroid plexus a transporter exhibiting broad selectivity for both purines and 40 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. pyrim idines w ith a 2:1 coupling ratio, designated as N3 (W u et a i. 1992), and another broad substrate selective subtype that excludes inosine, w ith a 1:1 coupling ratio designated as N 4 (G utierrez and Giacomini, 1993) (also know n as cib) was identified. Scheme 1-1 sum m arizes the type o f transporters and their distinguishing features in m am m alian tissues. 41 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T 3 O Q . C s Q . T 3 ( D W W o' 3 O Nucleoside Transporters C D 8 T 3 V C 3. ( Q 3 C D C D C D T 3 O Q . C S - . o 3 C D Q . T 3 C D C / J C / J o' 3 Equilibrative (Facilitated-difliision) Na^-indcpendent Concentrative (Uphill transport) Na-dependent es NBTI-sensitive nM inhibition (all nucleosides) ei NBTI-insensitive fiM inhibition (all nucleosides) Purines Adenosine, Guanosine, Inosine Pyrimidines Thymidine, Cytidine, Uridine Scheme 1-1: Classification of nucleoside transporters Purine selective N1 (cif) Na^:Nucleoside (1:1) (adenosine, uridine) Pyrimidine selective N2 (ciV) Na+iNucleoside (1:1) (adenosine, uridine) Broad spectrum Purine-pyrimidine N3 Na+:Nucleoside 2:1 (all nucleosides) Purine-pyrimidine N4 Na^:Nucleoside 1:1 (all nucleosides except inosine) 5.3. Equilibrative vs. Coneentrabve Transporters The transport o f nucleosides across cell mem branes is m ediated by three possible pathw ays: non-concentrative facilitated difhision, concentrative transport and non facilitated difhision (passive difhision). The m ost widely studied o f these is the non- concentrative facilitative difhision (or equilibrative transporter) in the hum an erythrocytes (Cabantchick and Ginsburg, 1977; Domin et aL, 1993). This class o f transporters functions independently ofNa% is kinetically sym m etrical and accepts a w ide variety o f purines and pyrim idines (Paterson etaL, 1981; Plagem ann, 1991). Those transporters that have a high affinity binding sites fo r NBTI (K d = 0.1 - 1 nM ) and are com pletely inhibited by nanom olar concentrations o f NBTI are designated as equilibrative-sensitive Çes) transporters, w hile those that lack high affinity binding sites and are effective inhibitors at higher concentrations (>1 ^M ) and designated as ei transporter. 5.4. Na*~dependence Concentrative nucleoside transporters are strongly dependent on Na% driven by the transmembrane Na^ gradient nucleosides are transported uphill into epithelial cells. This was verified by dissipation o f the Na^ gradient w ith the Na^K^ ATPase inhibitor ouabain that resulted in a 60% inhibition o f ^H-uridine uptake in rat liver parenchymal cells (M ercarder et aL, 1996). The ionophore gram icidin that abolishes Na^ and gradients decreased form ycin b uptake by 80% (Jakobs and Paterson, 1986). In epithelial cells 43 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. rem oval o f Na^ resulted in a virtually complete abolition o f nucleoside uptake in bovine renal BBM V(W illiams and Jarvis, 1991), intestinal epithelial cells (Jakobs and Paterson, 1986), choroid plexus (W u et al., 1994), and W alker 256 rat carcinom a cells (Crawford and Belt, 1991). This effect was selective for Na^, as replacement o f Na^ w ith inwardly directed gradients o f choline chloride and KCl failed to drive nucleoside uptake (W illiams and Jarvis, 1991; Vijayalakshmi and Belt, 1988). Other univalent cations like Li^ and Cs^ or buffered m annitol were ineffective in stim ulating nucleoside uptake (W illiams et aL, 1989; Jarvis, 1989) By m odifying the extracellular Na^ concentration the half-maximal activation o f nucleoside uptake appears to occur in the range o f 3-20 mM. In the rat intestinal BBMV and bovine renal BBM V the values reported were 13.3 mM and 11.4 mM, respectively (Iseki et aL, 1996; W illiam s and Jarvis, 1991). The coupling ration o f Na^ w ith the nucleoside in most studies was found to be 1:1. The only reported instances where a 2:1 Na^:nucleoside stoichiom etry was found was in the rabbit choroid plexus (Wu et aL, 1992). A Hill coefBcient o f 1 in each o f these studies suggests that one Na^ is recruited for the translocation o f one nucleoside to the interior o f the cell. 5.5. K/nefrc Parameters and Selectivity The initial velocities o f substrate accum ulation via nucleoside transporters is extrem ely rapid. There are tw o m ajor reasons why nucleoside uptake is studied over tim e frames that span seconds rather than m inutes. (1) Physiological nucleosides and their analogs are susceptible to rapid m etabolism . Anabolism and catabolism o f nucleosides could 44 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lead to intracellular trapping. Under these conditions, uptake curves may reflect the rates o f intracellular m etabolism such as phosphorylation o f the nucleoside rather than its transport velocity into the cell (Plagem ann et a i, 1988). (2) Several m am m alian cells posses equilibrative transporters that are also involved in the efllux o f nucleosides and their analogs. A steady state uptake profile over prolonged tim es (m inutes) could reflect a balance betw een the influx and efflux com ponents o f nucleoside transport rather than any one single event (Paterson et aL, 1981). Besides the difference in sensitivity to NBTI and requirements for Na% concentrative nucleoside transporters are distinguishable flom equilibrative transporters based on their kinetic parameters. Concentrative transporters exhibit a high affinity for substrates (Km <60 pM) and a low transport capacity, whereas equilibrative transporters have low affinities (>60 pM ) for substrates but a high transport capacity (Paterson et aL, 1981). This statement is supported by comparing the M ichaelis-M enton kinetic parameters o f nucleoside transport processes in several cell types listed in Table 1-3. 5.6. Nucleoside Transpotrters in Other Tissues 5.6.1. Lung The only known reports on nucleoside transport in the respiratory tract was reported by Shi and Young, (Shi and Young, 1986). The authors reported that membrane preparations 6om the guinea pig lung exhibited a high affinity binding o f the nucleoside transport inhibitors ^H-dipyridamole and ^H-NBTI (apparent Kd = 2 nM), suggesting interaction 45 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. w ith an es nucleoside transporter confirmed later by Jzennan et al., (Jzerm an et aL, 1992). In the isolated rat lung. Das and Steinberg, (Das and Steinberg, 1988), using the single pass circulation firom the pulm onary artery dem onstrated that ^^C-adenosine uptake in the lung was a saturable process (K m = 0.23 mM, Vm ax = 4.6 pm ole/g/m in) that was not a£fected by glucose or ouabain. These results suggested that a specific nucleoside transport process is present in the lung that mediates the blood to lumen direction uptake o f nucleosides (Das and Steinberg, 1988). Taken together, these observations indicate that an equilibrative nucleoside transporters may be present in lung membranes. However, their localization so as to be utilized in inhalation drug delivery, and their detailed characteristics and interaction w ith nucleoside analogs rem ain unclear at this tim e. The paucity o f inform ation on nucleoside transporters in the airw ay and alveolar epithelium, necessitate a detailed understanding o f the nucleoside transport process, as has been investigated in great depth in the intestine and kidney. These w ould serve as a guide for the elucidation o f the types o f nucleoside transporters, their localization and substrate selectivity. In the following section, the types o f transporters, their substrate selectivity, kinetic param eters and will be summarized. 5.6.2. Intestinal Epithelial Cells M ost o f the nucleoside transporter studies involving cultured intestinal epithelial cells, brush border membrane vesicles, or excised tissue indicate that nucleoside transport is primarily N a^-dependent In cultured lEC 6 cells form ycin uptake was concentrative, accumulating at a concentration 6-tim es higher than extracellular concentration (Jakobs 46 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and Paterson, 1986). In addition to the Na^nlependent uptake, an NBTI sensitive es transporter was also identified. In another study, Vijayalakshm i and Belt, 1988, determined the kinetic param eters (Km = 45 ± 3 pM , Vmax = 4.4 ±2.2 pmol/IO^ cells/s) and dem onstrated that two Na^-dependent nucleoside transporters N I and N2 were involved in the transport process. In Caco-2 cells, thym idine transport was saturable w ith a Km o f 44.6 pM and a V m ax o f 1.7 pmol/cmVs, m ildly inhibited by the absence o f Na^ (32%), but sensitive to N B TI inhibition (Hu, 1993), suggesting that nucleoside transport in Caco-2 cells is mediated by es nucleoside transporters. Furthermore, the nucleoside analog AZT was found not to be a substrate for the es transporter, opting instead for nonfacilitated diffiision. In rat and rabbit intestinal BBM V nucleoside uptake was Na^- dependent (1:1 stoichiom etry), m ediated by a broad selective N3 transporter, w ith comparable K m ’s ranging from 6.0 - 21 pM (Jarvis, 1989; Iseki et aL, 1996). In the isolated rat sm all intestine, B ronk and coworkers (Stow and Bronk, 1993; Bronk and Hastewell, 1988) dem onstrated that all purines w ere absorbed but extensively m etabolized down the degradative pathw ay to uric acid, while o f the pyrimidines uridine and thymidine w ere rapidly absorbed w ith very little appearing in the intact form on the serosal side. Cytidine was found to be the only nucleoside poorly absorbed (Stow and Bronk, 1993; Bronk and H astew ell, 1988). In the perfused rat intestine, nucleosides and d4T were transported by N 2 and N3 nucleoside transporters with Km o f 220 pM and 59 pM respectively. 47 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5.6.3. Kidney Epitheiiai Ceiis Sim ilar to the trend in intestinal epithelium, kidney epithelial cells also express both Na^- nucleoside and equilibrative transporters. Evidence o f this observation was reported in cultured LLC-PK, (GrifBth et al., 1992) and O K cell lines (D oherty and Jarvis, 1993). In both cell lines majority o f uridine influx was m ediated by facilitated difiusion {es nucleoside transporter). The apparent Km in O K kidney cells w as 160 ± 41 pM , and 34 pM in LLC-PKi cells. In either case, the Na^-dependent uptake w as consistent w ith N1 nucleoside transporter selectivity. In the rabbit (W illiams et al., 1989) and bovine kidney epithelial BBM V (W illiams and Jarvis, 1991) uridine uptake w as stim ulated by Na^, w ith a Km’s o f 12 pM and 10 pM respectively. In both species the NX and N2 transporters were reported to mediate nucleoside reabsorption in the renal epithelial cells. By contrast, human BBM V a single broad substrate selective transporter that accepts all purines and pyrim idine like the broad selective N3 transporter except it excludes formycin and inosine. Being unique from previously described transporters this was designated as N4 (G utierrez et a i, 1992). The characteristics o f this transporter were that it w as dependent on Na% with a 1:1 coupling ratio and saturable with a Km o f 4.8 ± 0.4 pM (Gutierrez and Giacomini, 1993). The kidney epithelium is one o f the only epithelia in w hich the overall vectorial movement o f nucleosides is understood. It w as proposed that uridine is translocated across the apical brush border membrane o f rabbit proxim al renal tubules by Na^-dependent pathways, but permeates the basolateral membrane by NBTI sensitive facilitated diffosion 48 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Williams et al., 1989). Evidence for this theory was supported by the finding that nucleoside influx across the surface membrane vesicles was Na^-dependent w ith a Km o f 12 ± 3 pM. In contrast basolateral uptake was not dependent on Na^ and highly sensitive to NBTI binding. Basolateral uridine uptake was saturable w ith a Km o f 137 ± 20 pM and a high affinity binding site for NBTI (Kd = 0.74 ± 0.5 nM. A sim ilar scenario was found in cultured OK cell line, where the basolateral uptake was m ediated by an es nucleoside transporter (Doherty and Jarvis, 1993). In both studies involvem ent o f the basolateral transporter in efflux was not investigated. [ 5.6.4. Brain Tissues Nucleoside transport in the brain was studied to understand the extent o f blood to cerebrospinal fluid (CSF) transport o f nucleosides and nucleoside analogs. Isolated rabbit choroid plexus was found to accum ulate and metabolize uridine (Spector, 1985). In brain tissue ATP-depleted slices (to prevent phosphorylation), nucleoside uptake was stimulated in the presence o fN a \ w ith a coupling ratio o f 2:1. Nucleoside uptake was saturable with Km values o f 18 and 13 pM for uridine and thymidine respectively, exhibiting broad substrate selectivity being inhibited by all physiological nucleosides (W u et a i, 1992). Further characterization o f the Na^-dependent nucleoside transporter N3 revealed that clinically relevant base m odified (5-fluorouridine, 2-chloroadenosine, and 5-iodo-2’- deoxyadenosine) but not ribose m odified nucleoside analogs (AZT, ddC, ddl) interact with this transporter and m aybe involved in carrier-mediated uptake to the brain (W u et ai, 1994). 49 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5.6.5. Other Cell Types Nucleoside transporters have been functionally and structurally characterized in a number o f non-epithelial tissue such as leukem ia cells (Craw ford et aL, 1990) lymphocytes, macrophages, bone marrow cells (Chan et aL, 1993), erythrocytes (Cabantchick and Ginsburg, 1977; Mahony et al., 1991) and carcinom a cells (Crawford and Belt, 1991; Boumah et aL, 1994). W ith respect to the nucleoside transporters, the leukemia cell line L1210 is one o f the most com plex. It was one o f the first tum or cell lines shown to have both Na^-dependent (N l) and Na^-independent (es and ei) nucleoside transporters (Crawford et al., 1990). Initial influx o f nucleoside w as determined to be via all three transporters, as the intracellular concentration approaches that o f the extracellular medium, only N l was capable o f continued nucleoside influx, driven by Na% while the es and ei transporters act predominantly as effiux routes (Craw ford et aL, 1990). O f the human effector cells, the rank order o f uridine uptake w as lymphocytes > macrophages > bone marrow cells (Chan et aL, 1993). The apparent K m ’s were 18.4,25.3, and 90.2 pM, respectively. For all three cell types facilitated diffiision o f the es type was the m ajor route for nucleoside transport The dideoxynucleosides (ddT, ddl and ddC) showed no detectable penetration in all three cell types whereas ddA permeated cells alm ost as efficiently as uridine. In W alker 256 carcinoma cells, nucleoside uptake was 10-fold greater in Na^-containing m edium than in Na^-fiee m edium and was inhibited by inosine and not thymidine. This is consistent with a N l nucleoside transporter. Furthermore, an equilibrative transporter that was not sensitive to dipyridam ole was also identified 50 I Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. indicating the sim ultaneous presence o f an ei and N l o f transporter (Crawford and Belt, 1991). 5.7. Expression and Cloning Isolated oocytes o f Xenopus laevis accurately translate heterologous mRNA and have been used in the expression cloning o f a number o f m em brane transport proteins for which antibodies are unavailable. In functional expression studies, polyadenylated mRNA isolated from the rabbit sm all intestine (Jarvis and G rifSth, 1991) and rat jejenum (Huang et al, 1993) injected into oocytes yielded a nucleoside transporter w ith characteristics resem bling the N2, and a combination o f N l and N3 respectively. Sequencing o f the N2 transporter yielded a 2420 base pair transporter with a relative m olecular mass o f 71,000 daltons. (Huang et al, 1994).Giacomini and coworkers dem onstrated that the human renal nucleoside transporters N 2 and N4 can be expressed m Xenopus laevis oocytes. M icroinjection o f 2-3 kb size-fractionated polyadenylated m RNA isolated from human renal cortex resulted in the expression o f a N 2 and N 4 (Giacom ini et al, 1994; Gutierrez and Giacomini, 1994). The transporter a fS n i^ constants Km o f 28 pM was sim ilar to that in BBM V (27 pM ), and the inhibition pattern consistent w ith the classical pyrimidine and broad selective transporter (Gutierrez and Giacomini, 1994). 51 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T 3 O Q . C S Q . T 3 ( D W W O 3 O Table 1-2: Nucleoside Analog interaction with Nucleoside or Nucleobase Transporters CD 8 T 3 V C 3. ( Q Model System Drug Interaction General Observations Rabbit intestinal BBMV 2-CA, 2’-dl, 2’-dU, S-FUrd Yes Reference (I) Aii nucleoside analogs inteiact with nucleoside transpoiteis (Jaivis, 1989) 3 C D Rat intesinal BBMV 6-MPR Yes (I) Multiple transporters, probably Nl and N3 involved (Iseki era/,, 1996) C D C D T 3 O Q . C S - . o 3 CD Q . T 3 CD C / J C / J o' 3 Caco-2 cells Human kidney BBMV Human kidney BBMV COS-1 kidney cell line Rabbit choroid plexus ddC.AZT AZT. ddC, Ara-C 2-CA,5-FUrd,IDU floxidine, IDU AZT, ddC, Ara-C dFdC,3TC 2-CA, IDU, FUrd AZT.ddl.ddC No No (I) No interaction with es transpotter (Hu, 1993) Yes Yes Yes (partial) No Yes No (I) Sugar modifications render nucleoside compound inactive, 2. (Brett era/,, 1993) 3 -OH important for interaction, 3 .2 -Trans-OH hinders interaction (1) Base modified analogs show strong interaction with nucleoside transporter ( I ) 2 -deoxyribosyl analogs most potent inhibitors (2) Sugar modified analogs least potent (Gutierrez and Giacomini, 1993) (Fang era/., 1996) ( I ) Base modified analogs show interaction with transporter (Wu et al., 1994) (2) Sugar modified analogs do not interact with transporter Rat macrophages 2-CA, 5-Furd, S’-dA Yes 3 -dA, ddA, 8-AZA, tubercidin. No S'-MTA, 6-MPR, Ara-A (1)Concentrative transporters mote restrictive than equilibrative (Plagemann, 1991) (2) 3 -OH important for substrate 7 3 ( D T 3 O Q . C S Q . T 3 ( D W W O 3 O (D 8 T 3 V C 3. ( Q Table 1-2 (continued) Lymphocytes, Macrophages ddA ddC.ddI.ddT Human erythrocytes OCV, IDU ACV Yes No Yes No (1) ddA the only dideoxynucleoside analog that is a substrate (Chan el al., 1993) (2) Other dideoxynucleosides cannot difTuse into cells ( I ) Substrate for nucleoside and nucleobase transporters (2) Substrate for nucleobase transporter (Mahony era/., 1991) 3 C D CD CD T 3 O Q . C S - . o 3 Human eiythrocytes Hhuman erythrocytes ddl.ddC ddG ddA,d4T ddC.ACV.GCV AZT, ddT Yes No No No No ( I ) Substrate for nucleoside and nucleobase transporters (Domin ei al„ 1 993) (2) Substrate for nucleobase transporter (3) Non facilitated diffusion (1) Substrate for nucleobase transporter (Gati el al.. 1984) (2) Non facilitated diffusion CD Q . T 3 CD C / J C / J o' 3 Abbreviations: ACV - acyclovir, GCV • ganciclovir, AZA - 8-azldoadenosine, AZT - 3 'azidothymidine, Ara-A - adenosine arabinoside, Ara-C - cytosine Arabinoside, 2-CA - 2-chloroadenosine, ddA - dideoxyadenosine, ddC - dideoxycytosine, ddl - dideoxyinosine, ddG - didexyguanosine, ddT-dideoxythymidine, d4T - stavudine, 2 -dA - 2 -deoxyadenosine, 2 -dl, - 2 -deoxyinosine, 2 -dU - 2'-deoxyuridine, 3 -dA - 3 -deoxyadenosine, 5-FUid - 5-fluorouridine, IDU - 5-iodo-2'-deoxyuridine„ dFdC - gemcitabine, 3TC - thiacytidine, 6-MOR - 6 meicaptopurine, 5 -MT - 5'-mcthylthioadenosine, w 7 3 ( D T3 O Q . C S Q . T 3 ( D W W O 3 O ( D O . T 3 V C ( Q 3 ( D ( D ( D T 3 O Q . C S - . o 3 CD Q . O C T 3 CD C / J C / J O 3 Table 1-3: Nucleoside Transport in Various Tissues Permeant Mechanism Transporters Km (pM) Vmax M atliial Formycin B Adenosine Uridine Thymidine, AZT Uridine, Ihymidine Inosine, guanosine Formycin B 6>Mercapropurine Thymidine Transporter Transporter Transporter NIandN2 4Sdb3 Na*-dependent nd Broad selectivity (N3, 6.4 ^ I. I N4) Transporter, diflusion Na^-dependent(N2) 44.6 ±11,7 Transporter Transporter Transporter Transporter ND ND NlandN3 4.4 ± 2.2 pmoles/IO cells/s nd 9.1 ± 3.6 pmoles/mg protein/s 1.73 ±0.17 pmoles/cmVs 61 ± 4.4,97.1 ± 5.7 320 ± 32,24.7 ± 1.4 nmole/min/mg protein. 318±45,41.4±3.6 es and Na^-dependent nd Expressed Transporter N2 103.7 ±16.9 36 ±12 79.3 ± 4.0,20.5 ± 0.05 pmole/min/mg protein nd 61 ± 15.4 pmoles/mg protein/10 s 15.5 ± 2.2 pmoles/oocyte/hr Model lEC cell line IEC6celi line BBMV Caco 2 cell line Isolated rat jejenum Isolated rat jejenu. lEC cells Rat Intestinal BBMV Rabbit intestinal Reference (Vijayalakshmi and Belt, 1988) (Jakobs el a/., 1990) (Jarvis, 1989) (Hu, 1993) (Bronk and Hastewell, 1988) (Stow and Bronk, 1993) (Jakobs and Paterson, 1986) (Iseki era/., 1996) (Jarvis and GrilTith, 1991) % 7 3 ( D T3 O Q . C S Q . T 3 ( D en en O 3 O C D I i 3. e o 3 CD C 3 - 3" CD C D T 3 O Q . C S - . o 3 T 3 O CD Q . O C T 3 CD en en O 3 Table 1-3 (continued) Uridine Expressed Transporter N2 Uver Uridine Adenosine M rA ie y Uridine Uridine Uridine Uridine Uridine Uridine Thymidine Thymidine Thymidine Transporter Transporter Transporter Transporter Transporter Transporter Transporter Transporter Transporter Nl Nl andes Nl andes NIandNZ Na'-dependent es N4 N4 Expressed Transporter N2 Expressed Transporter N4 37 ei and Na*-dependent 8-13 14 I60db4l 27 ±7. 41±7 I3± 4 I2± 3 137 ±20 4.76 ±0.4 27 nd 28.2 ±4.2 nd 246 pmoles/IO‘/3 min 142 pmoles/mg protein/s 610 ± 100 pmoles/mg protein/min Rat intestinal (Huang era/., 1993) Rat Liver parenchymal cells (Mercarder er al, 1996) Rat Liver canalicular membrane vesicles (Che era/., 1992) OK opposum kidney cell line (Doherty and Jarvis, 1 993) l.3±O.I2,2.5±O.I4 pmoles/cg protein/s LLC-PK, kidney cell line (Griflithe/a/., 1992) 8.1 0.9 pmoles/mg protein/s 3.9 ± 0.9 pmoles/mg protein/s 5.2 ± 0.6 pmoles/mg protein/s 6.42 ± 0.2 pmoles/mg protein/s nd nd 7.5 ± 0.8 pmoles/oocyte/30 min Bovine membrane vesicles (Williams and Jarvis, 1991 ) Rabbit renal BBMV Rabbit renal basolateral membrane vesicles Human kidney BBMV Human kidney BBMV Human renal cortex Human renal (Williams era/., 1989) (Williams era/., 1989) (Gutierrez era/., 1992) (Gutierrez and Giacomini, 1993) (Giacomini era/., 1994) (Gutierrez and Giacomini, 1994) C O 7 3 ( D T3 O Q . C S Q . T3 ( D W W O 3 O ( D 8 T3 V C 3. ( Q Table 1-3 (continued) Brain Uridine, ihymidine Guanosine Transporter Transporter N3 N3 18,1 ± 2,13*2,3 5,5 ± 0,3,1,0 ± 0,2 nmoles/g protcin/s Rabbit choroid plexus 30,5 ±4,2 4,4 ± 0,2 nmoles/g/s Rabbit choroid plexus (Wurto/,, 1992) (Wuero/,, 1994) 3 CD CD CD T3 O Q . C g, o 3 CD Q . T3 CD C / J C / J o' 3 Other Cell lypts Uridine Ganciclovir Acyclovir Iododeoxyuridine Uridine Uridine Uridine ddA AZT, ddl, ddC, ddT Transporter Equilibrative Nucleobase/Nucleosid Equilibrative e Transporter Nucleobase Equilibrative Transporter Nucleoside Equilibrative Transporter Transporter Transporter Transporter Transporter Diflusion Equilibrative Equilibrative Equilibrative Equilibrative 73 ±63 l4±3m M 260 ±20 ISO ± 1 0 18,4 ± 4,2 25,3 ±6,6 90,2 ±10,1 nd nd 0,53 ± 0,038 mM/min 31 ±2 pmoles/5 pl/s 33 ±6 pmoles/5 pl/s 540 ± 40 pmoles/5 pl/s 0,05 ± 0,01 pmoles/IO* cells/s 0,04 ± 0,01 pmoles/IO* cells/s 0,03 ±0,01 pmoles/IO* cells/s nd nd Human RBC Human RBC Human RBC Human RBC (Cabantchick and Ginsburg, 1977) (Mahony era/,, 1991) (Mahony era/,, 1991) (Mahony era/,, 1991) Human lymphocytes (Chan ei a/„ 1 993) Human macrophages (Chan er a/., 1 993) Human bone marrow (Chan er al., 1993) mononuclear cells lymphocytes, macrophages, (Chan er a/,, 1 993) mononuclear cells lymphocytes, macrophages, (Chan er al., 1993) mononuclear cells » Abbreviations: AZT - 3'-azidothymidine, ddA - dideoxyadenosine, ddC - dideoxycytosine, ddl - dideoxyinosine,, ddT-dldeoxythymidlne, II. STATEMENT OF THE PROBLEM 57 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Goals and Objectives The long term goal o f this project was to develop a rabbit airw ay cell culture m odel to study airw ay epithelial drug transport and to understand the m echanism o f nucleoside and nucleoside analog transport in the airw ay epithelium , so as to elucidate the essential structural requirem ents that dictate their interaction w ith nucleoside transporters. T his approach w ould present a strategy for the future design o f antiviral nucleoside analogs that w ould target cell surface nucleoside carrier-proteins thereby optim izing their delivery to cells enhancing their overall therapeutic effect. Central H ypothesis: T his research is based on the hypothesis th at specialized nucleoside transport processes on the apical surface m ediate the influx o f extracellular physiological nucleosides and structurally sim ilar exogenous nucleoside analogs into the conducting airway epithelial cells. The specific aims o f the project are: (1) To develop and characterize a rabbit airway epithelial cell culture nw delfor ion and drug transport studies, and to com pare its characteristics to that found in vivo and other culture models. The rabbit trachea was selected as the m odel for studying ion and drug transport in the airw ay for the following reasons: (1) rabbit tracheal epithelial cells appear to have 58 \ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sim ilar m orphologically characteristics exhibiting growth and differentiated features, electrical properties and m ucociliary function as hum an airway epithelia (Jetten et a i, 1987; W u and Sm ith, 1982; Z ietlin et al.. 1988); (2) excised rabbit trachea have been reported to be predom inantly Na^-absorbing in nature, sim ilar to the hum an airw ay epithelium (Zietlin er a/., 1988; Jam igan era/., 1983); (3) cell isolation from rabbit trachea provide a relatively high yield com pared to the small anim al m odels (rat, guinea pig, ferret and ham ster airw ay m odels) (Robison et al., 1993), while presenting an anim al handling convenience and availability restrictive in the larger anim al m odels such as dog, bovine and hum an airw ay epithelial m odels. Cultured rabbit airw ay epithelial cells have been reported to grow th and differentiation o f form pseudostratified tight epithelial layers with high electrical properties (Zietlin et al., 1988) and m ucociliary function (W u et al., 1985; Kim , 1985). In this study, the effect o f air-interfaced conditions on airway epithelial grow th and differentiation was investigated. A s depicted in Schem e 2-1, the overall research plan was to evaluate the epithelial m orphology, bioelectric properties, ion transport, cellular respiration, and drug perm eability characteristics in the air-interfaced and liquid- covered airway epithelial cell culture models. Focusing on th e (air-interfaced culture) m odel that m ost resem bles the native epithelium , the influence or solute size and lipophilicity on the transepithelial passive perm eability was assessed. U sing m acrom olecules o f varying sizes (FITC -dextrans 4 - 70 kD), the m olecular sieving cut-off lim it and the equivalent pore-size th at restricts the paracellular perm eability o f m acrom olecules w as examined. The influence o f solute 59 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lipophilicity on transepithelial drug transport was evaluated w ith solutes o f the same size but w idely differing log octanol w ater partition coefiQcient (log P -0.6 - 3.2), so as to determ ine the optim al log P for transcellular perm eability and to predict the perm eability o f a drug based on its partition coefficient. 6 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T 3 O Q . C s Q . T 3 ( D W W o' 3 O Specific Aim # 1 CD 8 T 3 V C 3. ( Q 3 CD CD i T 3 O Q . C S - . o 3 T 3 O CD Q . T 3 CD C / J C / J o' 3 Rabbit Tracheal Culture Development Permeability Morphology Bioelectric Properties Ion Transport Properties Biochemical Molecular size Lipophilicity Lactate release Electron microscopy Periodic acid Schiff Na+ current Cl current HCOt current PD, Rt and Isc Substratum Media serum Scheme 2-1: Scheme of specific aim # 1 O N (2) To investigate whether nucleoside and nucleoside analog uptake is m ediated by a specialized carrier-m ediated process in the airw t^ epithelial cells. N ucleoside transporter have been dem onstrated to be involved in the antiviral nucleoside analog transport into epithelial cells in the intestine and kidney (G utierrez and G iacom ini, 1993; B rett etaL, 1993; Iseki etaL, 1996; Jarvis, 1989). This provides a particularly useful approach to deliver these hydrophilic drugs across the restrictive apical m embrane. T he existence o f such nucleoside transporters in the airw ay epithelium is unknow n. Therefore, the aim o f this study was to establish evidence for active carrier-m ediated nucleoside transport process in the airw ay epithelium , by determ ining the uptake specificity for nucleosides, the dependence on N a \ and the overall kinetic properties (Scheme 2-2). Uridine w as used as the m odel nucleoside, as it is a com m on substrate for all known nucleoside transporters and is m etabolized to a lesser extent com pared to other physiological nucleosides (C abantchick and Ginsburg, 1977). The interaction o f nucleoside drugs w ith the nucleoside transporter was investigated by com petitive inhibition o f ^H-uridine uptake. A nalogs w ith m odification on the base and the sugar ring were tested to elucidate the m inim al structural requirem ents a substrate should possess in order to com pete w ith uridine for the nucleoside transporter. Furthermore, the displacem ent o f uridine binding or internalization w as confirm ed with a radiolabeled nucleoside analog that is a substrate 6 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. s for the transporter and another th at does not interact with the airw ay epithelial nucleoside transport system. T o determ ine whether nucleosides once internalized, are effluxed to the underlying tissue, the basolateral uptake and efQux o f nucleosides was studied. This w ould provide a better understanding o f the overall scheme o f nucleoside flow in the airway epithelium , and serve as an indicator o f the percentage o f nucleoside drug that can be delivered to the tissue underlying the surface epithelium . The possibility o f m ultiple transporter isoform s w as tested w ith functional inhibition studies using inosine as the m odel purine for the purine selective nucleoside transporter and thym idine as the model thym idine nucleoside. Since antiviral nucleoside drugs w ould be delivered to the conducting airw ay epithelium at the tim e o f infection, the final stage o f the project was to test w hether nucleoside transport activity is retained in cells infected w ith virus. Ad5 w as used as the model virus am ong others that firequently infects respiratory epithelial cells. Besides the possible alteration o f nucleoside transport function, the effect o f virus on epithelial barrier properties, its bioelectric properties and solute perm eability w as also evaluated. 63 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T3 O Q . C S Q . T 3 ( D W W O 3 O ( D 8 3. ( Q 3 ( D C 3 - 3" ( D ( D T 3 O Q . C S - . o 3 T 3 O CD Q . Specific Aim # 2 Nucleoside Transporter Transport Characteristics Substrate Selectivity Identification Virus-aiteration Localization O c " D ( D ( / ) o' D N m + dependence Na^:nucleoside Electrogenecity NBTI-sensitivity Satunibility Purines Pyrimidines Btase modified analogs Sugar modified analogs Inhibition pattern Thymidine, Inosine Na^-dependence Saturability Transport, Basolateral uptake and efflux Na^-dependence Time-dependent uptake g Scheme 2-2; Scheme of specific aim # 2 1 1 1 . MATERIALS AND METHODS 65 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1. Animal Model M ale N ew Zealand W hite rabbits weighing 2.5-3.0 kg were obtained from Irish Farm s (Los Angeles, CA). A ll studies involving rabbits in this project confr>rmed to the Guiding Principles in the C are and Use o f A nim al (DHEW Publication, NTH 80-23). A ll cell culture supplies such as disposable petri dishes (60 m m and 100 m m ), pipets, disposable centrifuge tubes (15 and 50 m l), cell strainers (70 pm) were obtained from Falcon (Lincoln Park, N J). Transwells (0.45 pm , 12 mm and 6.5 mm O.D .) w ere obtained from C om ing-C ostar (Cambridge, M A). A ll chem icals used were o f the highest purity com m ercially available. A ll m aterials and th eir sources for the prim ary culture o f rabbit tracheal epithelial cells, ion transport inhibitors, transport buffer reagents, perm eability m arkers, FITC-dextrans, fi-adrenergic agents, radiolabeled solutes, physiological nucleosides, and nucleoside analogs are listed in Table 3-1. Chem ical stm ctures o f inhibitiors and substrates used in the study, as well as the list o f chem icals and suppliers are located at the end o f this section. 2. Rabbit Tracheal Epithelial Cultures Rabbit tracheocytes were isolated by a procedure originally developed by W u and Sm ith (W u and Sm ith, 1982), and m odified by R obison et al. (Robison et al., 1993). N ew Zealand W hite rabbits (-2 .5 Kg) were injected w ith a lethal dose o f Eutha-6 (0.5 m l/kg) intravenously via the m arginal ear vein. The thorax w as opened and incisions m ade ju st 66 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. below the larynx and the bifurcation o f the m ain stem bronchi. The excised trachea was im m ediately placed in a sterile petri dish containing ice-cold Ca^^/Mg^^ -free H anks’ balanced salt solution (HBSS). The tracheal tissue w as w ashed several tim es in HBSS and the extraneous cartilage and connective tissue from the outer surface trim m ed ofT aseptically. The tissue was cut into sm aller pieces, opened longitudinally and incubated in 0.2% Pronase E in minimum essential m edium w ith E arl’s salts and L-glutam ine (S- MEM) at 37®C in 95% air / 5% COg for 90 min. The epithelial layer was gently scraped off* w ith a sterile scalpel blade (size 10). The isolated cells were m ixed in a solution o f 0.5 m g/m l DNase I solution containing 10% fetal bovine serum (FBS) preequilibrated at 37®C w ith 95% air / 5% CO2. The cell suspension w as m ixed using a sterile pipet and centrifuged at room tem perature for 10 m in at 210x g. The supernatant was rem oved and the cell pellet was suspended in S-M EM containing 10% FBS and centrifuged a second tim e for 10 m in at 2 l0 x g at room tem perature. The supernatant was suctioned o ff and the cell pellet suspended again in S-M EM containing 10% FBS, subsequently filtered through a 70 pm cell strainer, and centrifuged a third tim e at the sam e settings. The resulting cell pellet was suspended in PC-1 m edium (Biowhittaker, W alkersville, MD) (a serum free, low -protein defined m edium containing known amounts o f insulin, transferrin, growth and attachm ent factors and essential fatty acids, in a 1:1 Dulbecco’s m odified Eagle’s m edium (DM E) and H am ’s nutrient F12 m edium , DM E:F12) supplem ented w ith 2mM L-glutam ine, 100 units/m l penicillin, 100 pg/m l streptom ycin, 50 pg/m l gentam icin, and 1 pg/m l fungizone (Life Technologies, Grand Island, NY). 67 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The overall steps in the culturing o f rabbit tracheal epithelial cells is show n in scheme 3-1. 6 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T3 O Q . C S Q . T 3 (D W W O 3 O (D 8 T 3 V C 3. ( Q 3" i 3 Ç D ■ n c 3. 3" CD i T 3 O Q . C S - . o 3 T 3 O CD Q . O C T 3 CD C / J C / J O 3 Rabbit Tracheal Epithelial Cell Culturing Excised trachea 0.2% Protease treatment for 90 min at 37°C I Isolated cells washed and filtered through 70 micron cell strainer I Cells suspended in PC-1 growth medium I Plated on collagen-coated Transwell filters at 1.3 x 10^ cells/cm^ I Air-interfaced Cultures or Liquid-covered Cultures I Scheme 3-1: Flowchart of steps in the culturing of rabbit tracheal epithelial cells 3 Cell viability was estim ated w ith the exclusion o f 0.2% trypan blue, and the isolated tracheocytes (15-25 x 10^ cells/rabbit) w ere plated on 12 mm or 6.5 m m Transw ell filters inserts (Com ing Costar, Cam bridge, M A) coated for 4 hr w ith m ixture o f 5.3 pg/cm^ collagen (rat tail type I), 1.8 pg/cm ^ fibronectin (hum an), and 1.8 pg/cm^ bovine serum album in (BSA) in PC-1 m edium , at a density o f 1.3 x 10^ cells/cm ^. For 12 mm filters, 0.5 ml o f cell suspension w as added to the apical side o f the filter, and 1.5 m l to the basal side (day 0), w hile for 6.5 m m filters, the apical and basolateral volum es were 0.2 m l and 0.9 m l, respectively. These volum es elim inate the hydrostatic pressure across the Transw ell filter. The plated tracheocytes w ere allowed to adhere to the collagen-coated Transwell filter (37®C in 95 % air/5 % CO2 ) After 24 hr, the basal and apical m edia were rem oved, in that order, and the cell m onolayers were gently washed once w ith pre-equilibrated PC-1 m edium (37°C, pH 7.4). The washing fluid was gently suctioned o ff and the m onolayers cultured either in an air-interface having their apical surfaces directly exposed to air or under conventional liquid-covered conditions (LCC). For 12 mm filters, 0.8 m l PC-1 m edium was added to the basolateral reservoir, w hile the apical surface o f cell layer w as directly exposed to am bient conditions. The apical surface o f the air-interfaced cultures (A IC ) were spared firom desiccation by a thin film o f adherent liquid (— 10-20 pi) and a hum idified environm ent (5% C02/95% air) in the incubator. For liquid-covered cultures (LCC), 0.5 m l PC-1 medium was added to the apical com partm ent and 1.5 m l to the basolateral. For 6.5 mm filters, the volum es for AIC were 0.3 m l in the basolateral com partm ent, w hile for LCC, the 70 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. volumes w ere 0.2 m l in the apical cham ber and 0.9 m l in the basolateral chamber. Culture m edium was changed daily. 3. Bioeieetrie Measurements To assess the tim e course o f the bioelectric properties o f the tracheal m onolayer 48 hr after plating onw ards, the transepithelial electrical resistance (Rt) and spontaneous potential difference (PD) were m easured each day using an EVOM voltohm m eter device (W PI, Sarasota, FL). PD (apical fluid as reference) was expressed in m illivolts (mV) and Rt, in kohm cm^. The leq estim ated from the ratio, PD/Rt w as expressed as |xA/cm^. F o r the bioelectric m easurem ents o f AIC in 12 mm filters, 0.5 m l and 0.7 ml pre-equilibrated m edium (37°C, pH 7.4) were added to the apical and basolateral reservoirs, respectively, to facilitate measurements. F or 6.5 mm filters, 0.2 ml and 0.6 m l were added to the apical and basolateral cham ber, respectively. The m onolayers were allowed to develop a steady potential for a period o f 5 m in prior to the bioelectric m easurem ents. The background R^ and PD o f the blank filter and bathing m edium were subtracted ftom the actual bioelectric m easurements. 4. Electron Microscopy of Tracheal EpiÜtelial Cultures Tracheal cell m onolayers were processed for electron m icroscopy on days 0 (just after isolation), 1 and 5. Each cell m onolayer was w ashed three tim es w ith ice-cold phosphate-buffered-saline (PBS, pH 7.2) and fixed in 2.5% glutaraldehyde in PBS at 4°C for 2 hr, rinsed in ice-cold PBS five times and fixed in 1.5% OsO^ in PBS at 4°C 71 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. overnight. Osmicated cell m onolayers were rinsed twice w ith PBS, three tim es w ith distilled water, and stained w ith 1.5% nranyl acetate in 50% ethyl alcohol for 30 m in. Stained monolayers w ere dehydrated in graded ethyl alcohol’s o f 7 0 ,8 5 ,9 5 , and 100% for 5, 5, 5 and 10 m in, respectively. These dehydrated m onolayers were then em bedded in Spurr resin and sectioned for transm ission electron m icroscopy w ith a JEOL lOOC at 80 keV. 5. Periodic Acid Schiff (PAS) Staining of Tracheocytes Tracheocytes were fixed fo r staining to estim ate the secretory cell population on day 0 and confluent day 6 cultures. D ay 0 cells in suspension were fixed w ith 10% buffered neutral form alin, incubated w ith 0.5% periodic acid for 10 min, rinsed, incubated fo r 15 m in in a Schiff solution, w ashed in running cold and warm tap water, and finally dipped in clarifier and dehydrated in alcohol and xylene. Day 6 cells were incubated w ith 0.5% trypsin-EDTA for 15 m in at 37°C to detach cells firom the Transwell filter, and the cell suspension fixed and stained in the same manner. The firaction o f positively stained cells were counted under a light m icroscope. In addition, selectivity for any particular extracellular m atrix was studied by growing tracheocytes on a Biocoat variety pack consisting o f inserts coated w ith either rat tail collagen type 1, fibronectin, lam inin or m atrigel (Collaborative B iom edical Products, Bedford, MA). The bioelectric properties were m easured as m entioned earlier. 72 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6. Ussing Chamber Studies Confluent m onolayers w ith peak R t and leq (days 4 through 7) were m ounted in m odified Ussing cham bers on a central a d ^ te r with a soft silicone 0-ring (coated with vacuum silicone grease), establishing a tight seal. M onolayers were bathed on both sides with PC-1 m edium th at was pre-equilibrated w ith 5% CO% in air at 37®C (pH 7.4) and the ion transport properties were studied as described by Kom pella et al. (K om pella et a l, 1993) and R obison et a l (R obison and Kim , 1994). Briefly, the spontaneous potential difference (PD) across the m onolayer was m easured by a pair o f m atched calomel half-cells and the Isc by a p air o f m atched Ag/AgCl electrodes connected to an autom atic v o lt^ e clam p device DVC 1000, W orld Precision Instrum ents. The monolayers were continually short-circuited except for 3 sec periods o f constant voltage pulses (dV=IO mV) applied across the m onolayer every 50 sec. The background voltage and R t o f the collagen-coated filter and the bathing medium were adjusted to zero prior to the m ounting o f the tracheal m onolayers. Cell m onolayers were allow ed to reach steady-state Isc (usually w ithin 15 min) before a pharm acological agent was added to the medium in the apical or basolateral reservoir. Responses were m easured as a percent change from control baseline values, and reported as m eans o f at least 3 m onolayers ± the standard error o f m easurement. Pharm acological agents w ith their final concentrations used in this study w ere (a) ouabain, a N a^ ^ A T P a se inhibitor, 10 pM ; (b) amiloride, a Na^ channel blocker, 10 pM; (c) benzam il, an analog o f am iloride considered to be a potent Na^-channel 73 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. blocker, 10 nM ; (d) fiirosem ide, a blocker o f the N a^(K ^C f cotransporter, 10 pM ; (e) N -phenylanthranilic acid, a chloride channel blocker NPAA, 0.5 mM ; and (f) 4,4’- diisothiocyanotostilbene-2,2’disulfonic acid (DIDS) a bicarbonate-related transport blocker, 50 pM . These agents were dissolved in w ater except for fiirosem ide and N PA A w hich w ere dissolved in m ethanol and DIDS in 0.1 M KHCO3. No appreciable response w as observed w hen the vehicle was added to the monolayers as controls. 7. Transepithelial Drug Transport Studies A ll transport experim ents were perform ed in Bicarbonated Ringer's solution (BR S), a t a pH 7.3 and an osm olarity o f 300 mOSm/kg, com prised o f 1.8 mM CaCl2, 0.81 mM M gSO^, 5.4 mM KCl, 116.4 mM NaCl, 0.78 mM NaH2P0 4 , 5.55 mM glucose, 15 m M N -[2-hydroxyethyl]piperazine-N '-[2-hydroxypropanesulfonic acid], and 25 m M N aH C O j. Confluent rabbit tracheal epithelial m onolayers (day 4-7) that exhibited PD o f ~30 m V and R^ o f — 1,000 ohms.cm^ were used in the transport studies. Cell m onolayers w ere w ashed once w ith BRS supplem ented w ith 0.2% bovine serum album in (BSA) preequilibrated at 37°C in 5% CO2 and 95% air. Five hundred m icroliters o f BRS was added to the apical cham ber and 1.5 m l to the basolateral cham ber and the bioelectric properties m easured after 15-20 m in stabilization. The donor fluid was spiked w ith drug solution producing final concentrations of: 5 pCi/m l ^H-mannitol, 1 pC i/m l ^^C-sucrose, and 1 pC i/m l ^H-dexamethasone o f radiolabeled drugs; 0.5 m g/m l for FITC -dextrans; 1 m M for atenolol, sotalol, and albuterol; and 0.1 mM for m etoprolol, tim olol, betaxolol, 74 I Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and propranolol. The more hydrophilic 6 -adrenergic com pounds (sotalol, atenolol, and albuterol) were studied at a higher concentration to enhance detection sensitivity. The stock solutions o f all radiolabeled drugs firom suppliers w as evaporated under nitrogen to e lim inate any possible solvent effects on epithelial p erm eab ili^ and any possible exchange o f radiolabel w ith the solvent. In the case o f FITC-dextrans, 300 p i w as sam pled firom the receiver com partm ent (i.e. basolateral fluid) in the apical-to-basolateral (AB) transport experim ent, while ISO pi was sam pled firom the apical fluid in the basolateral-to-apical (B A ) transport experim ent for fluorescence m easurem ent o n a Perkin Elm er 650-1OS fluorescence spectrophotom eter (Norwalk, CT) at an excitation w avelength o f490 nm and an em ission w avelength o f 515 nm. The sam pling tim es w ere 30, 60, 120, 180, and 240 m in after spiking w ith the dosing solution. FITC -dextrans w ere used without further purification, as they were foimd to be firee o f sm all m olecular weight products by g el perm eation chrom atography on a W aters Protein-Pak 300SW colum n (8 mm x 300 m m , 10 pm packing m aterial, and m olecular w eight firactionation range o f 10,000 to 300,000 daltons) (M ilford, MA). The FITC content was in the range o f 0.003-0.02 m ole p er m ole o f glucose. In the case o f fi-adrenergic compounds, 200 pi w as sam pled from the receiver com partm ent at the predeterm ined tim e intervals and 2 0 0 p i o f acetonitrile was added to precipitate the proteins in the sample m edium . Samples w ere then centrifiiged at 3000x g a t room tem perature for 10 m in, and the supernatant w as evaporated under nitrogen. T he residue was dissolved in 200 pi o f internal standard (listed in Table 3-2) in water, 75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. vortexed and 50 injected into the HPLC (Shimadzu SCL-6 A, Shim adzu C o., Ltd., Kyoto, Japan). Sam ples were analyzed on a reverse phase C IS M icrosorb colum n (4.6 x 250 mm, particle size 5 pm , Rainin, W obum , MA) according to the conditions listed in Table 3-2. The cum ulative am ount o f drug transported was plotted against tim e. In transepithelial transport studies, the apparent perm eability coefficient (Papp) w as estim ated firom the transepithelial flux (slope o f the above plot), norm alized for the surface area (1.13 cm^ for 12 mm, and 0.331 cm^ for 6.5 mm Transw ell filters), and the initial drug concentration (C^,) in the donor compartment. Papp = Flux / (Co* A) Eq. 1 The Papp values o f the FITC-dextrans w ere fitted to the relationship m odified firom that o f Durbin et al. (D urbin et al., 1956) by Kim and Crandall (Kim and Crandall, 1983) using a nonlinear least square curve fitting program (Table Curve 2D, Jandel Scientific Inc., San Rafael, CA) on an IBM com patible personal computer. The perm eability (Papp) can be related to pore area Ap, the diflusivity D, the m em brane thickness, dx, and total surface area S, using Pick’s law : Papp/D = (Ap/dx)/S Eq. 2 The total num ber o f cylindrical pores can be estim ated firom the total pore area Ap by N pTcr^. By norm alizing against a m annitol as the reference solute the ratio o f perm eability o f each dextran to m annitol: 76 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Papp/Papp’ = (D/D ) * f(x)/f(x') Eq. 3 where Papp', D', and f(x') denote the values for a reference solute (m annitol); D is the free d iffiisiv i^ o f the solute; and f(x) is given by the R enkin equation (Renkin, 1954). f(x) = (1 - x)^ * (1 - 2.104X + 2.09x^ - 0.95x®) Eq. 4 In the above equation, x = s/r, where "s" is the solute radius and "r" is the radius o f the j equivalent aqueous cylindrical pore. Equation 4 takes into account the two factors proposed by the theory o f restricted diffusion, the steric hindrance at the m outh o f the pore (the first factor) and the frictional resistance as a m olecule m oves through the pore (the second factor) (Landis and Pappenheim er, 1963). A ssum ing spherical or nearly spherical m olecules, the frictional force opposing diffusion is given by Stokes' law that describes the m otion o f a sphere in a viscous medium. T he m olecular radius "s" can be estim ated from d iffu siv i^ using the Stokes-Einstein equation: D = kT/6icr|s, where k is the Boltzm an constant, T is the absolute tem perature, and T | is the viscosity (poise) (Landis and Pappenheimer, 1963). D iffusivity o f FD w as estim ated from the relationship D(MW)*^ = constant for m olecular w eight (M W ) exceeding 360 daltons, where the constant was calculated using bovine serum album in (BSA) as a reference (MW = 67,000 daltons, D = 0.88 x 10"^ cmVsec). The total pore area, Ap is calculated from the estim ate o f Ap/dx. The Papp values o f the B-adrenergic compounds were fitted to a sigm oidal model also using the aforem entioned softw are on an IB M com patible personal computer: 77 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Papp = a + b /((l + exp(-(log P - c)/d) Eq. 5 w here log P is the log octanol/pH 7.4 partition coe£Bcient, "a" represents the m inim um Papp, "b" the maximum Papp, "c" the log P corresponding to the h alf m axim al Papp, and "d" the transitional slope o f the increase in Papp from "a" to "b". 78 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T 3 O Q . C S Q . T 3 (D W W O 3 O (D 8 3. ( Q 3 (D C 3 - 3" (D (D T 3 O Q . C a O 3 T 3 O CD Q . O C T 3 CD C / J C / J O 3 Table 3-2: HPLC Conditions for Assaying fi-Adrenergic Compounds; Wavelength of UV Detection, Internal Standard, Flow Rate, and Composition of the Mobile Phase. Drug Wavelength (nm) Internal Standard (pg/ml) Flow Rate (ml/min) Acetonitrile* Sotalol 225 Atenolol 1 3% Albuterol 277 Atenolol 0.8 2% Atenolol 225 Sotalol 1 3% Metoprolol 280 Timolol 0.7 18% Timolol 290 Propranolol 0.7 18% Betaxolol 225 Levobunolol 1 0-8 min 22-27% 8-21 min 27% 21-23 min 27-22% Propranolol 290 Timolol 1 0-10 min 20% 10-15 min 20-30% 15-25 min 30% 25-27 min 20% ‘The aqueous portion of the mobile phase was 0.2% triethy lamine at pH 3.0, except in the case of albuterol. For albuterol, 25 mM ammonium dihydrogen phosphate buffer containing 1 mM dimethyloctylamine at pH 3.0 was used. « 8. Measurement o f Nucleoside Uptake: Confluent air-interfaced rabbit tracheal epithelial cell m onolayers were w ashed once in BRS o r sodium -free R inger’s solution (SFR) made iso-osm olar with choline chloride, and equilibrated for 15 m in at 37 ® C. Thereafter, ^H-uridine (10 pCi/m l, 45.2 Ci/m m ol) was added to the apical side and uptake estim ated over predeterm ined tim es (5 ,1 5 , 30, 60 and 120 sec), as indicated for the individual experim ent Uptake was term inated by rapid aspiration o f the radiolabeled solution followed by immersing the filter a total o f 6-tim es in three 100 m l volumes o f ice-cold SFR. The filter was excised and the cells lysed in 0.5% T riton X-100. A 20 p i aliquot was w ithdraw n for estim ation o f protein content (BioRad Laboratories, H ercules, CA) and the balance mixed w ith Econosafe scintillation cocktail (Research Products International, ML Prospect, IL) and the radioactivity associated w ith the cell estim ated by liquid scintillation counting (BecKman L S I801, Fullerton, CA). F or inhibition studies, in most cases except where stated, zero-trans uptake was estim ated where the radiolabeled substrate and inhibitor were sim ultaneously introduced to the cells for predeterm ined times. The extracellular and intracellular w ater spaces were calculated using *^C-mannitol and ^H-H2Û as described by Schafer and Barfuss (Schafer and Barfiis, 1986). Total cellular volum e w as estim ated by the difference between the w et weight and dry weight o f cell m onolayers. This consists o f both extracellular and intracellular space. The extracellular space was m easured fiom the ratio o f the am ount o f cell associated '^C-mannitol to the dosing concentration. Subtracting the extracellular volume firom the total cell m onolayer 80 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. volum e is the estim ated intracellular volume. A ll uptake results were corrected for the unw ashable cell adhered radioactivity in the extracellular spaces. Data Analysis The am ount o f drug taken up by cells was plotted against tim e, norm alized for the protein content p er m onolayer, and the radioactivity bound to the filter. The rate o f uptake w as estim ated firom the initial linear portion o f the curve. In transepithelial transport studies, the apparent perm eability coefficient (Papp) was estim ated as described earlier in Eq. 1. The kinetic param eters for nucleoside uptake were obtained by non-linear least square fitting o f the data follow ing the M ichaelis-M enton equation: V = Vmax C / (Km + C) + IQC Eq. 6 where v is the carrier-m ediated flux, Vmax is the maximum velocity. Km is the concentration at half-m axim al velocity, C is the concentration o f the nucleoside. IQ is the constant for non-selective difiusion and the term , IQC represents the total non specific diffiisional com ponent (estim ated by concentration dependence at 4°C). Param eter estim ates w ere obtained by a non-linear least square curve fitting program TableCurve (Jandel Scientific Inc., San Rafael, CA). Likewise, the stoichiom etry o f Na"^: nucleoside coupling was studied in the presence o f varying concentrations o f Na^ (0-140 mM) m aintained iso-osm olar with choline chloride. The data was fit to the Hill equation Flux = V m ax [Na^]" / Km ." + [N a^") Eq. 7 81 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. where K^a is the Na^ concentration corresponding to h alf m axim al velocity 0.5 V m ax and n is the H ill coefficient. Param eter estim ates were obtained by a non-linear least square curve fitting program TableCurve. 9. Metabolism o f Nueleosides: The chem ical nature o f the cell associated rad io activ i^ was m easured by thin layer chrom atography. C ell layers were exposed to ^H-uridine for 60 sec then w ashed by the procedure described above. The filter w as cut-out and placed in ice-cold 15% trichloroacetic acid for 30 m in. Follow ing centrifugation at 3,000 x g for 10 m in th e supernatant was neutralized w ith N aO H , concentrated under nitrogen and spotted onto cellulose coated plastic plates (0.1 m m thickness) impregnated w ith p2S 4 fluorescent indicator (EM Science). The TLC plate was chrom atographed w ith w ater (Jakobs and Paterson, 1986) giving Rf values o f 0.4, 0.6 and 0.9 for uracil, uridine and uridine nucleotides, respectively, where the spots were location o f the spots were visualized under a UV lamp. Each lane was cut into about 1 cm strips and placed in 0.5 m l solution o f 0.7 M MgCl2 in 0.02M Tris/HCl a t pH 3 overnight to extract the radioactivity, prior to the addition o f 5 m l scintillation cocktail. 10. Infection o f Epithelial Cell CulWres (a) Infectivity C onfluent cell layers grow n on Transw ell filter inserts or 24 w ell tissue culture clu ster plates were inoculated w ith 0.2 m l o f A d5 stock solutions at a m ultiplicity o f infection 82 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (M O I) o f 20:1 plaque form ing units/cell (pfii/celi) and allowed to adsorb to m em brane surfaces for 2 hr. Thereafter, the inoculum was aspirated and the cells were w ashed four tim es w ith culture m edia to rem ove non-intem alized virus particles. Growth m edium w as added to the cells w hich were then placed in an incubator a t 37°C in a hum idified atm osphere o f 95% air/ 5% CO^. A t 0 ,2 4 ,4 8 , 72 h r post-infection (PI), the total virus associated w ith the cells w as assessed by the virus yield assay. In th e virus yield assay, cells w ere lysed by tw o freeze-thaw cycles, and an aliquot o f lysate added to A549 cells (lung carcinom a cell line). A fter 7 days incubation the cells w ere fixed w ith a 3:1 m ethanohacetic acid solution and stained w ith 1% crystal violet. T he num ber o f plaques (w hite necrotic area) w ere counted com pared with controls containing no virus. (b) Measurement of Bioelectric Properties in Adenovirus Type 5 Infected Cells C onfluent cells grow n on perm eable filter supports were infected as described above at an M OI o f 20:1 (20 virus particles per cell). C ells w ere m aintained in an air-interface w ith 0.8 m l in the basolateral compartm ent and placed in a incubator at 37°C in a hum idified atm osphere o f 95% air and 5% CO^ A t the tim e o f m easurem ent (every 12 fo r the first three days and 24 hr subsequently), 0.2 m l PC-1 m edium was added to the apical surface for the m easurem ent o f epithelial bioelectric properties with an EVOM V oltohm m eter (W Pl, Sarasota, FL). Following a stabilization period o f 5 min, the spontaneous potential difference (PD), and the transepithelial electrical ^ were 83 I Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. estim ated. On completion o f the bioelectric m easurements the apical fluid w as rem oved and cells returned to an air-interface. The bioelectric properties between infected and m ock-infected cell layers w ere compared. (c) Measurement of Transepithelial Mannitol Transport in Adenovirus Type 5 Infected Cells Ad5 w ild type and m ock-infected cell layers 48 hr post-infection were w ashed w ith bicarbonated Ringer’s Solution (BRS). A fter 15 m in stabilization in buffer 0.2 m l ^H- m annitol (10 pCi/m l) in BRS was added to the apical surface o f epithelial cells. The transepithelial appearance in the basolateral fluid m easured by withdrawing 100 pi sam ples at 15, 30,60, 120, and 180 min. The basolateral volum e was m aintained by replacing the withdrawn sam ple w ith fresh buffer. The extent o f transepithelial and cell associated radioactivity w as compared betw een the infected and m ock-infected cell layers. (d) Measurement of Nucleoside Transport Function in Adenovirus Type 5 Infected Cells ^H-Uridine uptake was m easured as a model substrate to evaluate carrier-m ediated nucleoside transport function in airway epithelial cells under infected or m ock-infected conditions. Confluent m onolayers 48 hr post-infection w ere washed with either bicarbonated Ringer’s solution (BRS) or Na^-free Ringers solution (SFR) w here the Na^ was replaced w ith an equim olar concentration o f choline^. ^H-Uridine (10 pC i/m l) uptake w as initiated by adding 0.1 ml radiolabeled solution (in BRS or SFR) to the 84 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. apical surface o f cells for 30 or 300 seconds. The solution was rapidly aspirated and the filter insert im m ersed twice in three, 100 m l volumes o f ice-cold SFR solution. The filter was then excised and placed in 0.5 m l o f 0.5% T riton X-100 for Ihr. Twenty m icroliters was taken for protein estim ation and the balance mixed w ith 5 m l scintillation cocktail for liquid scintillation counting o f cell associated radioactivity. 8 S Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T 3 O Q . C S Q . T 3 (D W W O 3 O (D 8 T 3 V C 3. ( Q 3 (D (D (D T 3 O Q . C S - . o 3 C D Q . O C T 3 C D C / J C / J O 3 Table 3-1: Chemicals, Reagents and their suppliers Chem ical Cell Culture Reagents Hanks’ Balanced Salt Solution (BBSS) Eagles minimum essential medium (S-MEM) Fetal Bovine Serum (FBS) Pronase E (bacterial protease type XIV) Deoxyribonuclease (DNase I, type IV) Penicillin - Streptomycin Gentamicin Fungizone L-Glutamine PC-1 medium Collagen (rat tail, type I) Fibronectin (human type V) Supplier Life Technologies Life Technologies Life Technologies Sigma Chemical Co. Sigma Chemical Co. Life Technologies Life Technologies Life Technologies Life Technologies Bio Whittaker Collaborative Biomedical Prod. Collaborative Biomedical Prod. Location Grand Island, NY Grand Island, NY Grand Island, NY St. Louis, MO St. Louis, MO Grand Island, NY Grand Island, NY Grand Island, NY Grand Island, NY Walkerville, MD Bedford, MA Bedford, MA Ion Transport Inhibitors Ouabain Sigma Chemical Co. St. Louis, MO 7 3 ( D T 3 O Q . C S Q . T 3 (D W W O 3 O (D 8 T 3 V C 3. ( Q 3" i 3 C D C D i T 3 O Q . C S - . o 3 T 3 O C D Q . O C T 3 C D C / J C / J O 3 Amiloride Sigma Chemical Co. St. Louis, MO Furosemide Sigma Chemical Co. St. Louis, MO Benzamil Sigma Chemical Co. St. Louis, MO N-Phenylanthranilic acid Aldrich Chemical Co. St. Louis, MO DIDS Sigma Chemical Co. St. Louis, MO Buffer Reagents CaClz^HgO Sigma Chemical Co. St. Louis, MO Choline chloride Sigma Chemical Co. St. Louis, MO Choline bicarbonate Sigma Chemical Co. St. Louis, MO D-Glucose EM Science Gibbs Town, NJ HEPES Sigma Chemical Co. St. Louis, MO KCl Sigma Chemical Co. St. Louis, MO MgS0 4 . Sigma Chemical Co. St. Louis, MO NaCl EM Science Gibbs Town, NJ NaHCOj Sigma Chemical Co. St. Louis, MO NaHgPO^ Sigma Chemical Co. St. Louis, MO KH2PO4 Sigma Chemical Co. St. Louis, MO 5 7 3 ( D T 3 O Q . C S Q . T 3 (D W W O 3 O (D 8 3 . ( O 3 " i 3 (D C 3 - 3" (D (D T 3 O Q . C g . O 3 T 3 O (D Q . O C T 3 CD C / J C / J O 3 FITC-Dextrans (FD-size in kD) FD-4, FD-10, FD-20, FD-40, FD-70 Sigma Chemical Co. St. Louis, MO fi-Adrenergic Compounds Atenolol Sigma Chemical Co. St. Louis, MO Albuterol Sulfate Sigma Chemical Co. St. Louis, MO Timolol maleate Sigma Chemical Co. St. Louis, MO Metoprolol tartarate Sigma Chemical Co. St. Louis, MO Propranolol HCl Sigma Chemical Co. St. Louis, MO Sotalol Bristol-Myers Squibb Evansville, IN Betaxolol HCl the late professor Hans Bundgaard, Royal Danish School of Pharmacy Radiolabeled Solutes [l-^H(N)]-Mannitol (26.4 Ci/mmol) New England Nuclear DuPont Boston, MA [6,7-^H]-Dexamethasone (43.9 Ci/mmole) New England Nuclear DuPont Boston, MA [*'*C]-Sucrose (0.58 Ci/nunole) Amersham Lifescience Arlington, IL [5,6-^H]-Uridine (45.2 Ci/mmole) Moravek Biochemicals Brea, CA [Methyl-^H]-thymidine (2 Ci/nunole) Moravek Biochemicals Brea, CA [2,8-^H]-Inosine(31 Ci/mmole) Moravek Biochemicals Brea,CA 0 0 00 7 3 ( D T 3 O Q . C S Q . T 3 (D W W O 3 O (D 8 3. ( Q 3 (D C 3 - 3" (D (D T 3 O Q . C S - . o 3 T 3 O CD Q . O C T 3 CD C / J C / J O 3 [2,8-^H]-2’-Deoxyadenosine (29 Ci/mmole) [5-^H]-HPMPC (22.1 Ci/mmole) Physiological Nucleosides Uridine Thymidine Cytidine Adenosine Guanosine Inosine Uracil Ribose Uridine S'-triphosphate Phloridzin Nitrobenzylthioinosine (NBTI) Moravek Biochemicals Moravek Biochemicals Sigma Chemical Co. Sigma Chemical Co. Sigma Chemical Co. Sigma Chemical Co. Sigma Chemical Co. Sigma Chemical Co. Sigma Chemical Co. Sigma Chemical Co. Sigma Chemical Co. Sigma Chemical Co. Sigma Chemical Co. Brea, CA Brea, CA St. Louis, MO St. Louis, MO St. Louis, MO St. Louis, MO St. Louis, MO St. Louis, MO St. Louis, MO St. Louis, MO St. Louis, MO St. Louis, MO St. Louis, MO Antiviral Nucleoside Analogs 5-Fluorouracil (5-FU) 3 Sigma Chemical Co. St. Louis, MO 7 3 ( D T 3 O Q . C S Q . T 3 (D W W O 3 O (D 8 3 . (û 3 " i 3 (D C 3 - 3" (D (D T 3 O Q . C S - . o 3 T 3 O CD Q . O C T 3 CD C / J C / J O 3 S-Hydroxyuridine (5-HU) Sigma Chemical Co. St. Louis, MO 5-Iodouridine (5-IU) Sigma Chemical Co. St. Louis, MO 5-Iodo-2’-deoxyuridine (IDU) Sigma Chemical Co. St. Louis, MO 2’-Deoxyadenosine (2’-dA) Sigma Chemical Co. St. Louis, MO 3’-Deoxyadenosine (3’-dA) Sigma Chemical Co. St. Louis, MO 2’,3’-Dideoxyadenosine (ddA) Sigma Chemical Co. St. Louis, MO 2’,3’-Dideoxyinosine (ddl) Sigma Chemical Co. St. Louis, MO Cytosine aiabinoside (Ara-C) Sigma Chemical Co. St. Louis, MO Adenosine arabinoside (Ara-A) Sigma Chemical Co. St. Louis, MO Azidothymidine (AZT) Sigma Chemical Co. St. Louis, MO Acyclovir (ACV) Sigma Chemical Co. St. Louis, MO Cidofovir(HPMPC) Gilead Sciences, Inc. Foster City, CA § Ouabain J COON N M i S O , A M M —CM. _ 0 Furosem ide •H O c Benzamil hydrochloride H^henylandiranillc acid so. OH OH s-lp MM f S a iw y n * PUoridzia Chem ical structures o f ion transport inhibitors 91 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ç M fM HOCH HQ» HCOH HÇOH ÇHfiH C H ï O M * ï 5 AbMiHvISribiK 0 , 0 0 , 0 , K O ,!, .««4— < 1 o5 p e*v* ■o Chemical Structures o f M odel Solutes for Drug Transport 92 i Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. HO OH Uridine HO OH® Thynidhe HO HO OH Cytidine O NH, = N HO = N HÔ OH N .N Adenosine Ho' OH NHj N Guanosiie HO OH Inosine N N ^ O Base-Uracil OH Q M C hem ical Structures for Nucleosides and R elated Physiological Substrates 93 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chem ical Structures o f Nucleoside A nalogs R4 Base J R3 R2 R4 R3 R2 Uridine Drug R I B2 B1 BA Base Uracil H - - - Uracil S-Fluorouracil F - - - Uracil Uridine H OH OH OH Uracil 5-Hydro3ty uridine (5-HU) OH OH OH OH Uracil 5-Iodo uridine (5-IU) I OH OH OH Uracil 5-Iodo-2-deo)gruridine (IDU) I H OH OH Uracil Ara-A H t-OH OH OH Adenosine Ara-C H t-OH OH OH Cytidine 2% 3 -dideoxyinosine (ddl) H H H OH Inosine 2’,3’-dideoxyadenosine (ddA) H H H OH Adenosine 2 -deoxyadenosine (2’-dA) H H OH OH Adenosine 3’-deoxyadenosine ( 3 -dA) H OH H OH Adenosine 3 -azido 3 ’-deoxythymidine (AZT) CHs H N3 OH Thymidine NH, I O (S)-HPMPC HO N NH, A c y c l o v i r 94 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IV. RESULTS 9 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A. DEVELOPMENT OF AN AIRWAY EPITHELIAL CELL CULTURE MODEL 1. Primary Cultured Monolayers of Tracheal Epithelial Cells From th e N ew Zealand W hite rabbit thorax region, tracheas (the area below the larynx to below the bifurcation o f m ain stem bronchi) were excised ranging from 1.5 to 2.5 inches in length. The isolation o f epithelial cells by protease treatm ent com bined with w ashing and filtration steps yielded approxim ately 15-25 x 10^ epithelial cells from each anim al, w ith a cell viability >90%, assessed by trypan blue exclusion. Generally, an attachm ent efficiency o f approxim ately 50% was observed after 24 hr, resulting in confluent m onolayers that are clearly visible under an inverted light m icroscope. The overall reproducibility o f the cell isolation and culturing procedure considering anim al to anim al variation, anim al health, stability o f the enzymes and reagents used in the culture procedure, and the uniform ity in Transwell filter inserts, resulted in confluent cell layers w ith a success rate o f ~90%. 2. Morphology o f Airway Epithelial Cultures 2.1 Electron Microscopy Transm ission electron m icrographs (TEM) o f isolated tracheocytes in suspension at a magnification o f7,000x (Fig. 4-1 A ) show a heterogeneous population o f ciliated cells 96 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (C), secretory cells (S) and basal cells (B). C iliated cells show a distinct hair-like projections (cilia) on their apical surface, secretory cells (S) are characterized by a distinctly granular cytoplasm , and basal cells (B) identified by the narrow elongated or spindle cell shape. Transm ission electron m icrograph o f day 5 confluent AIC cell layers under a m agn ification of4,000x exhibited a pseudostratified appearance o f cells arranged in a single layer w ith overlapping cytoplasm ic regions (Fig. 4 -IB ). The same is seen w ith day 5 LCC cell layers except th at the cells appear a little more flattened in nature com pared to AIC (Fig. 4-lE ). A t a m agnification of30,000x under transm ission electron m icroscopy (Fig.4-lC ), intercellular junctional com plexes, in the location marked by the arrow head are observed in A IC cell layers. M icrovilli and cilia are visible on the apical m em brane, along with an abundance o f m itochondria. Scanning electron m icrographs under l,500x m agnification (Fig. 4 -ID) show ed the presence o f cilia on the apical surface o f ciliated cells in AIC cell layers, and less prom inent or denuded cilia at 2,200x m agnification in LCC cell layers (Fig. 4-lE ). 2.2 Periodic Acid Schiff Staining AIC and LCC cell layers were stained w ith periodic acid sch iff to determ ine two issues; the relative percent o f secretory cells in the total cell population, and the difference o f secretory cell population between AIC and LCC. Confluent m onolayers (day 6) o f both AIC and LCC show ed approxim ately 14% positive PAS stain, indicating that glycoprotein secreting cells are present at num bers com parable to that observed in vivo, and a sim ilar population o f secretory cells are present in either culture m odels. 97 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T 3 O Q . C S Q . T 3 (D ( / J W O 3 O (D 8 3. ( Q 3 (D C 3 - 3" (D (D T 3 O Q . C S - . o 3 T 3 O CD Q . O C T 3 CD C / J C / J O 3 Fig. 4-1: Morphological appearance of rabbit tracheal epithelial cells and their monolayers under the electron microscope. Plate A displays isolated rabbit tracheal epithelial cells in suspension (4,000x magnification), bar = 5 pm. The cell types seen are: ciliated cells (c), secretory cells (s), and basal cells (b). Plate B shows an air-interfaced cell monolayer (4,000x magnification), bar = 5 pm. Plate C shows the tight junctions (indicated by the arrowhead) in an air-interfaced cell culture (30,000x magnification), bar = 0.5 pm. Plate D shows a scanning electron micrograph of the ciliary appearance of an air-interfaced cell culture (l,500x magnification), bar = 5 pm. Plate E shows a scanning electron micrograph of liquid-covered cell cultures exhibiting denuded cilia on ciliated cells (2,200x magnification), bar = 5 pm. 3. Bioelectric properties o f Airway Epiritelial Ceil Cultures Plated at a density o f 1.3 x 1 0 ® cells/cm ^ both air-interfaced cultures (AIC) and liquid- covered cultures (LCC) appeared to reach confluence and began to exhibit a m easurable Rt by day 3 onw ards. Following the tim e course o f bioelectric properties o f AIC and LCC, the equivalent current (I^q) peaked at day 5, w hile the R t peaked at 7 days post seeding, and gradually declined from there on and was virtually abolished at day 12 (Fig. 4-2A and B). The peak bioelectric param eters o f AIC and LCC are listed in T able 4-1, with a ll the LCC values being significantly low er than AIC (p<0.05, n=20-24, at least 6 different rabbits). Com pared with LCC, A IC exhibited a 44% higher leq (74.1 ± 6.5 pA/cm^ vs. 51.6 ± 3.5 pA/cm^), and 17% higher PD (70.9 ± 2.8 vs. 60.5 ± 3.03 mV), but 28% low er R t (1.18 ± 0.03 kohms.cm^ vs. 1.51 ± 0.02 kohm.cm^). Cell m onolayers exhibiting PD and R t values greater than — 50 mV and >800 ohms.cm^, respectively, were used for characterization and drug transport studies. 99 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Day 1.5 1 i 0.5 0.0 0 5 10 15 20 Day Fig. 4-2: Bioelectric properties o f air-interfaced cultures (A IC , open squares) and liquid- covered cultures (LCC, solid squares). B ioelectric m easurem ents w ere m ade w ith an EVOM volthohm m eter each day 6 om day 2 to 18. Top panel (A ), is the equivalent short-circuit current (leq) change with culture age. Lower panel (B ) is the R t change w ith culture age. Data points are means ± s.e jn ., n = 21-25. 1 0 0 ) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T 3 O Q . C S Q . ■ D (D ( / J ( / J g Table 4~I: Bioelectric properties of rabbit tracheal epithelial monolayers under air-interface (AIC) and liquid-covered (LCC) conditions CD 8 3 . ( Q O 3 CD C 3 - 3" CD CD T 3 O Q . C g. o 3 T 3 O CD Q . T 3 C / J C / J o' 3 AIC LCC Native Tissue leq (pA/cm^) 74.1 ±6.5 51.6 ±3.5* 81.5 ±6.8" Rt (kohms.cm^) 1.18 ±0.03 1.51 ±0.02* 200 - 700'’ PD(mV) 70.9 ±2.8 60.5 ± 3.03* 12-18" Values are means ± sem, n = 20-25 o * Statistical difference between AIC and LCC, p <0.05 " Taken from reference (Jamigan et ai, 1983) CD b Reference (Welsh, 1987) f 3.1 Selectivity for Substratum (Extracellular Matrix) A irw ay epithelial cells are anchorage dependent cells that exhibit a preference fo r a particular extracellular m atrix. The trend in rabbit tracheocyte selectivity fo r attachm ent to various substrata consisting o f inserts coated with either collagen type 1 (rat tail type 1), fibronectin (human type V ), lam inin o r m atrigel (Biocoat variety pack. C ollaborative B iom edical Products, B edford, MA), was evaluated by m easuring the bioelectric properties o f cell layers. R abbit tracheocytes dem onstrate selectivity for fibronectin and rat tail collagen (type 1). In an air-interface, the leq peaked by day 6 for fibronectin and collagen coated inserts (23-31 |iA/cm^) and R t o f 0.7 - 1.7 kohms.cm^ by day 10. Lam inin coated filters attained a peak leq o f 12 pA/cm^ and R t o f 1.3 kohm s.cm ^ by day 7. C ells grown on M atrigel coated filters did not grow a t all. Under liquid-covered conditions, collagen and fibronectin coated filters developed an leq o f 12-14 pA/cm^ by day 4, w hereas, cells grow n on lam inin developed an leq o f 2.1 pA/cm^ and an R t o f 0.6 kohm cm^. As in the case o f air-interface cultures, M atrigel coating was th e least conducive to cell attachm ent (Table 4-2). In general, the tim e course o f bioelectric developm ent on biocoat filters w ith different substrata, follow a sim ilar trend, albeit, at low er overall values com pared to the Transwell-Col filters (Fig. 4-2A and B ) w ith the peak bioelectric properties reached betw een day 4-7. 1 0 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T 3 O Q . C S Q . T3 (D ( / J W O 3 O CD 8 T3 V C 3 . ( Q O 3 CD C 3. 3 " CD i T3 O Q . C a o 3 T3 O CD Q . O C T3 CD C / J C / J O 3 Table 4-2: Effect of Substrata on airway epithelial leq (pA/cm^) and Rt (kohm.cm^) Values are means ± s.e.m., n = 2-3 " represents the day of reaching peak leq and Rt, respectively. Matrigel values were similar to that of blank filter inserts AIC LCC leq Rt Day" Icq Rt Day" Collagen 23.5 ±7.1 0.7 ± 0.3 6,10 11.6 ±4.6 2.07 ±0.7 4.5 Fibronectin 31.4 ±5.6 1.74 ±0.8 6,6 13.9 ±58 3.23 ±0.8 4,4 Laminin 12.2 ±8.1 1.28 ±0.7 7,7 2.1 0.6 4,5 Matrigel - - - - - - S 3.2 Effect o f Fetal Bovine Serum (FBS) Anchorage dependent cells such as airw ay epithelial cells often exhibit greater differentiation and peak bioelectric properties when grow n in defined, low -protein, serum -free m edia. The airw ay epithelial cellular response to serum -containing m edium were exam ined by estim ating the bioelectric parameters o f cell layers grow n in the presence o f 0.5, 1 and 5% FBS. The effect o f FBS on airw ay bioelectric properties was more noticeable in AIC than LCC (Table 4-3). At high FBS concentrations, the leq in day 6 or 7 AIC w as stim ulated by — 40% , while the R t was lower by 20-25% com pared to low-serum treated cells. In LCC, no noticeable change in leq was evident w ith increasing FBS concentrations, how ever, R t was lowered by 20-25%. 104 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 0 C D ■ D O Q . C O C D Q . 3 " "D C D C / J W O 3 O 5 C D O O "D 3 . C Q 3" 3 C D 3 " C D C D "D O Q . C S-. o 3 "D O 3" g C D Q . 3 " O c ■ D C D C / J W O 3 TaMe 4-3: Effect of FBS on airway epithelial leq and Rt %FBS AIC LCC leq Rt Day* leq Rt Day" 0.5% 75.5 ±15 0.75 ±0.1 6,6 47.7 ±12.1 1.304 ±0.3 6,7 1 75.7 ±12 0.82 ±0.2 6,7 46.3 0.990 6,7 5 111.6±18 0.566 ±0.1 6,7 43.7 ±11.5 1.03 ±0.2 6,7 Values are means ± s.e.m., n = 2-3 ‘ Values represent the day of reaching peak leq and Rt, respectively. g B. ELECTROPHYSIOLOGICAL AND BIOCHEMICAL CHARACETRIZATION The io n transport characteristics were exam ined to tm derstand w hether AIC and LCC w ere absorptive or secretory in nature, and to validate their resem blance to the intact tissue. Confluent tight m onolayers m ounted in m odifled-U ssing cham bers developed a stable short-circuit current (Isc) and resistance (Rt) for m ore than 240 m in under baseline conditions. Typically, a steady-state baseline Isc w as reached w ithin a 20 m in equilibration period. A ll data are presented as a change from baseline values, prior to treatm ent w ith an ion transport inhibitor. 1. Effect of Ouabain The presence and localization o f the N a ATPase pum p was determ ined in the presence o f 10 pM ouabain added either to the apical o f basolateral side o f the epithelium (Fig. 4-3A and B). A pical treatm ent o f ouabain did not alter epithelial bioelectric properties in both AIC and LCC. By contrast, basolateral treatm ent in AIC (Panel A) virtually abolished the Isc (91.6 ± 1.2 % ) w ith a concurrent gradual increase in R t (48.8 ± 26%). The half-tim e for m axim al inhibition (ti/2) w as 21 m in. In LCC, basolateral 10 pM ouabain treatm ent (Panel B) reduced Isc by 90.8 ± 0.9% w ith a 20.7 ± 6.2% increase in Rt. The ti/2 for ouabain effect in LCC was 22 m in (Table 4-4). 1 0 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 200-1 150- %R I 100 50 OuatMin->i % l#c 30 50 70 -10 10 Time (min) 2001 Ouabaln->B ISO & %R 100 50 % I8C 50 70 10 30 -10 B Time (min) Fig. 4-3: Effect o f ouabain (10 pM ) on A IC (top panel. A) or LCC (low er panel, B) Isc and Rt. M onolayers were m ounted in m odified Ussing cham bers. Following stabilization o f baseline bioelectric properties, ouabain was first added to the apical then the basolateral com partm ent. D ata is reported as mean change for baseline Isc and R t values. Mean ± s.em ., n = 3-5. 107 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2. E f ^ t ofAmiloride and Furosemide The presence o f Na^-conductive pathways w as studied w ith am iloride, an inhibitor o f the Na^-channel (Benos, 1982). The presence o f Cl -entry pathways via the N a^(IC O K ^ symport, was verified by treatm ent with the loop diruretic furosem ide. Sequential treatm ent w ith these two inhibitors was studied to evaluate the relative contributions o f Na^ and C l' entry pathways to the overall ion transport o f airway epithelial cells. Apical treatm ent o f 10 pM am iloride in AIC (Fig. 4-4A ) inhibited Isc by 68.2 ± 5.1% w ith a ti/2 o f about 1 m in. A sim ultaneous increase in R t o f 67.7 ± 32% was observed. Basolateral am iloride treatm ent did not change baseline bioelectric properties. Following am iloride-sensitive Na^-current inhibition, addition o f 10 pM furosem ide to the basolateral side resulted in a S-10% decrease o f baseline current and corresponding 10% increase in Rt. In LCC, 10 pM amiloride treatm ent on the apical side yielded SO ± 5.3% inhibition o f Isc w ith a t ,/2 o f 1 min, and 26.6 ± 7.5% increase in R t (Fig. 4-4B) . In native rabbit tracheal epithelium , it has been reported that am iloride m ay induce C l'-secretion (Boucher and Gatzy, 1983). To verify the relative extents o f amiloride and furosem ide-sensitive Isc (from Fig. 4A and 4B) and rule out possibility o f amiloride inducing Cl'-secretion, cell m onolayers were treated first w ith furosem ide followed by am iloride. Apical furosemide treatm ent did not affect baseline bioelectric properties in either AIC or LCC. In AIC, basolateral furosemide treatm ent inhibited Isc by about 10 ± 4.0% w ith a txn o f 3 min, w ith no corresponding change in R * (Fig. 4-5A). Following stabilization o f the furosem ide-insensitive current, apical am iloride treatm ent 108 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. inhibited baseline Isc by an additional 68.4 ± 2.6% with a sim ultaneous 16 ± 13.1% increase in R t By contrast in LCC, furosem ide inhibited Isc b y 45.5 ± 8.02% w ith a ti/2 o f 3 min, accompanied by a transient increase in R t by about 10%. Follow ing stabilization, apical am iloride inhibited baseline Isc by an additional 30 ± 6.7% w ith no appreciable change in Rt (Fig. 4-5B ). 109 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 200- %R Amiloride->l 150- I < - Amiloride->A Furosemide>>B 100 SO SO 30 10 -10 Time (min) B ISO Amiloride->B %R & 100 Amiioride->A Furoeemlde->B I SO %lsc 30 SO -10 10 Time (min) Fig. 4-4: Effect o f am iloride (10 nM ) follow ed by furosem ide (10 fiM) on AIC (top panel. A) or LCC (low er panel, B) Isc and R t. M onolayers were m ounted in m odified U ssing cham bers. A m iloride was first added to the basolateral com partm ent and later to the apical com partm ent. Following stabilization o f Isc, fiirosem ide was added to the basolateral com partm ent. D ata is reported as m ean change for baseline values. M ean ± s.e m ., n = 3-5. 1 1 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 150 1 Furosemide->A 125 ■ %R 100 Amiloride->A & I 7 5 - u Furosemi<fe->B 50 25 - % l8C 30 50 10 •10 Time (min) 150-1 Amiiorlde->A Furo8emfde->A 100 I %R « 75 50 Furosemide->B 25 % l8C 30 50 10 •10 B Time (min) Fig. 4-5: Effect o f furosem ide (10 nM ) follow ed by amiloride (10 jiM ) o n AIC (top panel. A) or LCC (lower panel, B ) Isc and Rt. M onolayers were m ounted in m odified U ssing chambers. Furosem ide w as first added to the apical com partm ent and later to the basolateral compartment. Follow ing stabilization o f Isc, am iloride was added to the apical compartment. Data is reported as m ean change for baseline values. M ean ± s.em ., n = 3-5. I l l i Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3. Effect of BenzamU To confinn the predom inant am iloride-sensitive Na^ current in AIC, the relative extent o f inhibition w ith apical benzam il, a lipophilic (benzyl) analog o f amiloride was studied. Apical treatm ent w ith 10 pM benzam il inhibited Isc in AIC by 70 ± 4% w ith a ti/2 o f about 1 m in (Fig. 4-6A ). This was accom panied by a 16.6 ± 7.8% increase in Rt. In LCC, benzam il resulted in a 47.9 ± 7% inhibition in baseline Isc with a t,/2 o f about 1 min, and a corresponding increase in R t o f 20.9 ± 4.6% (Fig. 4-6B). 1 1 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 125- %R 100 & 75 50 25 %lsc 20 10 0 •10 Time (min) 125 %R 100 Î 75 o 50 % I8C 25 Benzamii->A 10 20 •10 0 B Time (min) Fig. 4-6: Effect o f benzam il (10 ^M ) on AIC (top panel. A) or LCC (low er panel, B) Isc and Rt. M onolayers were m ounted in m odified U ssing chambers. Benzam il was added to the apical suiface o f cells and Isc and R t m onitored over time. D ata is reported as mean change for baseline values. M ean ± s.e.m ., n = 3-5. 113 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. i 4. Effect o f N-Phenylanthranilic Acid The presence o f C l -conductive pathw ays w as exam ined w ith N -phenylanthranilic acid (NPAA), an inhibitor o f chloride channels in airway epithelial cells. As show n in Fig. 4- 7A and B, 0.5 m M N PA A apical treatm ent caused a biphasic decrease in Isc in A IC and LCC. The first response w as im m ediate and short-lived w ith a Xxn o f 1 m in, follow ed by a gradual decrease w ith a Xxn o f about 16 m in. In AIC, the fast response resulted in a 6.9 ± 2.3% decrease in Isc, follow ed by slow er inhibition o f about 43.3 ± 4.1% (Fig. 4-7 A). In LCC, the fast response yielded a 13 ± 5.1% inhibition follow ed by a 50.1 ± 3.7% decrease in baseline Isc (Fig. 4-7B). The increase in R t on prolonged exposure w as 27- 30% in AIC and LCC. 114 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 125- % R 100 I 75 u 50 % lsc NPAA->A 25 30 40 10 20 -10 0 Time (min) B 125 % R 10O & 75 u 50 % Isc 25 10 20 30 40 50 -10 0 Time (min) Fig. 4-7: Effect o f N-phenylanthranilic acid (NPAA, 0.5 mM) on AIC (top panel. A) or LCC (low er panel, B) Isc and R t. M onolayers were m ounted in m odified U ssing chambers. NPAA was added to the apical surface o f cells and Isc and R t m onitored over tim e. Data is reported as m ean change for baseline values. M ean ± s.e m ., n = 3-5. 115 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5. Effect of4,4*-DiisothiocyanotosWbene-2,2'disulfonic acid (DIDS) Epithelial cells are equipped with pH regulators such as CI /HCO3 exchanger or the Na^-HCO3 cotransporter. To test the possible existence o f these pH regulators in prim ary cultured airw ay epithelial cells, w e used the stilbene analog, DIDS to inhibit the bicarbonate-related current. In both AIC and LCC, apical treatm ent with 50 pM DIDS did not alter baseline bioelectric current. However, on basolateral addition, Isc was transiently stim ulated in A IC by 27.8 ± 5.8% , follow ed by a drop &om the baseline Isc by 12 ± 4.3% (Fig. 4-8A ), with a t,/2 o f 21 m in. Likewise in LCC, Isc transiently increased by 14.3 ± 6 % w ithin 15 min, follow ed by a drop in Isc by 15% below the baseline Isc (Fig. 4-8B). The Rt decreased by 14-25% w ith DIDS treatm ent (Table 4-4) Table 4-4, sum m arizes the response tim es for change in R * for each drug treatm ent and the tim e taken for reaching h a lf m aximal Isc response (ti/2). Shorter t (/2 as seen for am iloride, benzam il, and furosem ide suggests direct channel or transporter inhibition, and longer t ,/2 for ouabain and DIDS indicates a possible indirect effect due to cellular changes, such as pump inhibition or alteration o f intracellular pH. Increase in Rt occurs in m ost cases, due prim arily to a blockage in the conductive ion transport pathways. 116 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1251 %lsc 100 I %R 7 5 - DIDS->A 50 DiDS->B 2 5 - 10 20 40 30 0 -10 Time (min) B 125 %!sc 100 & 75 S %R DIOS->A 50 DIDS->B 25 20 30 0 10 -10 Time (min) Fig. 4-8: Effect o f 4 ,4 ’-diisothiocyanotostilbene-2,2’-<iisulfonic acid (DIDS, 50 ^xM) on AIC (top panel. A) o r LCC (low er panel, B) Isc and Rt. M onolayers were m ounted in m odified Ussing cham bers. DIDS was added to the apical surface follow ed by basolateral surface o f cells and Isc and R t m onitored over tim e. Data is reported as mean change for baseline values. M ean ± s.e.m., n = 3-5. 117 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T 3 O Q. C s Q. T3 ( D W W o' 3 O CD 8 T3 V C 3 . ( Q 3 " i 3 CD ■ n c 3. 3 " CD i T3 O Q. C S - . o 3 T3 O CD Q. O C T3 CD C / J C / J o' 3 Table 4-4: Half Time for Maximal Isc Response and Corresponding Rt Changes Treatment t , / 2 (min)" Rt (% change)*’ AIC LCC AIC LCC Ouabain (10 pM) 21 22 48.8 ±26 20.7 ±6.2 Amiloride (10 pM) 1 1 67.7 ±32 26.6 ±7.5 Benzamil (10 pM) 2 2 16.6 ±7.8 20.9 ±4.6 Furosemide (10 pM) 3 3 3.5 ±3.8' 10.5 ±10.9' NPAA (0.5 mM) 16 16 27 ±6.4 29.5 ±6.7 DIDS (50 pM) 21 15 -14.2 ±1.5 -25.2 ±1.4 ‘ Observed t | / 2 for mean Isc plots Means ± s.e.m., n =3-4. ® No change 0 0 I 6. Biochemical Characterization The am ount o f lactate (a product o f anaerobic respiration) produced and secreted by cells into the bathing m edium as an indicator o f the nature o f the cellular respiration was tested in AIC and LCC. The tim e course o f lactate release into the apical and basolateral fluid for AIC and LCC is show n in Fig. 4-9A. N orm alized for the m onolayer protein content the am ount o f lactate released by AIC and LCC increased linearly w ith tim e up to 36 hr. The overall rate o f lactate appearance in AIC was 0.27 ± 0.01 mg lactate/m g protein/hr, about 57% the rate in LCC, 0.47 ± 0.03 mg lactate/m g protein/hr (Table 4-5). A fter 48 hr, analysis o f the lactate content from apical w ashing o f AIC and apical fluid from LCC (Fig. 4-9B) revealed that ju st 0.8 ± 0.1% o f the total lactate secreted by AIC is localized in the apical m edium . W ith LCC, 29.9 ± 2.1% o f the total lactate secreted selectively to the apical m edium . 119 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 40 30 20 10 40 50 30 20 Time (hr) B 100 o 40 AIC LCC Fig. 4-9: Lactate release in tracheal cultures. Top panel (A ), tim e course o f total lactate release in AIC (open squares) and LCC (solid squares). Low er panel (B), localization o f lactate release to the apical (solid bars) or basolateral (hatched bars) com partm ents. M eans ± s.e.m ., n = 3. * Statistical difference betw een apical and basolateral lactate release, p < 0.05. 1 2 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T 3 O Q . C S Q . T3 ( D W W O 3 O ( D 8 T3 V C 3 . ( Q 3 " i 3 Ç D ■ n c 3. 3 " CD i T3 O Q . C S - . o 3 T3 O Table 4-5: Lactate release rate (mg lactate/mg protein/ hr) in AIC and LCC Rate of Release AIC LCC Ratio (AIC:LCC) p-value" Total (apical + basolateral) 0.27 ±0.012 0.47 ±0.03 0.57 <0.001 Basolateral fluid 0.25 ±0.01 0.30 ±0.02 0.83 <0.02 Apical fluid nd 0.16 ±0.01** CD Q. T3 CD * Statistics for significant difference between AIC and LCC values. ^ Statistically significant difference between apical and basal lactate rate in LCC, p < 0.001. Values are means ± s.e.m., n = 3. nd - not determined. ( / ) o' D N ) C. PERMEABILITY CHARACTERIZATION 1. Transport Characteristies o f Model Solutes in Air~interfaced (AIC) and Uquid^overed Cultures (LCC) T he perm eability characteristics o f the cultured airway epithelial cells was evaluated for th e first time using m odel hydrophilic (m annitol, sucrose and albuterol) and lipophilic (dexam ethasone and propranolol) solutes. The unidirectional (apical to basolateral) fluxes o f ^H-mannitol and ‘^C-sucrose w as linear throughout the duration o f the experim ent (Figure 4-lO A and 4-lO B , inset), w ith a lag tim e o f about 5-10 m in. While the flux o f the lipophilic dexam ethasone and propranolol, displayed curvilinearity, probably due to the collapse o f sin k conditions. As shown in Table 4-6, the Papp o f hydrophilic solutes w as about tw o orders o f m % nitude low er than lipophilic solutes. W hile both AIC and LCC are equally perm eable to the lipophilic dexam ethasone and propranolol, the LCC w as slightly m ore perm eable to hydrophilic solutes (p<0.05) than th e AIC. In AIC, the R t either increased a m inor extent or rem ained unchanged, for controls and m ost drug treated m onolayers. However, in LC C , a higher sensitivity to experim ental m anipulation through the 240 m in experim ent tim e course was noted for both control buffer as w ell as drug treated m onolayers. A ll th e drugs affected the in te g ri^ o f the LCC m onolayers b y reducing R^, with the beta adrenergic agents albuterol and propranolol eliciting th e m axim um response o f 37% drop. 122 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I I 03 45 02 o i 30 O O IC O 200 300 15 0 200 100 0 B 1 I 0 6 60 0 4 45 02 OO, 100 200 300 30 15 0 100 200 0 Time (min) Fig. 4-10: Tim e course o f solute transport across AIC (top panel. A) and LCC (low er panel, B). Inset enlarges the tim e course o f hydrophilic solute transport. K ey: m annitol (solid squares), sucrose (open squares), albuterol (open triangles), dexam ethasone (open circles), and propranolol (x). D ata points are m eans ± s.e m ., n = 3-6. 123 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I Table 4-6: Perm eability coefficients (Papp, x 10*^ cm /sec) o f model solutes across the AIC and LCC Drug log?" P app AIC * * (x 10" * cm /sec) L C C Mannitol -3.1 0.12 ± 0 .0 3 0.28 ± 0 .0 6 Sucrose 0.92 0.03 ± 0 .0 0 2 0.05 ± 0.003 Albuterol 0.11 0.22 ± 0.03 0.46 ± 0.42 Dexamethasone 1.83 9.8 ± 0.4 8.7 ± 0.3 Propranolol 3.2 31.7 ± 8 .0 23.8 ± 3.0 log octanol/pH 7.4 buffer p artition coefficient * * M ean Papp ± s.e m ., n = 3-6. Significant difference betw een Papp in AIC and LCC (p<0.05). 124 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2. Influence o f Solute Molecular Size FITC-Dextran Transport The transepithelial perm eability o f solutes based on m olecular size was studied using 0.5 mg/mi FITC-iabeied dextrans (FD ’s average m olecular w eight 4,000 - 70,000 daltons). The cum ulative appearance o f fluorescence in the receiver fluid was linear for the 4-br duration o f the experiment (Fig. 4-11). The rate o f transport decreased w ith m olecular size w ith FITC-label (MW 389) alone, exhibiting the highest perm eability. The transport o f the reference solute m annitol (Fig. 4-11 inset) and FITC-dextrans (Fig. 4-11) in the AB direction was not significantly difTerent firom th at in the BA direction (Table 4-7, p > 0.05). The symmetric paracellular transport w as verified by lowering the experim ental tem perature to 4°C. FD-4 exhibited no significant decrease in perm eability at 4°C com pared to 37°C with Papp’s o f 2.5 Qt 0.9) x 10'* cm /sec and 2.8 (± 0.4) x 10"*, respectively. In all cases, the bioelectric properties (PD — 50 m V , — 1,300 ohms.cm^) o f the FITC-treated monolayers rem ained sim ilar to that o f buffer treated control m onolayers, throughout the duration o f these experim ents. The extent o f metabolism o f FITC-dextrans was evaluated by HPLC analysis o f the receiver fluid was performed using a W aters Protein-Pak 300SW gel perm eation colum n w ith a m olecular weight fraction range from 10,000 - 300,000. Analysis o f the collected fractions revealed that the m ajority o f the fluorescence eluted in the void volum e w ith large m olecular w eight dextrans. In general, 3-8% o f the transported FITC- 125 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dextran was degraded to sm aller m olecular w eight species regardless o f transport direction (Fig. 4-12). Fig. 4-12A and 4-12C show degradation o f FD -20 and FD-70 in the AB direction, respectively, w hile Fig. 4-12B illustrates the apical appearance o f FD- 40 tested in the BA direction. N o degradation o f the dosing solution w as observed for all three dextrans tested after a 4 h r transport experim ent. A t a constant concentration o f 0.5 m g/m l, the Papp decreased sharply com m ensurate w ith increased m olecular w eight up to 2 0 ,0 0 0 daltons, rem aining relatively constant thereafter (Fig. 4-13). W hen plotted using the sam e approach as adopted by Renkin (Eq. 4), w ith m annitol as the reference solute, nonlinear least square curve fitting yielded a relationship consistent w ith a single pore population w ith an equivalent pore-radius o f about 4.98 run (Fig. 4-14) (r^ = 0.97, p < 0.01). The estim ated ratio o f total pore area to pore length (Ap/dx) is 0.026 cm. A ssum ing the pore length or thickness o f the barrier, dx is 10 pm (approxim ated firom electron m icroscopy), the calculated total num ber o f pores, N» is about 3.5 x 10*/cm^. 1 2 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4-7: M olecular W eight (M W ), Size, D ifiusivity (D, x 10^ cm^/sec) and Perm eability C oefficient (Papp) o f FITC-dextrans (0.5 m g/m l) in the Apical-to- Basolateral (A B) and Basolateral-to-A pical (BA) direction in A ir-interface Cultured Rabbit A irw ay Epithelial Cell M onolayers. Drug MW S ize(nm ) D Papp (x AB 10* * cm/sec)“ BA Mannitol 182 0.43 6.82 11.8 8.5 (±3.0) (±0.5) FITC 389 0 .6 8 4.84 11.2 12.1 (± 1.1) (±2 .1) FD-4 4,400 1.44 2.28 2.77 5.40 (±0.4) (± 1.6) FD-10 9,800 1.91 1.72 1.86 1.75 (±0 .6 ) (±0.4) FD-20 18,900 2.44 1.34 0.36 0.44 (±0 .0 1 ) (±0 .02) FD-40 35,600 3.15 1.04 0.32 0.50 (±0 .0 2 ) (±0.05) FD-70 71,200 3.81 0 .8 6 0.30 1.01 (±0.03) (±0 .1) “ Mean ± s.e m . for n = 3-6. 127 t Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6000 1^4000 1200C 1000 750 100 200 300 I 500 P 250 200 100 0 Time (min) Fig. 4-11: Tim e course o f FITC-dextrans transport across air-interfaced tracheal epithelial cell cultures. Inset shows the tim e course o f ^H-m annitoI transport. Key: FITC (solid circles), FD-4 (open circles), FD-10 (solid triangles), FD-20 (open triangles), FD- 40 (solid squares), FD-70 (open squares). D ata points are m eans s.e m ., n = 3-6. 1 2 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 10 20 30 Fraction* D onor Boc olvor 1000 - I 600 • 200 30 10 20 40 0 10 0 6 • 4 ; 2 I 0 B D o n o r R ocotvor Ftadkwi# 500 - 400 - 300 S 2 0 0 - 4 e 100 10 20 Fraction* 30 40 0 D onor Raeofvor Fig. 4-12: Size-exciusion chrom atography o f pooled receiver com partm ent sam ples (triangles) or dosing solutions (circles) after 4 hr transport. Apical dosing o f FD -20 (top panel. A), basolateral dosing o f FD -40 (m iddle panel, B) and apical dosing o f FD -70 (low er panel, C). 129 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I 12 9 6 3 0 20000 40000 60000 80000 0 M W (daltons) Fig. 4-13: R elationship betw een apparent perm eability coefficient (Papp) and m olecular weight (M W ) o f m annitol, FITC, FD -4, FD-10, FD -20, FD -40 and FD-70 in air- interfaced cultures. E rror bars are sm aller than the size o f the sym bols, n = 4-6. 130 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.2 1.0 0.8 0.6 0.4 0.2 0.0 4 5 2 3 0 1 Solute Radii (nm) Fig. 4-14: Equivalent pore-radius estim ation o f the Papp o f each solute norm alized to that o f the reference solute m annitol, as a function o f solute radius. 131 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3. influence o f Solute UpophiUcity ft-Adrenergic Compound Transport The role o f solute lipophilicity was exam ined by evaluating the transport characteristics o f a series o f fi-adrenergic agents that vary in the log octanol/pH 7.4 buffer partition coefBcient (log P) by about 4 log units. T he tim e course o f the appearance o f the drug in the receiver com partm ent was linear for com pounds w ith log P < 1.9 for the entire 4 hr period, whereas for those w ith log P > 1.9 transport gradually approached a plateau after 120 min (Fig. 4-15). A lag tim e o f <10 m in w as observed for hydrophilic B-adrenergic compounds, w hereas no appreciable lag tim e was observed for the lipophilic ones. HPLC analysis o f the transport sam ples revealed no evidence o f m etabolite form ation for B-adrenergic compounds. The Papp’s ranged from 1.7 x 10'^ cm/sec for atenolol to 287 X 10'^ cm /sec for propranolol, about a 170-fold difference (Table 4-8). The trend in the influence o f increasing solute log P on the airw ay epithelial Papp, is best described by a sigm oidal relationship (Fig. 4-16) (r^ = 0.99, p < 0.01). From the nonlinear least square curve fitting, the param eter estim ates o f the m inim um Papp "a" w as 2.22 ± 0.57 (x 10*^ cm /sec), the maximum Papp "b" w as 284.85 ± 1.14 (x 10^ cm /sec), the log P at the half-maximal Papp "c" was 2.08 ± 0.002, and the transitional slope "d" representing the gradual increase in Papp between "a" and "b" was 0.13 ± 0.002. The values represent predicted perm eability’s ± standard error o f the param eter estim ates are listed in Table 5-2. 132 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 15 1 10 ■ 5 - 100 Tim e (m in) 200 Fig. 4-15: Tim e course o f fi-adrenergic agent transport across airway epithelial cell cultures. Key in increasing order o f lipophilicity: sotalol (solid triangles), albuterol (open triangles), atenolol ( solid circles), tim olol (open circles), betaxolol (solid squares), propranolol (open squares). M eans ± s.e m ., n = 4-6. 133 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 300 1 4 ■ 200 ' o X I 100 ' 4 5 3 0 1 2 2 1 Log P Fig. 4-16: R elationship betw een apparent perm eability coefficient (Papp) and log octanol/pH 7.4 bu ffer partition coefficient (log P). K ey in increasing order o f lipophilicity: 1-sotalol, 2-albuterol, 3-atenolol, 4-m etoprolol, 5-tim olol, 6-betaxolol, 7- propranolol. 134 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T 3 O Q. C S Q . T3 (D W 1 I Table 4-8: o CD O I 5 I 3. I ( Q 3 CD CD 5 T3 O Q . C g . O 3 T3 O CD Q . O C T3 CD e n C / J O 3 Permeability (Papp, x 1 0 * ^ cm/sec)* of the Cultured Tracheal and Alveolar Epithelial Cell Monolayers to 0-Adrenergic Compounds of Varying Lipophilicity (log P). Drug LogP Papp (Trachea) Papp (A lveolar)** Sotalol -0.62 3.4 1.6 (±0.2) (±0.9) Albuterol 0.11 1.6 n r® (±0.2) Atenolol 0.16 1.7 2.0 (±0.2) (±0.3) Metoprolol 1.88 249.0 152.0 (±12.3) (±11.3) Timolol 1.91 63.2 194.0 (±1.3) (±22.1) Betaxolol 2.17 188.0 298.0 (±11.0) (±20.5) Propranolol 3.21 287.0 291.0 (±37.8) (±16.8) C ; Mean ± s.e,m .(n = 4-6). From Saha e /a /., 1994. Value not reported. L A D. EVIDENCE FOR CARRIER-MEDIATED NUCLEOSIDE UPTAKE The next stage o f the project the air-interface airw ay epithelial culture m odel was applied to understand physiological and therapeutically active nucleoside analog perm eation into the airway epithelium . The following points are addressed: (1) evidence for uridine influx into airw ay epithelial cells, (2 ) dependence on Na^ and the Na^ruridine stoichiom etry, (4) sensitivity to inhibitors: ouabain, phloridzin, and nitrobenzylthioinosine (NBTI), (5) uridine saturability, (6 ) inhibition w ith physiological substrates, (7) substrate selectivity w ith antiviral nucleoside analogs, (8) interaction with a known substrate (2’-deoxyadenosine), or non-substrate (HPM PC), (9) functional identification o f the transporters, ( 10) transepithelial transport and basolateral localization o f nucleoside transporters, and (11) effect o f Ad5 on nucleoside transport function. 1. Metabolism o f Nucleosides TLC analysis o f cellular radioactivity after one m inute uptake revealed that m ore than 90% o f the radioactivity coeluted w ith intact uridine, and less than <5% w ith the nucleobase uracil and uridine phosphorylated species (Fig. 4-17A). On prolonged exposure (60 m in), about 70% o f the radioactivity appears to be intact uridine w ith about 10% degraded to uracil and uridine nucleotides, respectively (Fig 4-17B). 136 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Therefore, in airway epithelial cells m etabolism o f uridine does not influence estim ation o f uptake over the duration o f most experim ents and th eir data interpretation. Uridine 1 2 0 0 - 800 I Nucleotides 400 Uracil 15 20 10 5 0 Fraction # Fig. 4-17: Enzymatic stability o f ^H-uridine (10 jiC i/m l) in cells w ithin 1 min. 137 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2. Uridine Influx Into airway epidiellal cells The tim e course o f initial uridine uptake (10 pCi/m l) by airw ay epithelial cells was rapid w hen introduced to the apical surface o f confluent cell m onolayers (Fig. 4-18). The initial uptake rate w as 33 ± 4 finoles/m g protein/sec. W ith basolateral dosing, uridine (10 pC i/m l) uptake (across th e basolateral m em brane) was linear fo r the 120 sec, w ith the uptake rate 4-tim es low er (7.8 ± 2.2 finoles/m g protein/sec) than th at w ith apical dosing (Table 4-9). This reflects a directional preference for lum inal-to- cytoplasm ic transfer o f uridine in the airw ay epithelium (Fig. 4-18). Influx o f uridine through the apical m em brane was strongly dependent on tem perature. Low ering o f the experim ental tem perature to 4°C yielded uptake value at 30 sec that is <10% o f the uptake at 37°C. The overall uptake rate w as reduced by 10-fold, to 3 ± 0.5 finoles/m g protein/sec (Table 4-9). 3. Na*"dependence o f Uridine Uptake U ridine uptake in Na^-containing bicarbonated Ringer’s solution (BRS) w as about 30 tim es higher than uptake in Na^-fi%e R inger’s solution (SFR). Replacem ent o f Na^ with an equim olar concentration o f choline^ yielded a 30 sec uridine uptake that was 7% o f that in Na^-containing BRS (Fig. 4-18). The uptake rate was 33 ± 4.0 finoles/m g protein/sec in BRS w hile that in SFR was 1.0 ± 0.3 finoles/m g protein/sec (Table 4-9). To determ ine the stoichiom etry o f N a^: nucleoside coupling, uridine uptake was studied in the presence o f varying concentrations o f Na^ (0-140 m M ) m aintained iso-osm olar 138 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. w ith choline chloride (Fig. 4-19). U ridine uptake increased w ith increasing concentration o fN a \ N on-linear least square curve fitting o f the data yielded a H ill coefS cient o f 1.02 ± 0.4 and a o f 14.8 ± 6 .8 mM. This result indicates th at one Na"" ion is coupled w ith the binding o f one uridine to the nucleoside transporter. 4.0 3.0 a. O ) 2.0 I 1.0 0.0 100 150 50 0 Time (seconds) Fig. 4-18: Uptake o f ^H-uridine (10 pC i/m l) in rabbit airw ay epithelial cells firom the apical surface in the presence (solid squares) o r absence (open squares) ofN a% at 4°C (triangles), and from the basolateral surface (circles). D ata points are m eans ± s.e m ., n = 3-5. 139 I Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.5 M m 2 « ■ 1.0 f I 0.5 g 0.0 120 80 40 0 Sodium Cone. (mM) Fig. 4-19: U ptake o f ^H-uridine (10 pCi/m l) as a function o f Na^ concentration (0-140 mM ) D ata points were fitted to the H ill equation yielding a K^a o f 14.8 mM and a Hill coefficient o f 1.02. D ata points are m eans ± s.e m ., n=3. 140 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4. Sensitivity to Inhibitors To ascertain whether uridine uptake was m ediated by a carrier-m ediated process, the effect o f the Na^K^ATPase pum p inhibitor, ouabain and the N a^-dependent sugar transport inhibitor, phloridzin was tested. In addition, the effect o f ion transport inhibitors, amiloride, fiirosem ide, DIDS, and the equilibrative nucleoside transport inhibitor, nitrobenzylthioinosine (NBTI) on uridine uptake was also evaluated. Sum m arized in Table 4-10, basolateral pretreatm ent w ith 10 pM ouabain for 45 m in and apical pretreatm ent w ith 100 pM phloridzin fo r 20 m in, both abolished uridine uptake by 75%. Apical pretreatm ent w ith 10 pM am iloride did not significantly affect uridine uptake (p>0.05). By contrast, basolateral fiirosem ide (10 pM ) pretreatm ent stim ulated uptake (~80% ). The reason for this in unknow n but it may involve the indirect stim ulation o f Na^-influx processes (including the Na^-coupled nucleoside transport) in order to maintain the hom eostasis o f cellular Na^. Further studies are required to verify th is hypothesis, however, as this was not the focus o f the study additional experim ents w ere not designed on these lines. Addition o f DIDS, resulted in very w eak (— 20%) inhibition o f uridine uptake, probably due to non-specific actions on intracellular pH. C ells pretreated w ith 10 pM NBTI for 20 m inutes show ed no significant inhibition in ^H-uridine uptake (Table 4-9). 141 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4~9: Uptake rate of ^H-uridine from the apical or basolateral cell membrane. T re a tm e n t U ptake R ate* (fm oies/m g p ro tein /sec) C ontrol apical uptake (N a^-containing buffer) 33 ± 4 B asolateral uptake (N a^-containing buffer) 7 i0 .5 ® 4®C uptake 3 ±0.1*’ N a^-free uptake 1 ± 0 .3 ‘ ’ N B T I inhibition'’ 34 ± 4 * M ean uptake rate ± s.e.m ., n = 3-5. * * C ells w ere pretreated w ith 10 pM N BTI for 15 min, prior to addition o f ^H -uridine (10 pC i/m l) and uptake m easured fo r 30 sec. Significant difference, p<0.05, by Student’s t-te st 142 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4^10: Effect of ion transport inhibitors on apical H-uridine uptake Inhibitor Cone. (HM) Uptake* (pmole/mg protein/30s) Control (BRS) 1.01 ± 0 .12 Ouabain* 10 0.24 ±0.01** Phloridzin'’ 100 0.25 ±0.05** Amiloride'* 10 1.21 ± 0.13 Furosemide*' 10 1.8 ±0.17** DIDS*' 50 0.79 ± 0 .1 2 M ean ± s.e m ., n = 3-6. ^ C ells w ere pretreated w ith 10 pM ouabain on the basolateral side for 45 min, before dosing w ith apical ^H-uridine (10 pCi/m l) for 30 sec. C ells w ere pretreated w ith 100 pM phloridzin on the apical side for 15 min, prior to dosing w ith H-uridine (10 pC i/m l) for 30 sec. C ells w ere pretreated w ith 10 pM fiirosem ide o r 50 pM DIDS on the basolateral side for 15 m in, prior to dosing w ith ^H-uridine (10 pCi/m l) for 30 sec. Significant difference from control estim ated by Student’s t-test, p<0.05 143 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5. Concentration Dependence The saturability o f the uptake process w as studied to determ ine the kinetic param eters o f ^H-uridine uptake in airw ay epithelial cells. The uptake in cell m onolayers was m easured at 30 seconds in the presence and absence o f a N a^-gradient. Fig. 4-20 shows that uridine uptake in the presence o f Na^ follows sim ple M ichaelis-M enton kinetics that can be resolved into tw o com ponents: a linear and a saturable component, determ ined by subtracting the linear, Na^-independent uptake from the total uptake. N onlinear least square curve-fitting on TableCurve program revealed that the saturable uridine component has a Km o f 3.4 ± 1 .8 pM and a V m ax o f 24.3 ± 5.2 pm oles/m g protein/30s. The non-specific difiusional component estim ated from the concentration dependence was subtracted from the total uptake. The uphill o r concentrative transport o f uridine w as exam ined by com paring the intracellular concentration o f uridine w ith that o f the extracellular bathing m edium concentration. C alculation o f the intracellular volume based on the apparent equilibrium space using ^H-H2 0 , and the difference between w et w eight and dry weight o f the cell layer, was estim ated to be approxim ately 10 pl/m g protein, w hich results in a calculated intracellular uridine concentration o f approxim ately 0.1 pM at 30 seconds, about h a lf the extracellular concentration o f ^H-uridine (0.2 pM ), w hile accum ulating to 5 tim es the extracellular concentration (1 pM ) after 10 min. 144 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 120 - Total « I _ 100 " I S . 80 - f * 1 6 0 I E 40 - & -Na+ +Na+ 20 - 60 80 20 40 0 Uridine {\iM) Fig. 4-20: C oncentration dependence o f uridine uptake. Key: total uptake (solid squares), N a^-independent uptake (circles), and the Na^-dependent com ponent (solid line). D ata points are m eans, n = 3-5. 145 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6. Inhibition Studios To ascertain the selectivity o f the nucleoside transport process, inhibition o f ^H-uridine uptake by excess unlabeled physiological substrates w ere studied. These were nucleosides (uridine, thym idine, cytidine, adenosine, guanosine and inosine), nucleobases (uracil), nucleotides (uridine triphosphate), and ribose sugar. As shown in Table 4-11, 100 pM com peting purines (adenosine and guanosine but not inosine) and pyrim idines (thym idine, cytidine and uridine) nucleosides strongly inhibited the uptake o f ^H-uridine (20-25% o f controls). Inosine dem onstrated partial inhibition at -50% . Possible interaction o f uridine w ith the N a^-sugar transporter w as ruled out by no com petition w ith 100 pM ribose (Table 4-11). Likewise, uracil did not inhibit uridine uptake elim inating interaction w ith a nucleobase transporter, w hile the nucleotide uridine 5 -triphosphate (UTP) exhibited very weak inhibition o f ^H-uridine uptake. A Dixon plot o f adenosine inhibition o f uridine uptake w as evaluated. A denosine inhibition w as exam ined in the range from 1-100 pM w ith uridine concentrations o f 5, 10 and 20 pM . R esults dem onstrated th at adenosine was a com petitive inhibitor o f uridine uptake w ith an apparent K; value o f 44 pM (Fig. 4-21). 146 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T 3 O Q . C S Q . T3 (D M , 5 Table 4-11: Inhibition of ^H-uridine, ^H-thymidine and H-inosine uptake with 100 pM physiological nucleosides and o physiologically relevant substrates CD 8 T3 VC 3 . ( Q 3 CD CD i T3 O Q. C g. o 3 T3 O CD Q. T3 CD 3 Inhibitor Cone. ^H-Uridine ^H-Thymidine ^H-Inosine Control 0.89 ±0.17 3.67 ±0.52 2 73 ± 0.42 PjirinMints lihymidine 100 0.2 ±0.06*’ 1.62 ±0.09” 2.16 ±0.06 Cytidine 100 0.23 ±0.04** 1.47 ±0.26” 1.73 ±0.45 Uridine 100 0.16 ±0.02” 0.93 ±0.06” 1.31 ±0.07” Purines Adenosine 100 0.23 ± 0.03” 1.79 ±0.5” 1.89 ±0.26 Guanosine 100 0.2 ±0.02” 2.5 ±0.32 1.63 ±0.38 Inosine 100 0.46 ±0.04” 2.8 ±0.17 1.58±0.19 Other Substrates Uracil 100 1.05 ±0.15 UTP 100 0.69 ±0.09 Ribose 100 1.01 ±0.07 ( / > 5 Cells were treated with H-nucleoside (10 pCi/ml) and 100 pM competing nucleoside simultaneously for 30 sec uptake. All values are means ± s.e.m., n = 3-6. ^ Significant difference from the control for each radiolabeled substrate using ANOVA, p<0.05 0.14-1 0.1 2 - 0.10 0.08- O ) 0.06- 0.04 I — '* I ■ lO O O — '—I —" —I —■ T " ~ * —I - — I ■ —I -80 -60 -40 -20 0 20 40 60 80 100 120 Adenosine (|iM) Fig. 4-21: D ixon plot o f ^H-uridine uptake at 30 sec w ith 5 pM (squares), 10 pM (triangles) and 20 pM (circles) unlabeled uridine against increasing adenosine concentration. Each data point represents the m ean o f triplicate determ ination. The apparent Kj is 44 pM . 148 r Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7. Substrat» Selectivity The interaction o f therapeutically active nucleoside analogs w ith the ^H-uridine uptake process w as investigated using analogs that have m odifications either on the base, or the ribose sugar m oiety (Fig. 4-22). N ucleoside analogs studied include: (1) the nucleobase analog 5-fluorouracil (S-FU), (2) analogs w ith substitutions on the base, 5-iodouridine (5-IU), and 5-hydroxyuridine (5-H U ), (3) analogs w ith substitution on the ribose sugar moiety, 5-iodo-2’-deoxyuridine (IDU), nucleosides w ith a 2’-/rans-O H on the ribose ring, cytosine arabinoside (A ra-C ), adenosine arabinoside (A ra-A ), 2 ’-deoxyadenosine (2’-dA), 3'-deoxyadenosine (3’-dA), 2%3' -dideoxyadenosine (ddA ), 2’,3’- dideoxyinosine (ddl), 3'-azidothym idine (AZT), and (4) acyclic nucleoside analogs, acyclovir (ACV) and cidofovir (HPM PC). Strong interaction w as observed for those analogs that had substitution o n the nucleobase m oiety, but an intact or unm odified ribose ring structure. 5-Hydroxy uridine (5-HU) and 5-iodouridine (5-IU ) at 100 pM inhibited ^H-uridine uptake by approxim ately 70-80% (p<0.05). Am ong the analogs w ith m odifications to the ribose ring, only the removal o f the 2 -OH group as in 2 ' -deoxyadenosine (2’-dA) and 5- hydroxy-2' -deoxyuridine (5-lD U ) resulted in a 65% and 70% inhibition o f uridine uptake (p<0.05), respectively. By contrast, analogs w ith a 2 ’-trans~OH as in A ra-A and Ara-C did not inhibit ^H-uridine uptake (p<0.05). Rem oval or substitution o f the 3 -OH group as in 3 ’-dA and AZT were ineffective inhibitors o f uridine uptake (p>0.05). 149 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Likew ise, the acyclic nucleosides HPM PC and acyclovir, both did not reduce the uptake o f ^H-uridine in airw ay epithelial cells (Fig. 4-22). 8. Uptake o f Nucleoside Analogs: 2*-Deoxyadenoslne and HPMPC Inhibition studies reflect the ability o f the com peting nucleoside analog to bind and displace the binding o f ^H-uridine. To determ ine w hether nucleoside analogs are able to not only bind the nucleoside transporter b u t also get internalized to the cytoplasm ic com partm ent (transportability), we directly m easured the uptake o f ^H-2’- deoxyadenosine (a substrate - based on inhibition studies. Fig. 4-23) and ^H-HPMPC (a nonsubstrate). The uptake rate o f 2-deoxyadenosine w as 4 tim es higher than HPMPC. 2’-D eoxyadenosine uptake was sensitive to the presence o fN a , and was inhibited by excess unlabeled (100 pM ) 2'-deoxyadenosine and uridine by 50-60% (Fig. 4-23). By contrast, ^H-HPMPC uptake was insensitive to Na^ and was not inhibited by excess unlabeled (100 pM ) HPMPC o r uridine (p<0.05) (Fig. 4-23). 150 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2-dA 3-dA 50 100 150 % Control Uptake Fig. 4-22: U ptake o f ^H-uridine (10 pC i/m l) in the presence o f 100 pM nucleoside analogs. Each bar represents the means ± s.e m ., n = 4-6. The m ean uptake value for the control is 0.9 ± 0 .1 7 pm oles/m g protein/30s. Abbreviations: 5-fluorouracil (5-FU ), 5- iodouridine (5-IU rd), 5-hydroxyuridine (5-HUrd), 5-iodo-2’-deoxyuridine (IDU), cytosine arabinoside (Ara-C), adenosine arabinoside (Ara-A), 2’-deoxyadenosine (2 ’- dA ), 3 ’-deoxyadenosine (3’-dA), 2’,3’-dideoxyadenosine (ddA), 2 ’,3’-dideoxyinosine (ddl), 3 ’-azido-3 ’-deoxythym idine (AZT), acyclovir (ACV), cidofovir (H PM PC). * Significant difference from control uptake tested by the unpaired Student’s t-test, p < 0.05. 151 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I I 0 . 6 0 0.45 0.30 0.15 0.00 BRS SFR 100 pM drug Urd Fig. 4-23: Uptake o f ^H-2’-deoxyadenosine (10 pCi/m i) (solid bars) and ^H-HPMPC (10 pCi/m l) (hatched bars) in the presence o f bicarbonated R inger’s solution (BRS), Na^-free Ringer’s solution (SFR), or 100 pM unlabeled drug, or 100 pM uridine. Data points are means ± s.e m ., n = 3-5. S i^iificant difference betw een 2 ’-dA treatm ents by the unpaired Student’s t-test, p < 0.05. * Significant difference betw een 2’-dA and HPMPC by the unpaired Students t-test, p < 0.05. 152 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. E. FUNCTIONAL IDENTIFICATION OF TRANSPORTERS: INHIBITION PATTERN WITH PHYSIOLOGICAL NUCLEOSIDES The possible involvem ent o f m ultiple nucleoside transporters w as exam ined by com paring the inhibition pattern o f ^H-uridine, ^H-thymidine and ^H-inosine. Uridine was used a substrate for all known nucleoside transporters, ^H-inosine a substrate selective for the purine selective (N l) transporter and ^H-thymidine a substrate selective for the pyrim idine selective (N2) transporter (V ijayalakshm i and Belt, 1988; Crawford et al.. 1990). O f the broad substrate Na"^-dependent transporters, N3 accepts all nucleosides (W u et al., 1992) whereas, N 4 discrim inates against inosine (G utierrez et al.. 1992). As seen in Table 4-11, ^H-uridine w as inhibited strongly by both purines and pyrim idines (75-80% ) w ith the exception o f inosine. ^H-Thym idine uptake was inhibited by ju st the pyrim idines and adenosine (6 8 -88 % ), w hile the purines dem onstrated weak inhibition (30-40% ). On the other hand, inhibition results w ith ^H- inosine showed weak interaction (approxim ately 20-40% inhibition) w ith both purines and pyrim idines. To confirm this result, w e com pared the N a^-sensitivity and saturability of^H - thym idine and ^H-inosine uptake (Table 4-12). In the absence o f N a \ ^H-thymidine uptake was reduced by about 50% w hile ^H-inosine uptake did not show any inhibition. Estim ation o f their kinetic param eters revealed that thym idine w as transported via a high afiSnity low capacity transporter (Km = 21 ± 5.3 pM , V m ax = 29 ± 4.5 pmoles/mg protein/30s), whereas, inosine uptake was m ediated by a low affinity high capacity 153 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. transporter (Km = 120 ± 24.8 pM , Vm ax = 135 ± 25 pm oles/m g protein/30s). These data suggest the involvem ent o f m ultiple transporters that mediate nucleoside transport in the airw ay epithelial cells takes place via: tw o high affînity transporters (w ith differing substrate preference), one pyrim idine selective and another w ith broad preference, and a low afSnity transporter that m ediates Na*-insensitive uptake. 154 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T 3 O Q . C S Q . T3 (D W W O 3 O (D 8 " O v c 3 . ( Q 3 " i 3 CD CD 5 T3 O Q . C g . O 3 T3 O CD Q . O C T3 CD C / J C / J O 3 Table 4-12: Sodium dependence (pmoles/mg protein/30 sec) and kinetic parameters of ^H-uridine, ^H-thymidine and ^H-inosine in airway epithelial cells. Nucleoside BRS* SFR* Km" Vmax" (pmoles/mg protelii/30 sec) (pM) (pmoles/mg protein/30sec) Uridine 1.51 ±0.15 o .ii± o .o r 3.4 ±1.8 24.3 ±5.2 Thymidine 6.16±1.19 3.32 ±0.3“ 21 ± 5.3 29 ±4.5 Inosine 1.87 ±0.13 2.12 ±0.47 120 ±24.8 135 ±25 * Mean uptake ± s.e.m., n = 3-6. * * Kinetic parameters based on Michaelis-Menton Equation. Mean ± s.e.m., n = 3-4. ° Significant difference compared to control BRS uptake using Student’s t-test, p<0.05 L A L A F. TRANSEPITHELIAL TRANSPORT AND LOCALIZATION OF NUCLEOSIDE TRANSPORTERS In transepithelial transport o f ^H-uridine, ^H-thymidine and ^H-inosine, the cum ulative appearance o f radioactivity in the receiver fluid was linear for the 2 h r duration o f the experim ent (Fig. 4-24). The apparent Papp o f nucleosides (Table 4-13) w ith a log octanoI/pH 7.4 buffer partition coefBcient (log P) ranging from -1 to -2, was about 80- tim es higher than m annitol (log P = -3.1) a paracellular transport m arker (Papp = 1.2 x 10’^ cm/sec). U ridine and thym idine exhibited sim ilar perm eability’s (-9 .5 x 10"^ cm /sec), whereas inosine was 35% low er (6 x 10"^ cm /sec). In the basolateral to apical (BA) transport direction uridine perm eability was 10-times lower (0.95 ± 0.01 x 10^ cm /sec) than in the apical to basolateral (AB) direction, though 8-tim es higher than m annitol Papp (Table 4-13). Unlike m annitol that exhibits bi-directional sym m etry in transport, nucleosides show a vectorial preference for AB transport. Since nucleosides are able to gain entry to the intracellular com partm ent from the basolateral fluid (Fig. 4-18), as w ell as, exit the basolateral epithelial membrane to the bathing basolateral fluid (based on transepithelial transport). The possibility that a basolaterally localized nucleoside transporter may be responsible for the basolateral m em brane translocation o f nucleosides w as tested. Basolateral uptake is insensitive to N a \ and inhibition w ith 1 pM N BTI, but is sensitive to inhibition by 10 pM NBTI (-70% ) (Fig. 4-25). Furtherm ore, cells preloaded w ith ^H- uridine were found to efBux radioactivity to the basolateral fluid w ith tim e. This efBux 156 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. process was found to increase w ith increasing am ount o f loading (Fig. 4-26A and B), suggesting the possible involvem ent o f bi-directional Na^-independent nucleoside transport across the basolateral membrane. I 15 400- 10 300 100 50 200- 1 0 0 - 100 150 0 50 Time (min) Fig. 4-24: T ransepitheliai transport o f 10 pC i/m l o f ^H-uridine, ^H-thym idine and inosine across airw ay epithelial cells. Inset enlarges the transport profiles o f uridine and inosine near th e x-axis. D ata points are m eans ± s.e m ., n = 3. 157 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4-13: T ransepitheliai perm eability (Papp x 10"* cm /sec) o f ^H-uridine, thymidine, and ^H -inosine. N ucleoside iogP * P ap p" Uridine (A-B) -1.98 9.5 ± 0 .1 5 Uridine (B-A) -1.98 0.9 ± 0 .0 1 ' Thymidine (A-B) -1.17 9.6 ± 1.3 Inosine (A-B) -2.08 6.1 ± 0 .1 5 ' Log octanol / pH 7.4 partition coefficient fi’ om Hansch et al., (H ansch et al.. 1990) Mean values ± s.e.m ., n = 3. Significant difference, p<0.05, one-way ANOVA. IS 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.6 ê 0.5 ? 1 0.4 o & E 0.3 " 5 % * o 0.2 E o. 0.1 0.0 C ontrol SFR 1 |iM NBTI 10 |lM NBTI Fig. 4-25: B asolateral uptake o f ^H-uridine in the presence or absence o f N a \ or in the presence o f the inhibitor NBTI. C ells were pretreated for 15 m in w ith 1 or 10 pM N BTI in the basolateral fluid, before dosing w ith H-uridine (10 pCi/m l) fo r 30 seconds. M eans ± s.e m ., n = 4. 159 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. s « « s ô E 0.4 0.3 0.2 0.1 0 .0 ' 4 40 60 20 0 Time (min) Fig. 4-26: B asolateral efflux o f H -uridine over tim e follow ing 10 m in (circles), 30 (tria n te s) and 60 m in (squares) apical loading. C ells w ere dosed from the apical side w ith % -uridine (10 pCi/m l). M onolayers were w ashed three tim es and efflux into the basolateral fluid measured a t 1, 5, 1 0 ,2 0 ,3 0 and 60 m in (Panel A). The total percent o f uridine effluxed to the basolateral com partm ent afler 60 m in (Panel B). D ata points are means ± s.e m ., n = 3-4. 1 6 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. G. ADENOVIRUS INFECTION OF AIRWAY EPITHELIAL CELLS The role o f infection on airw ay epithelial structure and function was investigated. The virus selected for these studies was the adenovirus type 5 w ild-type (A d5), a virus that frequently infects the respiratory, ocular and gastrointestinal epithelium . 1. Infection o f Airway Epithelial Ceils The susceptibility o f the airw ay epithelial cells to hum an Ad5 (infectivity) was evaluated by determining the virus grow th curve, inununofluorescent m icroscopy, and the Adenoclone enzym e immunoassay for the viral fiber antigen. V irus grow th at a m ultiplicity o f infection o f 1:1 and 20:1 (num ber o f virus particles per cell) w as studied. A typical grow th curve was found with Ad5 W T replication in airw ay epithelial cells (Fig. 4-27) w ith the grow th phase peaking at 48 hr post-infection w ith a tw o log increase in virus yield. A t higher MOI the virus yield was larger. The m axim um virus grow th occurred betw een 24 and 48 hr. Infection was confirm ed w ith im m unofluorescence staining o f the viral fiber antigen. Forty-eight hr post-infection, putative staining was observed throughout the culture, at lOx m agnification (Fig. 4- 28A). W hen observed at a m agnification o f 40x, fluorescence w as found to be predom inantly localized a t the nuclear regions, whereas fainter diffused fluorescence associated w ith the cytoplasm ic regions (Fig. 4-28B). 1 6 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Infection o f airw ay epithelial ceils and the extent o f virus release w as tested using the A denoclone enzym e imm unoassay. The viral antigen in the bathing m edia increased w ith tim e post-infection, exhibiting a growth curve-like profile (Fig. 4-29). As w ith virus replication in the virus growth curve (Fig. 4-27), the extent o f virus release was also dependent o n the M OI, displaying a higher virus output in the bathing m edia w ith higher virus infection. 10 ® 1 0 ® 4 8 2 4 72 0 Time (hr) Fig. 4-27: Growth curve o f adenovirus ^rpe 5 in airway epithelial cells at a m ultiplicity o f 1:1 (solid squares) o r 20:1 (open squares). D ata points are means o f duplicate determ inations titrated in A 549 cells. 1 6 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 4-28: Im munofluorescence staining o f Ad5 in airw ay epithelial cells. Top panel (A) m agnification lOx, low er panel (B ) m agnification a t 40x. 163 I Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 . 0 1.5 1.0 0.5 0.0 60 80 20 40 0 Time (hr) Fig. 4-29: Release o f Ad5 antigen firom airw ay epithelial cells to the apical bathing fluid. M ultiplicity o f infection 1:1 (solid squares) and 20:1 (open squares). 2. Effect o f Adenovirus Type 5 Infection on Epitheiiai Structure 2.1 Alteration of Epithelial Bioelectric Properties Changes to epithelial bioelectric properties following infection was com pared with m ock-infected m onolayers that w ere subjected to the identical treatm ent as the infected cells, except the virus solution w as substituted with m edium. Ad5 WT infection o f air- interfaced airway epithelial cells in culture did not drastically affect the bioelectric properties com pared to m ock-infected controls (M l-control, no virus). A m inor decrease in the spontaneous potential difference was observed for Ad5 WT infected cultures, about 15-25% from 24-36 hr post-infection, and 40-60% decrease 48-60 hr post- 164 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. infection (Fig. 4-30A). Ad5 W T infection transiently decreased the R^, about 40% decrease at 60 hr post-infection (Fig. 4-30B). 2.2 Infection and Paraceiiuiar Permeabiiity The perm eability o f the paraceiiuiar m arker m annitol was studied in AdS W T and M I- control m onolayers to exam ine the alteration o f tight junction function in restricting drug perm eability. The apparent perm eabili^ coefGcient o f m annitol in Ad5 infected cell layers was 8 .8 (±2.4) x 10'^ cm /sec, and 6.3 (±0.4) x 10*’ cm /sec in M l-controls, a 1.4 fold difference (Fig. 4-31 A ). These values are not statistically different (p>0.05). Likew ise there was no difference in the cell associated radioactivity betw een infected and m ock-infected m onolayers (Fig. 4-3 IB ). 3. Infection and Carrier-mediated Nucleoside Transport Function Ad5 infection resulted in about a 30-35% decrease in Na^-dependent uridine uptake com pared with M l-controls (Fig. 4-32). The cellular content o f ^H-uridine w ith 30 sec uptake was not significantly different (p>0.05) in AdS infected cells than in m ock- infected cell monolayers. How ever, with 5 m in incubation the difference in uridine uptake was significant (p<0.05). Likewise, the Na^-independent uptake was also affected w ith AdS. The uptake at 30 sec w as not significantly altered, w hile 5 m in incubation yielded a 3S% decrease in cellular transport in AdS infected cell m onolayers com pared w ith m ock-infected controls. 165 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 50 40 ^ 30 g 20 10 40 80 60 2 0 0 Time (hr) I 5 4 3 2 1 0 20 40 60 0 B Time (hr) Fig. 4-30: Change in spontaneous potential difference (PD) (top panel. A) or R* (low er panel, B) after Ad5 infection at an MOI o f 20:1 (solid squares) and m ock-infected controls (open squares). D ata points are m eans ± s.e.m ., n = 3-5. 166 i Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. î 1 8 4 2 0 200 100 150 50 0 Time (min) B AdS M i-CTRL Fig. 4-31 : Time course o f m annitol perm eability (top panel. A) in airw ay epithelial cells infected w ith Ad5 at an M OI o f 20:1 (solid squares), or m ock-infected controls (open squares). Cell associated ^H-mannitol after 3 hr. M eans ± s.e m ., n = 3-4. 167 ? Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. BRS-308 BRS-300S SFR-308 SFR-3008 1 2 3 4 pmoles/mg protein 6 Fig. 4-32: E ffect o f AdS on nucleoside transport (unction. H -U ridine uptake was evaluated in the presence or absence o f N a^ in infected (solid bars) and m ock-infected controls (shaded bars). M eans ± s.e m ., n = 3-4. 168 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. V. DISCUSSION 169 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A n airw ay epithelial culture m odel was developed to study ion and drug transport in the conducting airw ay. This culture m odel would serve as a m odel system to screen drug perm eability, en^rm atic stability, and toxicity o f drugs designed for local delivery to the respiratory epithelium . Initially tw o models w ere studied, an air-interfaced culture (AIC) m odel, th at m im ics the lum inal conditions experienced by airway epithelial cells in vivo, and a liquid-covered culture (LCC) m odel, developed under standard subm erged culturing techniques. AIC was found to resem ble native tissue (m orphologically and electrophysiologically) more so than LCC, hence, w as used for further evaluation o f airway epithelial perm eability characteristics including: the apparent m olecular cu t-o ff lim it for paraceiiuiar transport and the optim al lip o p h ilici^ for transcellular perm eation. The hypothesis that a nucleoside transporter m ay m ediate the uptake o f nucleoside analogs was tested. D issection o f th e uptake m echanism revealed that m ultiple N a^-dependent and N a^-independent saturable processes are present in airw ay epithelial cells. These transporters can be targeted to facilitate the delivery o f antiviral nucleoside drugs to cells. Finally, the feasibility o f this approach was tested under adenoviral in& cted conditions. Ad5, disrupts the barrier properties o f confluent airw ay epithelial cells, even the nucleoside transport activity to a m inor extent. Im portantly, transporter activity is not abolished. Hence, th e nucleoside transporter may still function in the accum ulating antiviral nucleoside drug in infected cells. 170 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A. DEVELOPMENT AND CHARACTERIZATION OF AIRWAY EPITHELIAL CELL CULTURE MODEL 1. Primary Culture o f Airway Epithelial Cells The culture procedure used in this study was a m odification o f the procedures originally i developed by W u and Sm ith (W u and Sm ith, 1982), and Robison e t a i, (Robison et a i, 1993). In this study, rabbit epithelial cells treated with 0.2% Pronase E (bacterial proteases type XTV) for 90 m in at 37°C was not harsh for cell isolation, as the cell viability was >90% . The overall cell yield o f 15-20 x 1 0 ® cells per anim al is larger than that w ith the sm aller anim al models such as rat, hamster, and guinea pig (<5 x 1 0 ® cells/anim al) (Robison et al., 1993), thus providing an opportunity for im proving experim ental stratagem w hile lim iting anim al usage. The reliability o f the cell isolation and culturing procedure was reflected in a 90% success rate in the generation o f confluent cell layers that exhibit structural and fimctional characteristics th at resemble the native tissue. 1.1 Effect of Media Supplements The optim um culture conditions for adequate cell anchorage, growth and differentiation revolves around the selection o f a suitable m edia containing the right m ix o f growth factors and differentiation potentiators as m edia additives, the right substratum and the an air-interface condition (V an Scott e t al.. 1991). Both serum -supplem ented (1-10%) and serum-fi%e m edia have been developed for culturing o f airw ay epithelial cells 171 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Table 1-1). Serum , in airw ay epithelial cells has been found to lim it the growth, differentiation, and functional characteristics o f epithelial cells in culture (W u and Smith, 1982; Lechner et a/.. 1981; Yamaya e ra /., 1992). H igh concentrations o f FBS (protein content 30-50 mg /m l) induces squam ous differentiation (flat scale-like appearance w ith m ultilayering) o f airway epithelium . M oreover, functional characteristics m ay be altered, as found w ith th e a c tiv i^ o f the Na^-nucleoside transporter that was reduced 6 -8 times in the presence o f serum com pared to cells cultured in a low protein serum supplem ent in the intestinal epithelial cell line, IEC-6 (Jakobs et a/., 1990). Serum alters the bioelectric properties o f rabbit tracheal epithelial cells in culture, and AIC are m ore susceptible to the influence o f FB S than LCC. High concentration o f FBS (5% ) induced a -40% increase in equivalent Isc in AIC, while it reduced R t by 20-25%. In LCC, only m inor reduction (20-30% ) in R t was noted (Table 4-3). This effect was sim ilar to that reported for cultured canine tracheal epithelial cells where R t was reduced by 35% and Isc increased by -50% , in the presence o f 5% FBS. The reduction o f R t was consistent with the finding o f M arm orstein et al. (M ortell et a i, 1993), who proposed that an epithelial perm eability factor (EPF) present in serum interacts w ith the cell m em brane causing a constriction o f the cytoskeletal network, resulting in condensation o f the perijunctional actin ring, thereby increasing transepitheliai perm eability a t specific sites along the tight junction. Therefore, given the variability o f com m ercially available serum , it’s propensity to induce squamous differentiation and m odify epithelial bioelectric properties, m akes a low protein (<530 172 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. p.g/ml) defined m edia like PC-1, consisting o f DM E:F12 nutrient base supplem ented with ‘proprietary’ amounts o f insulin, transferrin, grow th factors and essential fatty acids, the m ost effective in potentiating the overall growth and differentiation o f anchorage dependent cells like the rabbit airw ay epithelial cells. 1.2 Effect of Substrata Anchorage dependent airw ay epithelial cells require a substratum that resem bles the basement mem brane in vivo in order to facilitate cell attachment. O n plastic, airw ay epithelial cell cultures in canine (V an Scott e t al., 1991), rabbit (K im , 1985), and human (Wu et a i. 1990) have a low attachm ent efB ciencies (10-20%), and attain ultrastructural features (dedifferentiated and very flat) very different fiom native epithelium . In addition, they were found to lack ciliary function, ciliogenesis and mucus secretory function. A sim ilar scenario w as observed for rabbit tracheal epithelial cells in this study, where attachm ent to tissue culture grade plastic was very poor. On prolonged incubation (> 1 0 days), the m orphology w hen view ed under a norm al phase or phase- contrast showed islands o f very flat and stretched-out cells (possibly a selecting out o f basal cells), com pared to a heterogenous population o f colum nar o r cuboidal cell morphology when cultured on m atrix coated-perm eable filter supports. Therefore, extracellular m atrix proteins are clearly required for airway epithelial cells. In an air- interface, cells grown on collagen (rat tail type I) and fibronectin (hum an type V) were the most conducive to cell attachm ent. C ell cultured on lam inin developed a relatively lower leq, while those cultured on M atrigel failed to reach confluence. Likewise in 173 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LCC, collagen and fibronectin substrata yielded to highest bioelectric properties (Table 4-2), albeit, a t a considerably low er level than in A IC. Lam inin coated filters developed very low bioelectric properties, and M atrigel none a t all. Therefore, extracellular m atrix (ECM ) glycoproteins are necessary for rab b it airw ay epithelial cell attachm ent and differentiation. W hether interaction w ith specialized ECM receptors on the basolateral m em brane o f epithelial cells occurs is unknow n, but m ay in part explain the selectivity for collagen and fibronectin (Sim on and M cD onald, 1990). These ECM receptors provide transm em brane links betw een ECM com ponents and cytoskeleton, thus, besides anchorage, m ay transm it im portant inform ation firom the extracellular to cellular space thereby influencing cell shape, m ovem ent, grow th and differentiation (Sim on and M cDonald, 1990) 2. Airway Epithelial Cell Culture Models, AIC vs. LCC 2.1.1. Morphology and Cellular Differentiation Electron m icroscopy o f the cells in suspension revealed th at three m ajor cell types are isolated firom the tracheal tissue: ciliated cells, secretory cells and basal cells (Fig. 4- 1 A). C iliated cells are identified by the presence o f cilia (long hair-like structures) on their apical surface, their function being prim arily in the propulsion o f m ucus (Sleigh et al.. 1988). Secretory cells are identified by the characteristic appearance o f secretory granules in th e cytoplasm. B asal cells are elongated cells that play a role in the attachm ent o f other epithelial cells to the basem ent m em brane. Besides these three cell 174 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. types, several other less com m on cells may also reside in the airw ay culture, such as nonciliated o r interm ediate cells, brush cells and Clara cells (Plopper et a i, 1983). Culturing o f these isolated cells on perm eable filter supports resulted in a heterogeneous population o f cell in a pseudostratified appearance arranged in a single layer w ith overlapping cytoplasm ic regions (Fig. 4 -IB ). Basal cells interspersed between ciliated, nonciliated and secretory cells help in the anchorage o f other epithelial cell types to the filter support substratum . The polarized confluent cell layers develop tight junctional complexes visualized by the darkened pinched o ff area betw een cells at the apical junction o f adjacent cells as indicated by the arrowhead in Fig. 4 -lC , that encircle the perim eter o f each cell form ing a continuous barrier. Exposure o f airw ay epithelial cells to an air-interface has been found to stim ulate differentiation in a num ber o f species. In m ost studies, four distinct areas have been investigated as a m ark o f im proved ultrastructural differentiation w ith cells grown in culture: cell size, shape and arrangem ent (Jetten et al., 1987; W u and Sm ith, 1982; Yamaya et a l. 1992); ciliogenesis (C lark et a l, 1995; de Jong et a l, 1994; Kondo et a l, 1993); and m ucus secretory function (A dler et a l, 1990; K im , 1985; Fiedler et a l, 1991), and electrical properties (Yam aya et a l, 1992; K ondo et a l, 1991). Subm erged cultures result in cell layers w ith a low er epithelial thickness, a concom itant loss o f cilia, a decrease in the num ber o f ciliated cells, loss o f m ucin- secretory function and lowered bioelectric properties. C onfluent AIC cells appear taller, cuboidal in shape, polarized, and well differentiated, w ith abundant m icrovilli and cilia on the apical surface (Fig. 4 -lB and 4 -lD ). O n the other hand, LCC cell layers appear 175 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. relatively flattened phenotype, w ith few er ciliated cells or cells w ith denuded or less prom inent cilia (Fig. 2E). These typical characteristics have been found commonly in tracheal (K aartinen et a i. 1993; C lark et a l, 1995; Zietlin e t al., 1988) and bronchial cell cultures (de Jong et al., 1993; de Jong et al., 1994) o f several species. W hen bovine tracheal epithelial cells were cultured under liquid-covered conditions, the cell height was 1/lO th that o f the native epithelium (Kondo e t a i. 1993). B y contrast, when cultured in an air-interface in a serum -fiee defined m edia the cell height was ~50 pm , sim ilar to th at in the native epithelium . Likew ise, in human tracheal epithelial cultures air-exposure accounted for a height difference o f 50 pm vs. 10 pm , as well as increased protein and DNA content (Yam aya et al., 1992). Furthermore, differentiation o f ciliated cells (ciliogenesis) and m ucin secretion has been reported to be strongly suppressed in subm erged cultures (Kaartinen et al., 1993; C lark et al., 1995; de Jong et al., 1993; A dler et al., 1990). Using videom icroscopy and image analysis on imm unostained cultures, C lark e t al., (Clark et al., 1995) found that ciliated cell differentiation (identified by means o f a m onoclonal antibody specific for ciliated cells) was suppressed 25-fold in subm erged cultures o f rat tracheal epithelial cells. In this study (Fig. 4 -lD and 4 -lE ), stim ulated growth and differentiation as w ell as the appearance o f cilia only w hen cultured under air-interfaced conditions was confirm ed by electron m icroscopy. Studies have revealed th at cell culturing affects the differentiation capacity resulting in loss o f mucin-producing secretory cells. The latter underm ines the overall effectiveness o f the culture m odel in replicating in vivo conditions (Kim , 1985; A dler et 176 I Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. al.. 1990). W ith rabbit tracheal epithelial cells, PAS staining for glycoprotein secreting cells in day 6 cultures indicated a 14% PAS positive stain for both AIC and LCC. This suggests that the culture retains a sim ilar population o f secretory cells as found in vivo (described in th e section ahead), and the attachm ent o f secretory cells was not affected by air-exposure. Due to the lack o f com m ercially available m arkers o f differentiation, the relative abundance o f ciliated and basal cells in the m onolayer cell population could not be investigated. 2.1.2 Comparison with Native Tissue In vivo the airw ay epithelium is a pseudostratified colum nar epithelium (Plopper et al., 1983). T he ciliated and secretory cells extend firom the mucosal surface to the basem ent m em brane. The basal and interm ediate cells contact the basem ent membrane giving the epithelium a pseudostratified appearance (W elsh, 1987). C iliated cells are colum nar -2 0 -6 0 pm tall, w ide at the apical surface (4-7 pm ) w ith approxim ately 250 cilia covering the apical surface o f each cell. Interspersed w ith the cilia are m icrovilli that protrude about 1-2 pm above the m ucosal surface. The luminal surface is also coated by a glycocalyx, that plays a role in protecting the epithelial cells. Secretory cells are identified by their characteristic electron dense secretory granules. Secretory cells are about a fifth o f the ciliated cells but their relative distribution varies between species and their location (proxim al or distal airw ays) (W elsh, 1987; Harkem a et a l, 1991), w ith the secretory cell population increasing in the distal airw ays. Basal cells are sm all 177 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. undifferentiated spindle shaped cells that aid in the anchorage o f ciliated and secretory cells to the basem ent m em brane m atrix (W elsh, 1987). This density o f secretory cells in AIC and LCC was noticeably sim ilar to the relative abundance o f secretory cells in the rabbit trachea in vrvo, com prising of: 17.6% C lara cells and 1.3% m ucus cells (both known to secrete glycoprotein), 28% basal cells, 43% ciliated cells and 9.4% unidentified cells (Plopper et al.. 1983; Harkem a et a i, 1991). Several types o f secretory cells are present in various species that vary in their relative distribution through the conducting airw ay. G oblet and serous cells appear to populate the upper airw ay, w hile C lara cells populate the bronchiolar airways. In the rabbit, C lara cells are found throughout the airw ays (Plopper e t a i, 1983). In the hum an airw ay, ciliated cells com prise nearly h a lf o f the epithelial cells at all levels down to the bronchioles, w hereas goblet cells are rare (Rennard et al., 1991). 2.2. Bioelectric Properties Increase in the overall cell differentiation o f cultured m onolayers reflects improved electrical properties (Y am aya et al., 1992; Kondo e t al., 1991). This observation was confirm ed in rabbit airw ay epithelial cells, by exam ination o f the baseline bioelectric properties (PD , Isc and R t) under open circuit and short-circuit condition. The spontaneous potential difference (PD) is the difference betw een the apical and basolateral m em brane potential generated by the flow o f cations and anions. The equivalent short-circuit (leq) current is an index o f the overall capacity o f a cell layer to 178 'i Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. i transport ions. The R t, is a m easure o f the resistance to the passive flow o f charge, an index o f barrier properties or integrity o f the epithelial cell layer. AIC and LCC are capable o f active cation absorption and/or anion secretion across the airway epithelium . The hig h PD indicates a distinct vectorial m ovem ent o f ions, accentuated in cell layers cultured in an air-interface. The AIC exhibited a greater c a p a c i^ for ion transport w ith a 44% higher leq compared to LCC, and a 28% lower Rt (Fig. 4-2A and Fig 4-2B ) (Table 4-1). The sm aller cuboidal differentiated phenotype o f A IC m ay have a higher overall tig h t junctional area per unit area, leading to a greater passive flow o f ion, accounting fo r a reduced Rt. Alternately, the tight junctional arrangem ent (num ber o f strands, distance betw een strands or equivalent pore size in the junctional area) may also play a ro le (Schneeberger, 1980; Inoue and Hogg, 1977). H ow ever, since the PD in AIC w as still larger than in LCC, it is likely that air-interface conditions elevates the cellular intrinsic capacity for active ion transport. These differences in the bioelectric properties have been observed in cultured human and dog tracheal epithelial cells and appear to be the typical response air-exposure elicits in airw ay epithelial cells (Yam aya e t al., 1992; Kondo et a l, 1993; de Jong e t a l, 1994). D ifferences in the bioelectric properties m ay also be reflected in the cell density. In vivo the num ber o f cells per unit area m ay be considerably higher than in culture. Thus, the cell morphology and size, cell num ber and tight junctional arrangem ents each may in p art account for the differences in bioelectric properties between native tissue and cultured cells and betw een AIC and LCC. 179 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.3. Ion Transport Properties The ion transport characteristics o f airw ay epithelial cells were studied using the voltage-clam p technique. First developed by U ssing and Zehran (Ussing and Zehran, 1950), this technique is based on the principle that the net transport o f ions across an epithelium gives rise to a current, called short-circuit cuirent (Isc). Using identical solutions at equal tem perature and pressure on either side the spontaneous open-circuit voltage generated is nullified by clam ping the transepitheliai spontaneous potential to 0 mV, thus abolishing the driving force for passive transport o f ions. Therefore, with this technique the differences in the active ion transport characteristics betw een AIC and LCC can be elucidated, including the predom inant ions being transported (N a \ CT, and HCO 3 ') and their relative contribution to the overall current. Inhibitors o f specialized ion transport processes responsible for Na^ current. C l current, or HCO 3 -related current, block the baseline Isc, thereby reflecting the contribution o f that pathway to the overall ion transport. Com pared to other epithelia such as the alveolar, conjunctival, and intestinal epithelia, the high equivalent Isc suggests that the airway has a very high capacity for active ion transport. The nature o f the airw ay epithelium w hether Na^-absorbing or CT- secreting or both, varies w idely w ith species. For instance, the hum an, dog, pig, sheep, guinea-pig and m onkey airw ay epithelia are predom inantly Na^-absorbing, the fetal sheep and dog airw ay epithelia are predom inantly Cl -secreting, w hile the rabbit, ferret. 1 8 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. bovine and cat trachea exhibit a com bination o f N a^-absorption and C l'-secretion (Van Scott e ra /.. 1991). 2.3.1. Electrogenic Na*-current in Airway Epitheiiai Ceiis In Na^-absorbing epithelia such as the airway epithelium , Na^ enters cells prim arily via Na^ channels. The low intracellular concentration o f Na^ is m aintained by the energized efflux o f Na^ across the basolateral mem brane by the Na^K^ATPase pum p (Jam igan et a i, 1983; Van Scott e t al., 1991; Robison and Kim, 1994; Z ietlin et al.. 1988). Cultured rabbit airw ay epithelial cells show the same characteristics. The lumen n e g a tiv i^ o f the airw ay epithelium spontaneous potential difference suggests a distinct cation absorption o r anion secretion. N a^ entry was verified w ith the Na^-channel inhibitors, am iloride (know n to block apical Na^-conductive pathw ays), and benzam il a m ore potent lipophilic analog o f am iloride (Benos, 1982). Na^ exit w as studied by treatm ent w ith the Na"^K^ATPase inhibitor ouabain. Basolateral ouabain treatm ent resulted in a near com plete dissipation o f Isc in AIC and LCC (Fig. 4-3A and 4-3B) and an decrease in the overall conductance. This effect w as gradual, but strictly selective for basolateral treatm ent, indicative o f the existence o f an ouabain-sensitive Na^K^ATPase pum p, the driving force fo r all ion transport in the cell, localized on the basolateral cell m em brane. The inhibition o f Isc w ith the Na^-channel inhibitors in AIC and LCC w as rapid w ith a parallel increase in Rt, consistent w ith channel inhibition. In AIC, baseline Isc w as inhibited b y — 70% on apical treatm ent w ith am iloride (Fig. 4-4A). Basolateral application in both AIC and 181 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LCC, did not elicit any inhibition in Isc, im plying that Na^ entry into airw ay epithelial cells prim arily occurs at the luminal surface. A m iloride inhibited furosem ide-insensitive current to a sim ilar extent (Fig 4-5A) th at was further confirm ed with apical benzamil treatm ent (Fig. 4-6A ). Taken together, these data indicated that Na^-current accounted fo r 70% o f the total Isc o f AIC airw ay epithelial cells. B y contrast, in LCC, apical am iloride treatm ent resulted in only 50% inhibition o f baseline Isc. This relatively w eaker response to am iloride observed w ith LCC w as confirm ed w ith benzam il where 50% o f the baseline Isc is an electrogenic Na^-current. Therefore, air-exposure makes airw ay epithelial cells predom inantly Na^-absorbing, w hile submerged culturing conditions decreases the N a-absorbing capacity o f airw ay cells. 2.3.1. Electrogenic C l* current in Airway Epitheiial Cells The possibility th at a C f-current, m ay be responsible a t least in part, for the unaccounted Isc w as tested. Further, it w as speculated th at Cl'-secretion in LCC may be elevated, as predicted by the dim inished Na^ current. A pical localization o f the Na^K"^(2C1*) or N aC l cotransport process in AIC and LCC was ruled out by the lack o f furosem ide sensitivity on apical treatm ent (Fig. 4-5A and 4-5B). In AIC, inhibition o f C r entry w ith basolateral furosem ide treatm ent resulted in — 10% inhibition o f baseline Isc. A sim ilar response was observed w ith furosem ide treatm ent o f am iloride-insensitive Isc. In LCC, sim ilar treatm ents as described above resulted in a 45% inhibition o f Isc (Fig. 4-5B ), w hile furosem ide treatm ent o f am iloride-insensitive current yielded a 26% reduction in current (Fig. 4-4B). The data obtained by alternating the order o f addition 182 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o f am iloride and furosem ide revealed that am iloride and fiirosem ide inhibitory effects w ere additive indicating the distinct nature o f the relative inhibitions. M oreover, there results suggest that am iloride does not stim ulate chloride secretion in cultured rabbit tracheal epithelial cells. The existence o f apical Cl -conductance pathways in AIC and LCC was confirm ed by inhibition w ith NPAA. In both AIC and LCC, a biphasic response was observed. A n initial rapid inhibition possibly reflecting C l -channel inhibition, follow ed by a slow er and prolonged inhibitory effect that may be attributed to an indirect CT- channel m odulation o r non-specific inhibition o f Isc (Fig. 4-7A and 4-7B). In AIC, N PA A inhibition o f baseline Isc was about 7% eventually reaching a total inhibition o f 43% . In LCC, the im m ediate inhibition o f Isc was about 13% that eventually yielded a 50% inhibition o f Isc. Thus, cultured airw ay epithelium is capable o f Cl’-secretory function, albeit to relatively different extents based on the culture condition. In all cases, am iloride, furosem ide, benzamil and N PAA accounted for about 75- 80% o f the total baseline Isc suggesting that cultured rabbit tracheal epithelial cells transport ion com prising prim arily o f Na^ and Cl*. The involvem ent o f H CO 3' to the Isc balance sheet was exam ined w ith DIDS, an agent that blocks HCO3 -related transport (H C0 3 */Cr antiport and/or a N a-H C O 3 sym port). A 15% inhibition in Isc for both AIC and LCC was observed w ith basolateral treatm ent, which supports the idea that pH regulators are present in both the AIC and LCC. M oreover, these im portant pH regulators appear to be localized exclusively on the basolateral m em brane o f airway epithelial cells. O ther pH regulators reported in other species that may be present in 183 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. rabbit tracheal epithelial cells include the N a HCO3 sym port and the Na^/H"^ antiport localized at the basolateral m em branes. These ion transporters are depicted in proposed schem atic (Fig. 5-1) w ith “?” as Anther studies are required to confirm their existence. Based on the inhibition data with am iloride, benzam il, furosem ide, N PA A and DIDS the putative m odel for ion transport in airw ay epithelial cells is show n in Fig. 5-1. Conceivably, Na^ present in the airway epithelial lum inal environm ent may enter cells via the Na^-channels, where the low intracellular concentrations are m aintained by the energized efflux through the Na^K^ATPase across the basolateral membrane to the bathing fluid. Na^ m ay then m ake its way back to the apical surface by difiusion through the junctional spaces betw een cells (Van Scott e t al., 1991). CT m ay enter the cell by m ore than one m echanism , the basolateral Na^-(K)^-CT or the NaCl cotransport process o r by a basolateral HCO3' related process like the HCO3 /C l exchange. C l exit occurs presum ably by a Cl -conductive pathway in the apical m em brane that m ay play a role in the active secretory processes in the airway epithelial cells. This CT-secretion along w ith Na^ diffusion to the apical surface creates an osm otic gradient draw ing water to the airway lum en responsible for m aintaining the fluid environm ent and hydration o f the m ucus blanket in the airw ay lum en (Diamond, 1979). Taken together, these characteristics underscores the importance o f the airw ay epithelium in m aintaining the electrolyte and fluid balance in the airway lum en m icroenvironm ent, and clearly em phasizes the preservation o f ion transport characteristics in cultured airw ay epithelial cells. 184 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.4. Comparison with Native Tissue and Other Ainvay Cuiture Modeis Com pared to the native rabbit tracheal epithelium (Table 4-1), the AIC retained a strikingly sim ilar high level o f active ion transport w ith a peak leq o f 75 pA/cm^ vs. 81 pA/cm^ (Jam igan et al.. 1983). The peak R t o f cultured m onolayers is higher (1,180 ohms.cm^ vs. 200-700 ohm s.cm^) (Jam igan e t al., 1983). Elevated R t com pared to native tissue is com m only observed in prim ary cultures in a num ber o f species (W elsh, 1987; Yam aya era /., 1992; Z ietlin era/., 1988). Sim ilar to the explanation provided for AIC com pared to LCC in the previous section, this may be ascribed to geometric difTerences o f cells and tight junctional regions. A lower density o f tight junctions per unit area in culture com pared to tissue, associated w ith the cuboidal nature o f AIC versus the colum nar appearance o f epithelial cells in tissue, or a decrease in the length o f tight junctions per unit area o f membrane. The same reasons m ay explain the lower lumen negative PD (12 mV) reported for tissue (Jam igan e t al., 1983) as opposed to 70 mV in AIC. In excised rabbit tracheal epithelial native tissue, about 70% o f Isc was attributed to Na^ transport along w ith a sm all net secretion o f Cl (Jam igan et al., 1983). The net Na^-absorption, based on ^N a^ flux (2.2 pEq/cm^/hr), accounted for 70% o f the total ion flux. A sm all net CT-secretory com ponent was observed however, though the value w as not significant (0.4 pEq/cm^/hr). Thus, the electrical properties o f native epithelia are strikingly sim ilar to that observed in the rabbit AIC. In cultured rabbit 1 8 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. airw ay tracheal cell layers (Z ietlin e t al., 1988), the PD , Isc and R t were 12 mV (lumen negative), 32.2 |xA/cm^ and 442 o h m s.cm \ respectively, w ith the relative Na^- absorption (am iloride-sensitive) accounting fo r 24% and C l'-secretion (forosem ide- sensitive) for 53% o f baseline Isc (Z ietlin et at., 1988). Therefore, LCC exhibits alm ost identical Na^ and Cl transport characteristics to other established rabbit tracheal cultures reported in literature. However, these subm erged m odels show considerably low er baseline electrical properties and squam ous-like differentiation that is quite different from the native epithelium . T hus, it appears th at air-interfaced conditions induce rabbit airw ay epithelial cultures to an in vivo like phenotype and functional ion transport characteristics. The bioelectric and ion transport properties o f excised airway epithelium show a w ide regional and species variation. V an S cott e t al. found that PD and R t decrease as one moves from the proxim al to the distal airw ays. M oreover, the relative Na^- absorption and C l'-secretion is region and species selective (V an Scott et a i, 1991). In canine trachea the PD and R t are significantly higher (approxim ately 3-tim es) in the trachea than in the bronchus. N a^-absorption accounts for 45% o f the total Isc in the tracheal region and 78% in the bronchial region (V an Scott et al.. 1991). In the human trachea, a R t o f 125 ohms.cm^ and a baseline Isc o f 28 pA/cm^ was reported, that was virtually entirely Na^ absorption (Y am aya e t a/., 1992). In sum m ary, the striking difference betw een the tw o culture m odels w as that m orphologically, AIC appeared m ore differentiated, w ith the bioelectric and ion transport properties elevated due to enhanced Na^ absorption and reduced CT secretion. 186 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Therefore, culture conditions significantly affect the overall ion transport m echanism s altering th e functionality, or expression, or regulation o f individual ion transport processes such as channels, cotransporters o r exchangers involved in Na^ absorption and c r secretion. This was ascribed to the lack o f hydrostatic pressure exerted by the apical bathing liquid on the cell surface, o r alternately, enhanced difiusion o f oxygen presented in air-interfaced conditions (Johnson et al., 1993). 187 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I I ^ 3C Î O i i I ë c 1 I § I J If * ob Z 188 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3. Biochem ical Comparison In canine airw ay epithelial ceils, Johnson e t al. (Johnson e t al., 1993) dem onstrated that air-exposed cell layers exhibit elevated Na^ absorption coinciding w ith an enhanced O2 consum ption com pared to subm erged cultures. This the authors found was associated w ith an increase in cellular ATP levels largely due to the ability o f the cells to perform oxidative phosphorylation under aerobic conditions. C ells cultured in a submerged conditions develop hypoxia leading to an increase in glycolytic m etabolism (anaerobic respiration) and consequently an increased lactate production. This hypothesis was tested in rabbit tracheal epithelial AIC and LCC. The overall rate o f lactate appearance in the bathing fluid o f LCC w as about 43% higher than th at in AIC (Table 4-5) (Fig. 4- 9A). This finding is consistent w ith that observed by Johnson et al. (Johnson et al., 1993), w here the exposure o f airw ay epithelial cells to a ir m ay shift cellular respiration from a com bination o f glycolytic and oxidative (seen in subm erged cultures) to a m ore aerobic respiration (oxidative). This change in cellular respiration had a sim ilar stim ulatory effect on Na^ current in AIC as evidenced by the 44% elevated equivalent Isc. The level o f O2 consum ption was not perform ed in th is study, but the data is in agreem ent w ith that found in the canine bronchial epithelial lactate study where air- exposure facilitates oxygen delivery to cells that results in increased oxidative m etabolism th at translates to a low er lactate production. Interestingly, in AIC m ost o f the lactate (99% ) w as selectively released to the basolateral fluid, while in LCC, about 70% was found to be basolaterally localized (Fig. 189 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4-9B). This suggested that lactate was predom inantly secreted across the basolateral cell m em brane in airw ay epithelial cells in either o f the culture m odels, although the rate o f release in AIC w as 20% low er (p<0.02) than in LCC (Table 4-5). This profound vectorial preference for basolateral release in AIC, m ay have significance in the clearance o f glycolytic products such as lactate and pyruvate by blood. W hen considering the alm ost nonexistent apical lactate release in AIC, two possibilities exist: either there m ay not be any cellular m echanism to m ediate the release o f lactate to the apical fluid, or, alternately, a specialized carrier m ay be present that functions in the rapid and efGcient absorption o f these organic anions (Fig. 5-2). In the case o f the form er, basolateral efflux follow ed by system ic clearance m ay be the principal m echanism for the rem oval o f cellular m etabolites. In the case o f the latter, a specialized carrier process such as a m onocarboxylic acid transporter, M CTI or M CT2, m ay fim ction in the efGcient salvage o f secreted lactate. However, there is no evidence to support either hypothesis at th is stage. In skeletal m uscles, G arcia et a l, (G arcia et a i, 1995) reported that M CTI and M CT2, were expressed in m itochondria-rich oxidative fibers, and neither were detectable in m itochondria-poor glycolytic fibers. A long the same lines, it m ay be possible that air-exposure as in AIC (rich in m itochondria) may induce aerobic respiration th at m ay result in the expression o f M CTI and/or MCT2 th at could be deficient in LCC. Conceivably, a num ber o f cellular events m ay be affected w ith the nature o f cellular respiration. In this study, coupled w ith the enhanced aerobic respiration, Na^ absorption and lactate transport w ere all affected and consequently m ay all be interrelated. A shem atic representation o f this scenario is depicted in Fig. 5-2. 190 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. M CTI o r 2 Na- apical anaerobic Gluconeogenesis pyruvate ^ ^ lactate basolateral Lactate MCTI or 2 Na apical anaerobic / aerobic Gluconeogenesis lactate pyruvate basolateral Lactate carrier-mediated ^ passive diffusion Fig. 5-2: Schem atic representation o f cellular respiration and lactate release. Top panel is the proposed schem e o f cells cultured in an air-interface (AIC). Lower panel is the proposed scheme in liquid-covered cultures (LCC). 191 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4. Effect o f Culture Condition on Drug Transport The transport o f drugs across airw ay epithelial barrier could occur in four possible ways: ( 1) by transport via specialized m em brane-bound carrier-proteins, (2 ) by transcellular passive difiusion through the phospholipid bilayer, (3) by paraceliular difiusion, involving the passive transport o f drugs through the intercellular tight- junctional areas betw een cells; (4) transcytosis, involving uptake into vesicles, follow ed by their m ovem ent from the apical to the basolateral surface or vice versa. In this study the characteristics o f paraceliular, transcellular difiusion and carrier-m ediated transport o f drugs in the airw ay epithelial cell culture m onolayers were investigated. The hydrophilic drugs, m annitol, sucrose and albuterol are expected to be excluded from the lipophilic cell m em branes and perm eate through the aqueous paraceliular spaces. In rabbit tracheal epithelial cells m onolayers the hydrophilic solutes (m annitol, sucrose and albuterol) exhibited relatively low perm eability (Fig. 4-lOA and B inset). Com paring the Papp’s o f m annitol (Table 4-6) to those in establish models like Caco-2 and MDCK, the Papp was about 10-times low er at 1.17 x IC^ cm /s and 2.1 x IC^ cm/s, respectively (Adson et a i, 1994). Thus it appears that the intercellular junctions present a m ajor barrier that restricts the free difiusion o f hydrophilic drugs across the epithelium . LCC exhibited a higher perm eabili^ to m annitol, sucrose and albuterol, relative to AIC. The reason for this is unclear, but may be due to tight junctional differences betw een the tw o culture m odels alluded to in the preceding bioelectric and ion transport section. Com pared to excised tracheal preparations (Jam igan et al., 1983; W angensteen 192 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. et al., 1993), the Papp o f m annitol and sucrose in culture is alm ost 10 tim es lower. Indeed, in our hands, the Papp o f m annitol in excised rabbit trachea m ounted in Ussing cham bers displayed a value 5-tim es higher, 5.7 x 10’^ cm /sec (n = 3) w ith a R t o f 445 ± 35 ohms.cm^ w hich is com parable to the m annitol Papp in canine tracheal epithelium (Al-Bazzaz and Cheng, 1979). The reason for these differences is not know n but could be related to differences in junctional structure, electrical resistance (native tissue being reported to be considerably more leaky than A IC), and differences in experim ental technique, i.e. a Transw ell difRision cham ber system versus an U ssing s ^ le difiusion cham ber or cannulated tracheal tube preparations (W angensteen et al., 1993). The barrier properties o f cell m onolayers is o f forem ost im portance when considering paraceliular transport. T his is often overlooked and m ay present itself as a crucial artifact o f experim ental design or, drug effects th at facilitate their own transport. In general, m ost drugs tested did n o t com prom ise the integrity o f A IC, w ith the exception o f albuterol, that decreased by 12% (probably due to fi-adrenergic stim ulation). On the other hand, at the sam e concentrations, R , in LCC decreased for all drugs studied w ith the B adrenergic agents albuterol and propranolol recording a 37% decrease in Rt. Therefore, these results indicated that LCC w as m ore sensitive to the drugs tested in this study. As toxicity was not apparent in A IC, it appears that m onolayer integrity o f LCC was com prom ised due to experim ental m anipulation such as m onolayer w ashing, dosing and sam pling during transport experim ents. This may also partially account for the higher perm eability o f m annitol, sucrose and albuterol in LCC relative to AIC. 193 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Transport o f lipophilic drugs is very rapid in rabbit tracheal epithelial m onolayers and does not seem to be affected by the culture condition. This was anticipated as their perm eability is difhision-lim ited w ith partitioning into the apical phospholipid m em brane and out o f th e basolateral m em brane being the prim ary determ inants o f tran sp o rt D exam ethasone, a m oderately lipophilic drug (log P = 1.83) and propranolol, a highly lipophilic drug (log P = 3.21) are — 200 tim es m ore perm eable w hen com pared w ith m annitol. T ight junctions play a secondary role in determ ining the perm eability o f lipophilic drugs. Evidence for th is based on the lack o f any correlation betw een the Papp o f dexam ethasone and propranolol and R*. In addition, no statistical difference was found betw een their Papp in A IC and LCC. Taken together, this suggested that the transport o f lipophilic drugs w as prim arily governed by the barrier properties o f the plasm a m em branes that exhibit sim ilar perm eability’s throughout the respiratory tract. 194 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CD T3 O Q . C s Q . T3 CD C / J C / J Ô Table 5-1 o CD I I 3. ' CO o o CD c 3. 3 " CD CD T3 O Q . C a o 3 T3 O CD Q . O C T3 CD C / J C / J o' 3 Comparison of Papp (x 10'^ cm/sec)" of Polar Solutes in the Rabbit Nasal and Tracheal Epithelial Barriers Drug Nasal Tissue'’ Tracheal Monolayers Aveolar Monolayers' Mannitol 62 ± 12 ll.8±3 18 FITC n r* * ll.2± l.l nr" FD-4 342 2.77 ±0.4 1.29 ±0.06 FD-IO 127 1.86 ± 0.6 1.16±0.09 FD-20 nr 0.36 ±0.01 1.13±0.17 FD-40 67.7 0.32 ±0.02 0.35 ±0.06 FD-70 18.1 0.30 ±0.03 0.15 ±0.01 Propranolol 2554'’ 2870 ±378 2910 ±168 ‘Mean ± s.e.m., n = 3-6. ^ From Kubo et al., (Kubo et al, 1994). ‘ From Matsukawa et al, (Matsukawa et al, 1997), not reported. « B. PERMEABILITY CHARACTERISTICS OF AIRWAY EPITHELIUM O f the two culture models, the resilience o f the AIC to experim ental handling, together w ith bioelectric, electrophysiological and biochem ical characterization studies, strongly supported A IC as the model o f choice. From this point on, all studies involved the application o f the air-interface culture for the understanding o f transport phenomena in the airway epithelium . The perm eability o f the air-interfaced cultures to small solutes and m acrom olecules was characterized by studying two o f the key factors that influence passive transepithelial drug transport, m olecular size and solute lipophilicity. The perm eability o f the air-interfaced rabbit tracheal epithelial cell m onolayers to hydrophilic m acrom olecules was dependent on m olecular size (Fig. 4-11) and show ed no direction (Table 4-7) o r tem perature dependence. The m olecular weight cu t-o ff was about 20,000 daltons (Fig. 4-13). This m olecular sieving m ay be explained by hypothetical w ater-filled cylindrical pores or channels o f about 5 nm in radius present in the tight junctional area or intracellular regions that perm its the restricted fiee diffiision o f hydrophilic drugs across the epithelia (Fig. 4-14). These pores are hypothetical disruption, also referred to as gates, in the linear fusions o f adjoining m em branes o f neighboring cells. Evidence for their existence is yet to be confirm ed m orphologically o r electrophysiologically (Cereijido et al., 1978). Solutes whose Stokes-Einstein radius approaches that o f the pore size may 196 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. experience increasing restriction due to steric hindrance on entering the pore and frictional resistance during passage through the length o f the pore (Landis and Pappenheimer, 1963). Perhaps due to their rod-like characteristics, the difiusion o f m acrom olecular solutes > 2.44 nm in size (FD-20, FD-40 and FD -70) was greatly restricted, as indicated by th eir Papp's (-0.3 x 10'* cm /sec) that are not significantly different from each o ther (p > 0.05). Therefore, hydrophilic solutes near and above the threshold o f 2 0 ,0 0 0 daltons m ay experience sim ilar restrictive difiusion when considered for airway drug delivery. The density o f equivalent pores m ay play a role in the overall perm eability o f hydrophilic solutes and the barrier properties. T he total number o f equivalent pores in each airway epithelial cell m onolayer o f 1.13 cm^ was found to be about 3.5 X 1 0 ® w hich translates to a pore density o f 3.1 x 1 0 ® /cm ^. In the alveolar epithelial cells the calculated pore density was slightly low er a t about 2.5 x 10®/cm^ (M atsukawa et a l, 1997). Thus, the short difiusional path length (0.5 pm) in alveolar cells com pared to the airw ay epithelial cells may be offset by a low er number o f pores eventually leading to sim ilar overall perm eability coefBcients. The perm eability coefficients obtained in the air-interfaced rabbit tracheal epithelial culture m odel (Table 4-7) are com parable to those o f sim ilar sized solutes in excised tracheal tissue and other epithelia. In isolated rabbit tracheal tissue preparations (W angensteen et a l, 1993; Guo et a l, 1995), Papp's o f 3.5 x 10** cm /sec and 0.4 x 10** cm/sec were reported fo r inulin (5,500 daltons) and FD-20, respectively, as com pared with 2.8 X 10** cm /sec fiar FD -4 and 0.36 x 10"* cm /sec for FD -20 found in this study. Although the pore-size o f rabbit tracheal tube preparations w as n ot estim ated, excised rat 197 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and guinea pig tiacheas w ere reported to have pores radii o f 3.5-4 nm. In general, the perm eability o f the air-interfaced rabbit tracheal cell m onolayers correlates reasonably w ell w ith excised native rabbit trachea. The m inor differences in the perm eability results and pore-radius estim ations betw een the airw ay culture m odel in this study and the excised tissue m ay be related to the differences in experim ental design (cultured m onolayers vs. tracheal tube preparation), and the extent o f tissue edge dam age. High variability in excised tissue R* ranging from 125 ohms.cm^ (W angensteen et al., 1993; Guo et al., 1995) to as high as 700 ohms.cm^ (Jam igan et al., 1983) has been reported w hich m ay have a profound effect on the perm eability to hydrophilic drugs. Sim ilar to the FlTC -dextrans, the hydrophilic B-adrenergic com pounds also appear to opt for the paraceliular o r pore pathw ay for transport. The Papp's o f sotalol (M W 308.8), atenolol (M W 266), and albuterol (M W 239.3) are sim ilar (Table 4-8) and com parable to that o f m annitol (Table 5-2). This was also the case in the alveolar epithelium (Saha et al., 1994), w here the perm eability o f polar B-blockers is sim ilar to th at o f polar peptides, 1 mM G ly-L-Phe (M W 222.2), Gly-D-Phe (M orim oto et ai, 1993) and '^1-thyrotropin releasing horm one (M W 363.4) (M orim oto et al., 1994) w ith Papp's o f 3.3 X 10'^, 1.6 X 10"\ and 1.6 x 10'^ cm /sec, respectively. The m oderately lipophilic com pounds (m etoprolol and tim olol) and the highly lipophilic com pounds (betaxolol and propranolol) are transported m ore rapidly and to a greater extent than the polar fi-adrenergic com pounds (Fig. 4-15). A sigm oidal relationship best describes the influence o f drug lip o p h ilici^ on transepithelial transport. The effective half-m axim al Papp occurred at a log P o f about 2.08 (Fig. 4-16). This w as 198 I Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sim ilar to the value o f 1.95 reported for rat alveolar epithelial cell cultures (Saha et al., 1994), 1.63 for the excised rabbit conjunctival tissue (W ang et a i, 1991), and 1.98 for the isolated rabbit com eal tissue (W ang et at., 1991). T he above trend concerning the sigm oidal influence o f lipophilicity on drug absorption has been well docum ented in the rat lung following intratracheal instillation in several studies (Schanker, 1978; Enna and Schanker, 1972; Lanm an et a i, 1973). O ther reports in excised ham ster buccal epithelium (Tavakoli-Saberi and Audus, 1989), rabbit intestinal m ucosa (Narawane et al.. 1992), Caco-2 cell m onolayers (A rtursson, 1990), and cultured bovine cerebrovascular endothelial cells (van Bree et al., 1988) also report a sim ilar sigm oidal dependence on the lipophilicity o f drugs. A possible exception to the sigm oidal relationship was m etoprolol (Fig. 4-16). As in the case with the conjunctiva (W ang et al., 1991), cornea (Schoenwald and Huang, 1983), and Caco-2 cells (A rtursson, 1990), the Papp for m etoprolol in tracheal epithelial cell m onolayers was larger than predicted (249 x 10'^ cm /sec) when com pared to that o f tim olol (63.2 x 10'^ cm /sec) w ith a sim ilar log P. T he reason for this is not known, but could be related to the ability o f m etoprolol to fluidize phospholipid m em branes (Shi and Tien, 1986) resulting in the anom alously faster perm eation. In conclusion, the air-interfaced rabbit tracheal cell culture m odel w as successfully applied to elucidate the perm eability o f tracheal epithelial cell m onolayers to m odel m acromolecules and sm all solutes that opt fo r the paraceliular or transcellular diffosional pathway. It appears that the cultured rab b it tracheal epithelial cell monolayers absorb drags in a sim ilar m anner as do native rabbit trachea and other epithelia. 199 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Comparison of Airway and Alveolar Epithelial Cells Culture Models To present an overall perspective o f drug transport characteristics o f respiratory epithelial cell in culture, the sim ilarities and the differences betw een the conducting airw ay and alveolar epithelial cells in prim ary cultures w ill be contrasted (Saha et a i, 1994). They both com prise a heterogeneous population o f cells that exhibit structural and biochem ical characteristics known to exist in vivo in their respective regions o f the respiratory tract. N onetheless, from a m orphological perspective, the two culture m odels are very different, the m ost prom inent being the relative shape and size o f cells. C onfluent tracheal epithelial cells are prim arily cuboidal or colum nar in shape w ith a height o f 10-50 pm , w hereas alveolar epithelial cells differentiate into type 1-like cell m orphology and are very flat (about 0.5-1 pm ) covering a surface area believed to be 9 tim es th at o f the cuboidal type II cells. The electrical properties o f airway epithelial cells appear to be higher than the alveolar epithelium w ith a spontaneous potential difference o f 70 m V vs. 9.7 mV and an Isc o f 75 vs. 4 pA/cm^ (M athias et al., 1995; Kim et a i, 1991). The Rt in the alveolar cultures is higher — 2,200 ohms.cm^ compared to 1,200 ohms.cm^ in airway epithelial cultures. This is not surprising, given that the large surface area o f type-l-like cells in the alveolar cultures probably have a considerable low er tight junctional area per unit area, for the passive m ovem ent o f ions com pared to the m ore com pact airway epithelial cells. Interestingly, however, the perm eability’s o f both culture m odels to the solutes studied as discussed earlier is not very d ifferen t N ot only are their perm eability coefficients comparable (Table 5-1 and 5-2), but also do they both show a 200 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sim ilar restrictive difiusion o f hydrophilic m acrom olecules (a pore-radius o f 4-5 nm) (M athias et al., 1996; M atsukaw a et at., 1997) and a sim ilar sigm oidal influence on perm eability to solutes w ith increasing lipophilicity (M athias et al., 1996; Saha et al., 1994). The calculated pore density in airw ay epithelial cell m onolayers was found to be about 3.1 X 10^ /cm^. In the alveolar epithelial cells the calculated pore density was slightly low er at about 2.5 x 10^/cm^ (M atsukaw a et a i. 1997). Thus, the short difiusional path length (0.5 pm ) in alveolar cells com pared to the airw ay epithelial cells may ofiset the low er estim ated pore density eventually leading to sim ilar overall perm eability coefBcients. This suggests that the epithelia o f the conducting airw ays and lower respiratory tract m ay share sim ilar barrier properties to the transport o f macromolecules. By contrast, the nasal m ucosal epithelium , a leaky tissue w ith Rt o f about 40 ohms.cm^ (Hosoya et al., 1993), has been reported to have Papp values o f FITC-dextrans about 60- 200 tim es greater than those observed from the tracheal and alveolar epithelium (Table 5- 1). From the published Papp's an equivalent pore o f about 13-15 nm in radius in the nasal epithelium was estim ated using Eq. 3 and 4. Therefore, the nasal epithelium is considerably more perm eable to hydrophilic m acrom olecules than the conducting airway and alveolar epithelium . O n the other hand, the penetration o f a lipophilic drug is largely governed by partitioning into the phospholipid m em branes. This w as evident from the perm eability coefBcients o f propranolol that was noted to be sim ilar in all three region o f the respiratory tract (Table 5-1) Thus, in spite o f m ajor m orphological differences, the conducting airway and alveolar epithelium exhibit sim ilar paraceliular and transcellular passive transport 201 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. processes. It is likely that in the case o f hydrophilic drugs the reduced tight junction d iS usional area in the alveolar m odel m ay be offset by the very low difiusional distance com pared to the airw ay epithelial cell m onolayers. Lipophilic drugs, b y contrast, are less sensitive to the aforem entioned epithelial cell differences as the area fo r difiusion is relatively consistent. Therefore, the perm eability o f B-adrenergic com pounds in both the airw ay (M athias et aL, 1996) and alveolar (Saha et al., 1994) epithelium both showed a sim ilar sigm oidal relationship to drug lipophilicity and are astoundingly w ell correlated (F ig. 5-3). Table 5-2 compares the calculated param eter estim ates based o f the perm eability data for both the airw ay and the alveolar m odels using the sigm oid equation (E q. 5) w here: “a” represents the m inim um predicted perm eability (1-2 x 10’^ cm /sec) th at estim ates the paraceliular transport o f hydrophilic m olecules , “b” is an estim ate o f th e m aximal transepithelial perm eability (285-295 x 10'^ cm/sec) th at represents the transcellular perm eability o f lipophilic m olecules, “c” is the log P corresponding to the h a lf m axim al perm eability (2.08 vs. 1.79), and “d” is the slope or the transition betw een “a” and “b” (0.13-0.3), respectively. O n the other hand, active transcellular transport processes m ay be quite different betw een airw ay and alveolar epithelial cell m onolayers. U nfortunately, a num ber o f transport processes relevant to drug delivery have no t yet been w ell characterized in eith er one o r both o f the respiratory epithelial cell m onolayers. 2 0 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. loooon I I 1000- » o 1 0 0 - < 5 10 - 1000 10000 100 10 1 Papp(tracheal) xIO" (cm/sec) Fig. 5-3: C orrelation betw een fi-adrenergic agent p erm eab ili^ in airway and alveolar epithelial cell m onolayers 203 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 5-2: Param eters for sigm oidal relationship* between apparent perm eability o f P- blockers and their lipophilici^ in the tracheal and alveolar epithelia. Param eters Rabbit tracheal cell Rat alveolar cell* M inimum Papp (x 10’^ cm /s) (a) 2 .2 2 1.2 M axim um Papp (x 10"^ cm/s) (6 ) 285 295 Log P at half-m axim al Papp (c) 2.08 1.79 Transitional slope representing the gradual increase in Papp (d) 0.13 0.31 *Papp = a + 1 + exp [-(lo g P - c y d \ Reference (Saha et ai, 1994) 204 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T3 O Q . C S Q . T3 (D W W O 3 O (D 8 3 . ( û 3 " i 3 (D C 3 - 3 " (D (D T3 O Q . C a O 3 T3 O (D Q . Table 5-3: Comparison of pulmonary absorption rate of hydrophilic compounds in various experimental models Approximate Molecular W eight Tracheal Cells Tracheal Strips Alveolar Cells A549 cells Isolated lung In vivo Drug g PappxlO (cm /s) Drug PappxlO * (cm/s) Drug PappxlO * (cm/s) Drug PappxlO * (cm/s) Drug PappxlO * (cm/s) Dnig % absorbed w ithin 120 min 200 m annitol" 11.8 mannitol * 57 mannitol 18 mannitol ^ 4.46 400 sucrose^ 3.0 sucrose" 51 T R H ' 4.7 sucrose '' 2.41 phenol red " 37.0 1,000 vasopressin ^ 3.6 bradykinin^ 840 5,000 FD4* 2.77 inulin " 3.5 FD4* 1.39 FD4* 340 FD 4" 7.2 10,000 FDIO* 1.86 FDIO* 220 FD IO " 2,0 20,000 FD20* 0.36 dextran" 0,4 FD20* 120 FD 20' 0.051 40,000 FD40* 0.32 HRP*' 0.7 FD40* 75 70,000 FD70* 0.30 albumin ' 24 FD70* 61 albumin " 0.15 FD70* 0.4 150,000 FD I50* 0.13 FD150* 25 O c " D ( D ( / ) C O o' 3 References: " (Mathias et ai, 1996), * ’ (Mathias et ai, 1995), “ (Wangensteen et ai, 1993), * * (Kim et ai, 19911, * (Morimoto et ai, 1994),‘ ^(Yamaharaetfl/., 1994),* (Matsukawa et a/., 1997),’ '(Matsukawa et a/., 1996), ’ (Kim et n/., 1995), ^ (Kobayashi et n/., 1995), (Berg et ai, 1989), ’ (Goodman et ai, 1994), (Wangensteen and Yankovich, 1979), " (Morita et ai, 1993) S c. NUCLEOSIDE TRANSPORT STUDIES 1 Canier-mediated Uridine Uptake in Airway Epitheiiai Ceiis 1.1 Uptake Rate, Directionality, Na* and Temperature dependence Initial evidence th at nucleoside uptake was m ediated by an active carrier-m ediated process was obtained firom the findings that uridine uptake was rapid, tem perature- sensitive, and asym m etric in nature (Fig. 4-18). T he uptake rate from the basolateral m em brane was four tim es low er than from the apical membrane, suggesting a vectorial preference for lum en to cytoplasm ic transfer o f nucleosides across the apical plasm a m em brane (Table 4-9). Uridine uptake was strongly dependent o n the presence o f Na*, being nearly abolished when Na* was replaced w ith choline in th e bathing medium (Fig. 4-18). The coupling o f Na* w ith uridine based on th e H ill equation (Eq. 7) revealed a 1:1 stoichiom etry (Fig. 4-19), w hich suggests one Na* is required for the translocation o f a uridine m olecule across the apical cell m em brane. The apparent concentration for Na* activation o f uridine uptake (K ^ J was 14.8 m M , sim ilar to the values (13.3 mM) reported in ra t intestinal brush-border m em brane vesicles (BBM V) (Iseki et al., 1996)and bovine renal BBMV (11.4 mM) (W illiam s and Jarvis, 1991). These results, along w ith 75% inhibition observed w ith basolateral ouabain (10 pM ), and apical phloridzin (100 pM ) pretreatm ent (Table 4-10) are consistent w ith the notion that uridine uptake in airw ay epithelial cells was m ediated by an electrogenic carrier- m ediated transport process. These characteristics o f ^H-nucleoside uptake have also 206 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. been observed in other cultured epithelial ceils such as: isolated intestinal cells and lEC- 6 rat intestinal cell line (Jakobs and Paterson, 1986; V ijayalakshm i and B elt, 1988), LLC-PKi kidney cell line (G riffith er a/., 1992), and OK kidney epithelial cell (Doherty and Jarvis, 1993). This suggests th at th e airway epithelium like the absorptive intestinal and kidney epithelia (all o f which possess strong inw ardly directed electrochem ical Na^ gradients), display sim ilar characteristics in Na^-coupled nucleoside transport processes. Based on kinetic analysis, tw o Qrpes o f transporters are distinguishable, high afGnity transporter o r those w ith a low Km (<60 pM ) and low a ffin i^ transporters w ith a high Km (>60 pM ) (Paterson et al., 1981). In airw ay epithelial cells, total uridine uptake could be resolved into two com ponents: a N a^-dependent and a Na^-independent com ponent (Fig. 4-20). A saturable com ponent that conform s w ith M ichaelis-M enton kinetics, with a Km o f 3.4 ± 1.8 pM and Vmax o f 25 ± 5.2 pm oles/m g protein/30 sec was found. At lower concentrations th e Na^-dependent com ponent plays the m ajor role, and a t higher concentrations uptake appears to shift over to Na^-independent m echanism s, as the concentration gradient energizes the influx o f nucleosides via the equilibrative transporter. The nonspecific diffusional com ponent estim ated by a concentration dependence study at comprised o f <10% o f total uptake. The Km value o f the uridine transport in airw ay epithelial cells w as sim ilar to that o f Na^- dependent nucleoside transport reported in human kidney BBM V (4.76 pM ), but lower than bovine renal BBM V (10 pM ) (W illiam s and Jarvis, 1991), the rabbit choroid plexus (18 pM) (W u et al., 1992), and intestinal cultured cells (45 pM ) (Vijayalakshm i and B elt, 1988). The high afBnity/low capacity N a^:nucleoside cotransport generally 207 I Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. exhibit a 10-100 fold lower Km than that observed fo r the low afOnity/high capacity Na^-independent equilibrative transporters in hum an erythrocytes (Domin et al., 1993; M ahony et al., 1991), hum an lymphocytes (Chan et al., 1993), and rabbit kidney epithelia (W illiam s et al., 1989). 1.2. Inhibition Studies O f all the physiological substrates tested, m axim um inhibition o f ^H-uridine (83% ) was observed w ith 100 pM unlabeled uridine. The rem aining 17% probably represents the non-specific absorption o f uridine, the extent o f radioactivity associated w ith the cell m onolayer when uridine uptake was m easured a t 4°C o r that taking place via the Na^- independent component. The other physiological nucleosides inhibited uridine uptake to a sim ilar extent (75-80%) w ith the exception o f inosine (— 50% ). The nature o f inhibition w ith physiological nucleosides was com petitive in nature w ith an apparent K; o f 44 pM (Fig. 4-21). This K; was unexpectedly large com pared to the Km o f uridine, although this could be due to a low er afGnity o f adenosine (the inhibitor used in estim ating the K; by m eans o f a Dixon Plot) for the nucleoside transporter or could represent a hybrid value w ith contributions from two or more transporters (W illiam s and Jarvis, 1991). Interaction o f ^H-uridine w ith the Na^-driven sugar transporter, or the Na^-driven nucleobase transporter was ruled out as 100 pM ribose and uracil, each failed to inhibit uridine uptake (Table 4-11). Poor recognition o f nucleosides by the sugar and nucleobase transporters have been suggested due to specific interaction w ith the D- ribose m oiety. This is consistent w ith reports (Plagem aim et al., 1988) that state that 208 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. nucleoside transporters are selective for the (uranose rather than the pyranose sugar ring. Secondly, sugar transporters require a carbon-1 hydroxyl group w hich is absent in nucleosides. Thirdly, the nucleobase m oiety may sterically hinder interaction w ith the sugar transporter. Likew ise, in airway epithelial cells, it is likely that the D -ribose may hinder the binding o f nucleosides to the nucleobase transporters w hich have yet to be studied in the airw ay epithelia. Partial interaction was observed w ith D TP. How ever, it is likely th at m etabolism o f LH P by cell surface 5’-ectonucleotidases (C rack et aL. 1995) could generate firee uridine that may result in the 30% inhibition in uridine uptake. 1.3. Substrate Selectivity of the Nucleoside Transport Process Selectivity o f the nucleoside transporter to nucleoside analogs and physiological nucleosides (Fig. 4-22) em phasize the m inim um structural requirem ents for a substrate o f the nucleoside transport process. Firstly, nucleoside substrates do not interact w ith the sugar (ribose) or nucleobase (uracil and 5-fluorouracil) transporter. Second, nucleoside analogs w ith m odifications on the nucleobase (5-HUrd and 5-IU rd) are generally w ell tolerated. Third, the ribose m oiety was absolutely necessary for interaction w ith the nucleoside transporter system. Ribose sugar moiety^ w ith a 2'-trans- OH arabinose sugar (A ra-A and Ara-C) did not elicit appreciable interaction. Fourth, o f all the substituents on the ribose m oiety, the 3 -OH group w as crucial, rem oval (3 ’-dA, 2’-dA, ddA , and ddl) or substitution (AZT) at this position abolished interaction w ith the nucleoside transporter process. Finally, a cyclic sugar ring-structure appears to be 209 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. preferred over an acyclic sugar m oiety (ACV and HPM PC). Despite the conform ational flex ib ili^ o f acyclic analogs, w herein the hydroxyl side chain m oiety m ay be positioned sim ilarly to the 3 ’-OH in a nucleoside, interaction w ith the transporter is m inim al. This suggests a need fo r the com bined interaction w ith all the intact ribose-ring substituents for optim al interaction w ith the nucleoside transporter. Sim ilar observations on the m inim al structural requirem ents for a nucleoside transporter substrate have been reported previously (Plagem ann, 1991; B rett et al, 1993; Fang et a l, 1996; W u et a l, 1994). D iscrim ination against substrates that lack the 3 -OH group, and exhibit low tolerance for analogs w ith m odifications to the sugar have been reported in m ouse erythrocytes (G ati et a l, 1984), hum an renal BBM V (Brett et a l, 1993), rabbit choroid plexus (W u et a l, 1994), in m onkey kidney cells transiently expressed w ith a recom binant Na^inucleoside N 2 transporter (Fang et a l, 1996) and rat m acrophages (Plagem ann, 1991). Furtherm ore, the need for cyclic as opposed to acyclic sugars has been found in ganciclovir uptake by hum an erythrocytes (M ahony et al, 1991). 1.4. Uptake of Nucleoside Analogs ^H-2-Deo)^adenosine and ^H-HPMPC Since inhibition o f uridine uptake does not necessarily im ply uptake o f the com peting analog, the internalization (lum en to cytoplasm ic transport) o f antiviral analogs by direct m easurem ent o f ^H-2 ’ -deoxyadenosine (a transporter substrate) and ^H-HPMPC (a non-substrate based on inhibition studies) w as studied (Fig. 4-22). U sing ^H-2- deoxyadenosine and ^H-HPMPC, both therapeutically active analogs, the uptake o f ^H- 210 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. HPMPC uptake was about 5 tim es low er than ^H-2’deoxyadenosine (Fig. 4-23). Further, ^H-2’-deoxyadenosine but not ^H-HPMPC uptake was sensitive to Na^ and com petition w ith excess unlabeled (100 pM ) drug and uridine. Thus radiolabeled antiviral analogs that m eet the minimum criteria for a nucleoside transporter substrate were able to gain entry into th e intracellular compartm ent by a com m on m echanism as physiological nucleosides. T hose analogs that do not conform w ith the m inim um structural requirem ents, such as HPM PC, exhibit a m arkedly reduced uptake as a consequence o f their in a b ili^ to utilize the nucleoside transporter. L ike other polar solutes, HPM PC probably gets transported into cells gradually by nonspecific m echanism s such as fiuid-phase endocytosis (Connelley et aL, 1993), o r non-facilitated passive difiusion (Zim m erm an et aL, 1987; Domin et aL, 1993), dow n a concentration gradient. 2. Functional Identification of (he Nucleoside Transporters Functionally, m ultiple Na^ : nucleoside transporters have been reported in a num ber of cell types: For instance, in bovine kidney BBMV (W illiam s and Jarvis, 1991), tw o Na^- dependent transporters, N1 and N 2, that exhibit differing substrate selectivity, w ith guanosine and thym idine as m odel substrates, respectively, displayed a single saturable component w ith an apparent K m value for uridine o f ~10 pM . Likew ise, in m ouse intestinal epithelial IEC6 cells, evidence for N1 and N2 was reported exhibiting a sim ilar substrate selectiv i^ as the kidney BBMV, w ith a Km o f 45 pM (Vijayalakshm i and Belt, 1988). Recently, in ra t intestinal membrane vesicles, the transport mechanism 211 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o f nucleosides and 6 -m ercaptopurine riboside were found to show an inhibition pattern explained by the presence o f N1 and N3 N a^rnucieoside transporters (Iseki et aL, 1996). O n the lum inal surface o f airway epithelial cells, the functional existence o f two Na^znucleoside transporters N2 and N 4 w ere found. U sing the radiolabeled uridine, thym idine and inosine, inhibition with physiological nucleosides present evidence for at least tw o N a^-dependent nucleoside transporters w ith differing, but overlapping substrate selectivity’s (Table 4-11). ^H-Uridine uptake was strongly inhibited by thym idine, cytidine, uridine, adenosine and guanosine but not by inosine. This inhibition pattern is consistent with that observed in hum an kidney membrane vesicles (G utierrez et al, 1992; G utierrez and G iacom ini, 1993) where a broad substrate transporter N4, accepts all nucleosides as substrates except inosine. This broad substrate selectivity o f ^H-uridine inhibition pattern is different from the classical N3 transporter that accepts all nucleosides including inosine, reported in Caco-2 , and the m yeloid cell lines HL-60 and U937 (B elt et al, 1993), and the choroid plexus w ith a 2:1 Na^znucleoside stoichiom etry (W u et a l, 1992). The inhibition pattern o f ^H-inosine and ^H-thymidine suggested a transporter that excludes inosine (N 4), and one selective for pyrim idine nucleosides (N2). This observation was supported by the lack o f inhibition o f ^H-inosine uptake by purines and pyrim idines, that elim inates the involvem ent o f the classical purine-selective N 1 transporter (V ijayalakshm i and Belt, 1988; B elt et a l, 1993). On the other hand, ^H- thym idine uptake w as inhibited by the thym idine, cytidine and uridine (>70%), but not by purines guanosine and inosine (suggesting the involvem ent o f N 2). This data was 212 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. consistent with the inhibition pattern o f the classical N 2 transporter expressed in the rabbit intestine (Jarvis and G riffith, 1991), mouse intestinal cells (V ijayalakshm i and B elt, 1988), the rat and hum an renal Na^znucleoside transporter expressed in Xenopus laevis oocyte system (Giacom ini et aL, 1994), and the pyrim idine selective transporter in COS-1 monkey kidney cells (Fang et aL, 1996). These results were supported by the lack o f ^H-inosine sensitivity to N a \ together w ith the high Km o f 120 ± 24.8 pM and V m ax = 135 ± 25 pm oles/m g protein/30 sec, suggesting a low afGnity Na^-insensitive saturable uptake (Table 4-12) rem iniscent o f the facilitated difhision, ei type o f transporter (Paterson et aL, 1981). W hile a decrease in ^H-thymidine uptake in the absence o fN a \ along w ith the low K m = 21.5 ± 5.3 pM and high Vm ax = 29 ± 4.5 pm ols/m g protein/30 sec, reflects a selectivity for pyrim idine transport in airway epithelial cells (Table 4-12). It should be noted th at the m ethod o f distinction betw een the functional activity o f the transporters was based on inhibition studies which are not absolute. More specific protein structural identification m olecular techniques are needed to confirm these functional results. 3. Basolateral Localization Transepithelial fluxes o f nucleosides emphasize the overall absorptive nucleoside transport. Since the transepithelial perm eability o f uridine is 80-tim es higher than th at o f the paracellular m arker m annitol (T able 4-13), the intracellular uridine m ust have a m eans o f exiting the cell via the basolateral m em brane to reach the basolateral bathing fluid. Hence, the hypothesis that uridine once accum ulated w ithin the cells, on apical 213 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. application, is effluxed to the basolateral com partm ent dow n a concentration gradient via facilitated difhision, w as tested, sim ilar to the overall nucleoside transport scenario in the rabbit kidney (W illiam s et al., 1989). W illiam et al., (W illiam s et al.. 1989) reported that uridine uptake across the brush-border surface o f renal proximal tubules w as m ediated by N a^-dependent pathw ays (Km = 12 ± 3 |iM and Vmax = 3.9 ± 0.9 pm oles/m g protein/s), b u t perm eation o f the basolateral m em brane was m ediated by N a^-insensitive (K m = 137 ± 20 pM and Vm ax = 5.2 ± 0.6 pm ols/m g protein/s), but N BTI-sensitive facilitated difftision nucleoside carriers. N a^-independent facilitated difiusion transporter on the basolateral m em brane was also suggested in rats and rabbits utilizing single-pass intestinal perfusion studies (W aclaw ski and Sinko, 1996). In airway epithelial cells, basolateral uridine influx w as insensitive to Na^ but inhibited by 70% w ith high concentrations (10 pM ) o f N B T I (Fig. 4-25), suggesting that possibly an ei type o f equilibrative transporter m ay m ediated transport across the basolateral m em brane. E fflux experim ents w ith preloaded ^H-uridine dem onstrated efflux (from apically loaded cells m onolayers) into the basolateral fluid increased w ith tim e and w ith increased loading (Fig. 4-26A and 4-26B). These tw o lines o f evidence suggest that airway epithelial cells m ay possess a basolaterally localized bidirectional nucleoside transporter. It is likely that a passive com ponent to the overall basolateral efflux also exists, how ever, w ith the experim ental setup used in this study, the relative passive efflux com ponent cannot be determ ined. Com pared to the influx rate (tw o m inute tim e fram e), th e efflux rate (60 m in tim e fram e) w as m ore than 10-tim es low er at 0.5 finoles/m g protein/sec. 214 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. There are three lim itations in the evaluation o f basolateral efQux in cultured cells besides the w idely different experim ental tim e fram es. Firstly, prolonged incubation o f ^H -uridine w ith airw ay cells results in increased m etabolism . A bout 70% o f the radioactivity efQuxed to the basolateral fluid was intact uridine, about 10% coeluted w ith uracil and another 10% w ith nucleotide m etabolites on the TLC plate, the rem aining 10% was unidentified products. It is possible to block m etabolism with inhibitors o f en^rm es involved in nucleoside catabolism . However, the success o f this strategy in evaluating efflux still rem ains to be ascertained judging from the data in literature. Secondly, due to the rapid nature o f nucleoside transport, the intracellular concentration cannot be controlled and m aintained at high enough concentrations to observe linearity or saturability o f the efflux m echanism . Thirdly, from this data w hether a single m em brane protein is responsible for both events (basolateral influx and efflux) cannot be confirm ed. Thus, a direct com parison o f influx and efflux rates should be interpreted w ith caution, as further characterization o f basolateral nucleoside transport process is necessary. N onetheless, evidence o f basolateral nucleoside transport process in this study supports the involvem ent o f a carrier-m ediated process as one o f the m echanism s o f nucleoside release firom the basolateral m em brane. 4. Nucieoside Transportar Sites The relative abundance o f the concentrative and equilibrative transporters cannot be elucidated due to the lack o f a selective Na^-nucleoside transporter ligand. In the case o f equilibrative es transporters, using ^H-NBTI as a radioligand, the es transporters were 215 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. found to have binding constant in the range o f 0.1 to 1 nM in other cell types. Scatchard plots o f the site specific binding for NBTI revealed that the num ber o f es binding sites per cell varies w ith the cell type, ranging from 10^ to 10^ per cell in erythrocytes (Jarvis et ai, 1982) to >10^ sites in BeW b hum an choriocarcinom a cells (Boum ah et aL. 1994). In cultured intestinal epithelial cells, 1.5 x 10^ sites per ceil were estim ated w ith a dissociation constant K d o f 1.4 nM. Thus it is likely that airway epithlelial cells that were found to exhibit sim ilar nucleoside transport characteristics as intestinal epithelia, may share sim ilar observations. However, NBTI being lipophilic may perm eate the cell m em brane thereby binding specific intracellular membrane sites, leading to an overestim ate in the abundance o f cell membrane es transporters (Roovers and Meckling, 1996). A study on the dynam ics o f the es transporter in cultured chrom affin cells using photoafBnity labeling (Torres et a l, 1992) revealed that in the internalization step about 50-60% o f the transporter was destroyed, while the rem aining was recycled back to plasm a membranes (Torres et aL, 1992). 5. Nucleoside Transporter Regulation N ucleoside transporters are regulated by cellular events. Phorbol esters (phorbol 12- myri state 13-acetate, PM A) caused a direct increase in the Na^-dependent uridine transport in HL-60 leukem ia cells attributed to a dram atic increase in transport affinity (1.2 vs. 44 pM ) but not velocity (Lee et aL, 1991). O n the other hand, PM A had the opposite effect on facilitated difiusion, resulting in a corresponding rapid decrease in the rate o f facilitated difiusion. These modulations o f Na^-nucleoside transport activity were 216 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reversed w ith staurosporine suggesting regulation via PKC (Lee et al., 1991). PMA effect on the es facilitated difiusion transporters w as sim ilar in chromafGn cells where no apparent change in the afSnity for NBTI uptake was observed, whereas the total num ber o f NBTI binding sites decreased (Delicado et aL, 1994). These findings indicate that PKC closely regulates the balance between nucleoside transport mechanisms in mammalian cells. Purinergic regulation o f nucleoside transporters has also been reported. Using chromafGn plasma m em brane vesicles, ^ -u rid in e uptake was increased via the es transporters on addition o f ATP and other nonmetabolizable purinergic agonists (?% receptors). This increase was observed only in the Vmax but not the Km o f nucleoside transporters (Delicado et al, 1994), suggesting upregulation o f the num ber o f nucleoside transport sites. In contrast to the above findings, nucleoside uptake was not affected w ith P( purinergic agonists. ^H-Uridine uptake in guinea pig erythrocytes was not affected by P) adenosine receptor (A ; receptor) agonists. This suggests two im portant issues, first, nucleoside transporters are not regulated by P % purinergic receptors and second. Pi agonists have an independent structure-activity relationship for A2 receptors and the site for nucleoside transporters (Balwierczak et aL, 1989). Therefore, it appears that nucleoside transporters are exclusive fix)m the adenosine transporters. Concentrative nucleoside transporters are regulated by hormonal control (Gom ez- Angelats et aL, 1996). Na"^-dependent uridine uptake is stimulated by glucagon w ithin 10 min, and its action is m im icked by agents th at hypetpolarize the membrane (dibutyryl cAM P). Insulin on the other hand produces a more gradual stable induction o f nucleoside 217 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. transport activity. Therefore, hepatic concentrative nucieoside transporters are under short term horm onal regulation w ith glucagon, and im der long-term control by insulin involving the synthesis and insertion o f m ore carriers into the plasm a membrane (Gomez-Angelats et ai. 1996). O ther reported mechanisms o f nucleoside regulation include: polyunsaturated fatty acid th at stimulate purine but not pyrimidine transporter function in L1210 leukem ia cells (M artin and M eckling, 1996); long-term ethanol feeding decreased equilibrative nucleoside transport in hepatocytes independent o f a change in transporter number (W annam aker and N agy, 1995); and colony stimulating factor (CSF-1) w as found to upregulate es nucleoside transporter sites in activated but not in quiescent S 1 macrophage cultures (M eckling et al., 1993). 6. Possible Physiological Role of Nucleoside Transporters The physiological relevance o f nucleoside cotransporters is unknow n but, in general, it may serve tw o purposes: first, involves the rem oval o f solutes from the airway surface liquid. The clearance o f solutes from the surface fluid would prevent their accum ulation which otherw ise w ould have resulted in an osm otic drag o f fluid into the lum inal surface. Thus the nucleoside transporters (and other solute transport processes) along with the ion transporters m ay be involved in the m aintenance o f electrolyte and fluid balance, a crucial role o f the airw ay epithelium . Second, nucleoside transporters m ay be involved in the conservation o f nucleosides fix>m the lum inal environm ent. N ucleoside transporters have been hypothesized to be involved in nucleoside salvage in the intestine (He et at., 1994; W aclawski and Sinko, 1996), hepatocyte canalicular membrane (Che et 218 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. al.. 1992), kidneys (W illiam s and Jarvis, 1991) and the brain (W u et al., 1992). In intestinal epithelial cells, a high turnover rate for epithelial cells results in a high intrinsic requirem ent for nucleosides for DNA and RN A synthesis (C he et al., 1992). In the airways, aged and damaged surface cells o f the epithelial lining are constantly replenished by secretory stem cells or basal cells (N ettesheim et al., 1990). However, it replicates at a slow er rate than the cells o f the intestinal epithelium (Fulm er and Crystal, 1976), approxim ately 4% daily in the larger airways (K ilbum , 1974). D espite the lower epithelial ceil turnover, their intrinsic requirem ent fo r nucleosides m ay not be met by de novo synthesis o f nucleosides. Consequently, salvage o f luminal extracellular nucleosides may be necessary. In the conducting airw ay, the extracellular source o f these nucleosides could potentially arise firom nucleosides, nucleotides o r DNA associated with cell debris o f sloughed-o£f cells or dead m icroorganism s in the lumen environm ent o f the m ucus blanket, o r those secreted along with m ucus glycoproteins firom secretory cells and submucosal glands. M em brane-bound ectonucleotidases may play a crucial role in this salvage operation. In the rabbit tracheal epithelial luminal surface, 5 ' -ectonucleoddase activity has been reported (C rack et al., 1995), w hich may fim ction in the hydrolysis o f luminal nucleotides and polynucleotides leading to form ation o f nucleosides th at can be conserved. This has been tested w ith the ectonucleotidase inhibitor FPL 67156. Interestingly, Che et a/. (Che et al., 1992) dem onstrated that ectonucleotidases on the lum inal surface (cananicular m em brane) ofheptocytes are fim ctionally linked to nucleoside carriers. M oreover, 5’-ectonucleotidases activity together w ith nucleoside 219 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. uptake were demonstrated to provide all the cellular purine requirem ents in m itogen- stim ulated T cells and hum an B lym phoblastoid cells (Thompson, 1985). Therefore, it is likely that the residual nucleoside content o f the airw ay lumen m ay be salvaged v ia the high afBnity Na^znucleoside cotransporters on the epithelial apical surface, however, the role o f the facilitated difhision transporters rem ains unclear. Im portantly, firom a therapeutic standpoint the presence o f carrier-m ediated nucleoside transport process on the lum inal surface o f the conducting airway, can be exploited to optim ize the delivery o f antiviral nucleoside analogs to the airways. 7. Rational Design o f Antiviral Analogs A curious question exam ined was how sim ilar are the structural requirements that restrict entry o f nucleoside analogs to cells com pared to those found in structure-activity relationships (S AR>? SAR studies o f nucleoside analogs revealed that for optim um anti herpes sim plex virus (HSV) and anti-hum an imm imodeficiency virus (HTV) activity, the substituent at the 5-position on the base should be unsaturated, not larger than 2-4 carbon atom s in length, w ith a hydrophobic electronegative m oiety (Goodchild et a i, 1983; C hu et al. 1989). Substitutions a t C-5 w ith H , CH j, and C2H5 yielded analogs w ith high antiviral activity, whereas, substitutions greater than C2 including brom ovinyl significantly reduced activity. Changing o f the nucleobase fiom thym idine to uracil o r cytosine produced analogs w ith high potency ( E C 5 0 o f 0.2 to 1.2 p M) and reduced toxicity. Studying the binding affinities o f sugar-m odified analogs to HSV-1 and HSV-2 thym idine kinase in cell free isolated enzyme preparations, Cheng and coworkers (Zou 220 i Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. et aL. 1984) concluded that m ost 3’ and 5'-substituted analogs o f brom o- o r chloro- deoxyuridine show ed little antiviral activity. The Ki values varied depending on the source o f target viral thym idine kinases (TK), suggesting th at the T K 's d iffer in term s o f th eir catalytic binding site. M oreover, anti-HSV activity did not closely correlate w ith the respective binding afBnities. This suggested that phosphorylation and o th er factors m ay contribute to selective antiviral action (Zou et aL, 1984). In a later study, triphosphate nucleotide analog incorporation by viral DNA polym erase at prim er term ini (that elim inates confounding phosphorylation steps) was exam ined by Frank et aL, (Frank et aL, 1985). In general, analogs w ith m odification on the base w ere less effective inhibitors o f DNA elongation com pared to sugar m odified analogs. W ith acyclic nucleoside analogs such as acyclovir and ganciclovir phosphate the high afOnity for the viral enzym es m ay be explained by the steric freedom o f these analogs to conform ationally accom m odate the viral enzymes. Therefore, the structural requirem ents for binding to viral enzym es and / or activity for a nucleoside analogs is in reasonable agreem ent w ith the structural requirem ents for cellular uptake via the nucleoside transporters. The m ajor differences in the structural requirem ents between uptake and activity appears to be in the presence o f a structural substitution at the 3 -OH position and an intact sugar ring v s. acyclic sugar nucleoside analog. The lack o f 3 -OH o r substitution at that position, and acyclic sugar nucleoside analogs show dim inished interaction w ith the nucleoside transporters therefore, excluding cellular entry via active m echanism s. O n the other hand, these analogs may exhibit good binding affinity for viral enzym es and/or antiviral activity. 221 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Thus it appears that there is a higher stringency for cellular uptake w hile activity is dictated by a criterion th at is broadly selective typical fo r a particular target viral anyone. 8. Feasibility o f Airway Antiviral Nucieoside Drug Delivery To date, ribavirin is only antiviral agent approved by the US Food and Drug A dm inistration for hum an use as an aerosol for infants w ith serious infections due to the respiratory syncytial virus. R ecently there has been considerable interest in the utilization o f ribavirin in the treatm ent o f infections due to HTV. The deposition o f aerosolized ribavirin for 6 m onth old infants and 25 year old adults was approxim ately 55*70% o f inhaled drug (K night and G ilbert, 1988). For 6 m onth old infants, about 57% o f the inhaled drug was deposited in the tracheobronchial area and ju st over 5% in the distal pulm onary area. W ith adults, about 27% w as found in the tracheobronchial are and about 25% in the distal pulm onary area. The sm aller dim ensions o f the infant tracheobronchial tree accounts for the relatively higher deposition in the conducting airway. A fter a 8-12 hr period o f treatm ent w ith 200 pg/1 ribavirin aerosol, the concentration in the respiratory lining fluid (the site o f deposited inhaled drug) was very high at 1.2 m M , w ith a peak lung concentration o f 165 nm oles/gm in hom ogenized lung tissue in 2-4 hr, 8 -tim es higher that the m axim um serum concentrations o f 21 pM after 12 hr treatm ent (Knight and G ilbert, 1988). In another antiviral aerosol delivery study. Debs et ai. (D ebs et ai., 1988) exam ined in a m urine cytom egalovirus model the activity o f aerosol vs. oral or system ic ganciclovir and phosphonoform ic acid. The authors 222 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. report th at aerosol delivery was specific for the lung, being ineffective in inhibiting virus in the salivary glands. The overall level o f antiviral activity in the lung was equivalent to system ic drug adm inistration. These studies clearly dem onstrate th at local delivery o f antiviral nucleoside analogs to the respiratory tract by inhalation results in significant absorption fiom the airw ay lum en leading to im proved antiviral efficacy for local respiratory diseases w hile avoiding extrapulm onary toxicity (Debs et aL, 1988; M ontgom ery er a/., 1987). T he oral bioavailabilities o f nucleoside analogs varies w idely due to factors such as hepatic first-pass m etabolism , variability in the gastrointestinal tract, patient health, and interaction w ith food and other m edications. O ral bioavailability o f AZT, d d l and acyclovir w ere found also to be low and variable due to a com bination o f their inability to traverse the apical epithelial m em brane, low enzym atic and chem ical stability in the gastrointestinal environm ent (W aclawski and Sinko, 1996; Knupp e t a i, 1991; Y archoan e t a i, 1990). In a random ized double-blind placebo-controlled study in HIV- positive hum an subjects, oral delivery o f HPM PC at a maximum dose o f 10 m g/kg had a very poor bioavailability o f <5% (W achsman e t a i, 1996). The low intrinsic perm eability o f HPMPC in airw ay epithelial cells and Vero cells via slow nonspecific diffiisional or fiuid-phase m echanism s (Connelley et a i, 1993) m ay occur in intestinal epithelia thereby possibly leading to low bioavailability. Stavudine (d4T) on the other hand, has a high bioavailability (70-90% ) w ith low variability, accounted for by saturable Na^-dependent and Na^-independent nucleoside transport mechanisms (W aclaw ski and Sinko, 1996). Taken together, these observations emphasize that 223 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. b ioavailabili^ is a function o f epithelial absorption. Therefore, by enhancing delivery o f nucleoside analogs to cells, the clinical or therapeutic concentrations would be achievable optim izing therapeutic benefit o f the drug. D. VIRUS INFECTION STUDIES OF THE AIRWAY EPITHELIUM 1. Infection o f Airway Epithelial Cells The selection o f a specific virus and a target system are im portant in studying viral in&ction. In cell culture often, cell layers are resistant to infection by virus, this m ay be either due to infection o f a particular cell type that m ay not support viral replication, or the lack o f specific cell surface receptors th at m ay m ediate the internalization o f virus particles (Tucker and Compans, 1992). In this study, the infection o f rabbit airway epithelial cells w ith hum an Ad5 w as exam ined. Human Ad5, infected rabbit airway epithelial cells grow n in culture. The extent o f infection (infectivity) was determ ined by estim ating viral replication (virus grow th curve), im m unofluorescent staining o f viral antigen, and the tim e-dependent detection o f viral antigen released into the bathing media. V irus grow th occurred at a m ultiplicity o f infection (MOI) o f 1:1 and 20:1 (Fig. 4-27 ). A ^ i c a l grow th curve was foimd w ith Ad5 W T replication in airway epithelial cells w ith the grow th phase peaking at 48 hr post-infection w ith a two log increase in virus yield. Com paring the two virus concentrations used to infect cells, virus yield 224 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. increased a t higher M OI. T his suggests that cultured rabbit airway epithelial cells express specific cell surface receptors to facilitate the internalization o f hum an Ad5 over the tw o hour adsorption period. Virus infection o f cultured airway epithelial cells was confirm ed w ith imm unofluorescence staining. Putative staining was observed throughout the culture (Fig. 4-28A and 4-28B), on closer exam ination at a higher m agnification the fluorescence was found to be localized to the perinuclear and nuclear regions o f infected cells about 24 h r post-infection. This is in agreem ent w ith reports that viral gene products accum ulate in the nuclear region, th at the further dem onstrated that the pathogenesis o f infection is a consequence o f adenoviral early gene function (G insberg e / ûr/., 1990). The extent o f virus release from infected cells to the bathing m edium increased over tim e in a sim ilar m anner as did the virus growth curve (Fig. 4-29). A fter a 24 hr latent period the production o f viral antigen was abundant reaching a m axim um after 72 hr. Thus this technique provides a rapid and convenient (w ithin 70 m in) w ay o f estim ating virus infection (Lipson, 1989). However, there are a few draw backs. It is sem iquantitative, ju st the relative increase or decrease in viral antigen released to the surrounding m edium can be assessed, as there is no standard to estim ate the precise concentration o f virus protein in sam ples. Secondly, the E lA is a m easure o f virus protein, not intact assem bled virus particles. Thus, serves as an indirect indicator o f virus replication. Further, an excess am ount o f virus protein m ay be synthesized than required for virus assem bly. Only 10% o f viral DNA and structural proteins synthesized are packaged into new virions, the rem ainder accum ulates in the nuclear region 225 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (G insberg et aL, 1990). N onetheless, th is data confirm ed the production and release o f virus protein in a tim e dependent m anner. O n visualization under norm al o r phase-contrast m icroscopy, the cell layers did not appear to exhibit any signs o f cytophathic effects (CPE) in early Ad5 infection. The absence o f CPE in m onolayers despite virus replication can be explained by the possibility o f non-lytic virus effects o n epithelial cells. In Caco-2 and M DCK cells (Svensson et aL. 1991), filter grow n cells w ere surprisingly resistant to CPE w ith rotavirus infection com pared to cells cultured on an im perm eable support (plastic). Therefore, though CPE m ay serve as easy visual indicator o f infection, the lack o f CPE does not im ply the lack o f infection. 2. Barrier Pmperties and Adenovirus Type 5 Inflection V irus infection has been reported to alter epithelial structure and fim ction. In the airw ays infection is associated w ith a disruption in the epithelial barrier properties causing an increase in epithelial perm eability. In the airw ay epithelium , infection results in abnorm al secretions via alteration o f active and passive ion transport m echanism s (C loutier et aL. 1989), virus-induced m ediator release and sm ooth-m uscle hyperreactivity (Sterk, 1993; B ecker e t aL. 1993). In the cultured rabbit airway epithelium Ad5 infection m ildly affects epithelial barrier fimction at early stages o f infection. This conclusion is based o n changes in the bioelectric properties o f airway cell cultures and on the transepithelial perm eability o f the paracellular m arker, ^H- m annitol. AdS W T infection o f airw ay epithelial cells at an MOI o f 20:1 decreased the 226 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. spontaneous PD by 40-60% , 48-60hr post-infection (Fig. 4-30A ) compared to mock- infected (M I) control cell m onolayers that were inoculated w ith firesh medium containing no virus. This change in PD was accom panied by a 40% decrease in R t 60 hr post-infection (Fig. 4-30B). A decrease in the epithelial R t did not coincide w ith a decrease in the Papp o f mannitol in AdS infected cell layers. Forty-eight hr post-infection, although mannitol perm eability increased 1.4 tim es over M l-control m onolayers (Fig. 4-31 A), these values are not statistically different (p>0.05). This suggests a trend th at the airway epithelial junctional properties m ay be slightly altered under infected conditions, but not severely enough so as to increase solute perm eability. The cell associated ^H-mannitol content after a 3 hr transport experim ent was not altered in infected cells (Fig. 43 IB ), which reflects no change in the integrity o f the cell membrane. Sim ilar results in the lack o f change in m arker perm eability was found in bronchial epithelial cells infected with Pseudomonas aeroginosa w hen compared to uninfected controls (Stutts et aL, 1986). Taken together, it is reasonable to conclude that Ad5 affects airw ay epithelial tight junctions is selective. The fact that the epithelial Rt is decreased, but not solute flow argues that barrier function is retained w hile its structure m ay be slightly altered. However, further studies are necessary to confirm this observation. The effect o f virus on an epithelia appears to be specific for the virus (based on the viral pathogenesis) and the target cells (cellular response and inununi^). In the cotton rat, respiratory syncytial virus infection resulted in a 28% decrease in Isc and a 100% increase in Rt. The decrease in Isc was attributed to a decrease in the Cl* secretion 227 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. w hile the increase in R t was associated w ith significant restriction o f passive movement o f Na^ and Cl through the paracellular pathw ay induced by a change in the tight junction structure (Cloutier et aL, 1989). In bronchial epithelium . Pseudomonas aeroginosa decreased R t by ~40% ascribed to a 72% drop in net Na^ flux and a 50% decrease in unidirectional C l'-fiux, but no change in ^^C-mannitol perm eability (Stutts et aL, 1986). In African green monkey kidney cells (M A-104), rotavirus infection alters epithelial membrane perm eability to Na^ and The perm eability o f the cell membrane was increased, as activity o f the N a ^ ^ ATPase pump or the N a^^2 C T cotransport activity was not affected. These changes in th e cell plasm a membrane leakiness coincided w ith the onset o f viral protein synthesis (del C astillo e t aL. 1991). Collectively, it is possible that virus infection o f airw ay epithelial cells disrupts epithelial tight junctions by inhibiting synthesis o f tight junctional proteins at later stages o f infection (>60 hr). T his speculation, however, needs to be confirm ed by direct m easurem ent o f the structural presence and am ounts o f junctional proteins such as ZO-1, cadherins, desm odgleins, and connexins. 3. Inflection and Carrier-madiated Nucleosida Transport Function Viruses have been reported to induce changes in epithelial barrier properties, ion transport properties, epithelial perm eabilization (Carrasco, 1978), and macromolecule synthesis (Carpio et aL, 1981). In airway epithelial cells, evidence was foimd that nucleoside transport function w as altered in m onolayers infected w ith AdS. The difference in uridine uptake betw een infected and M l-m onolayers was significant 228 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (p<0.05) only on prolonged uridine incubation (5 m in) and n o t on short-duration o f incubation (30 s) (Fig. 4-32). M oreover, the decrease in ^H-uridine uptake o f 30-35% was not dependent on the presence o f N a \ Thus AdS infection o f airw ay epithelial cells does not selectively affect Na^-dependent or N a^-independent nucleoside transport processes (Fig. 4-32). These prelim inary data suggest that AdS infection does cause m inor alteration in nucleoside carrier-m ediated transport function. Since a decrease in uptake seem s to be noticeable only on prolonged incubation it is possible that AdS infection affects the velocity (V m ax) o f uridine transport. This could be interpreted as a decrease in the num ber o f transporters due to adenoviral disruption o f cellular protein synthesis, or direct or indirect m odulation o f functional nucleoside transporters or the alteration o f nascent transporter trafBcking and insertion in the apical cell m em brane. However, further evidence is needed to substantiate this speculation. In H eLa S3 cells infected w ith herpes sim plex virus (H SV -1) the rate o f thymidine uptake increased 1-8 hr post-infection (Palu et aL, 1990). D espite the inhibition o f host cell protein synthesis beyond 8 hr post-infection, the number and affinities o f nucleoside transporters were unchanged. Furthermore, increased thym idine transport was correlated well with the uptake o f the membrane impermeable solute sucrose. This suggested that the es nucleoside transporter in HeLa S3 cells rem ains intact and th at passive diffusion was increased w ith the progression o f viral infection (Palu et aL, 1990). A long the same lines o f reasoning, in airway epithelial cells, it is likely that the m em brane leakiness o f infected cells is increased. Since uridine concentrations w ithin cells is lai^er than that in the 229 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. extracellular fluid after S m in it is likely th at uridine may difluse out o f the cells com m ensurate w ith virus release thus giving rise to a lower cell associated radioactivi^ com pared to uninfected airw ay epithelial cells. However, the lack o f accumulation o f m annitol and uridne (in the absence o f N à ^ in infected cells argues against this explanation. Therefore, m ore studies need to be done to ascertain the basis for the change in nucleoside transport follow ing viral infection. S im ilar reports on decrease in carrier-m ediated uptake has been reported in early influenza infection o f M DCK cells (G urevitz e t al., 1987). Three hours after infection, the uptake o f inorganic phosphate, 2 -deoxyglucose and a-am inoisobutyruc acid decreased by 10-20% com pared to uninfected controls. The kinetic param eters for saturable phosphate uptake revealed th at the Km was not significantly changed, but the Vm ax w as reduced by 30% . Likewise, in porcine sm all intestinal brush-border m em brane vesicles, transm issible gastrointestinal virus (TGV) resulted in a decrease in the N a-d ep en d en t alanine and glucose influx (Rhoads et al, 1989) com pared to those o f uninfected controls. In conclusion, this virus study in airw ay epithelium dem onstrated that hum an AdS w as capable o f infecting and replicating in rabbit airway epithelial cells grow n in culture. T herefore, could serve as an infection-m odel to study viral infection on host cell structure and function. This study also dem onstrates that nucleoside transport, although reduced, w as present despite viral infection and could still accum ulate antiviral 230 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. nucleoside analogs to a degree, that are substrate for the nucieoside transporters in airw ay epithelial cells. E. IN VITRO-IN VIVO CORRELATION E pithelial cell culture m odels for drug transport studies could serve as a predictive tool to estim ate the in vivo absorption behavior based on in vitro results. In order to optim ize the utility and validity o f a cell culture m odel, a thorough understanding o f the factors influencing in vivo drug absorption at both cellular and m olecular levels is required. The absence o f system ic blood flow and nervous control in the cell culture system s along w ith the lack o f m otility o f the lung and m ucociliary clearance and other physiological factors m ay play a signiflcant role in correlating drug transport in culture to that in vivo. N onetheless, given that the rate-lim iting step in drug absorption is the barrier properties o f the epithelial lining, it should be possible to com pare drug perm eability in epithelial cells cultured under conditions that m im ic the in vivo environm ent and that closely resem ble the native tissue, w ith in vivo absorption. There have been a num ber o f studies on characterization o f barrier properties o f respiratory epithelium in both in vivo and isolated tissue models. Table 5-3 sum m arizes the relationship between m olecular size and transepithelial transport o f hydrophilic com pounds. The perm eability o f cultured respiratory epithelial cell cultures correlated reasonably w ell with those o f in vivo and other in vitro models for drugs that opt for the 231 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. paracellular pathw ay. Several equivalent pore m odels have been suggested to account for the observed p erm eab ili^ o f hydrophilic compounds, depending on the species and on the nature o f drug studied in vitro and in vivo o r isolated lung preparations. A lthough heterogeneity o f equivalent pore populations in respiratory epithelium has been w idely suggested, tw o m ajor pore populations are believed to exist, a sm all pore populations with radius o f 0.5-1 m n allow the passage o f ions and other sm all electrolyte solutes, and a larger pore population w ith a radius o f 2-25 nm, perm its the translocation o f m acrom olecules (K im and Crandall, 1983; Taylor, 1990). In Table 5-3, th e perm eability o f A549 lung carcinom a cells to FITC-dextrans is strikingly different from other m odels. The Papp values appear to be about 100-fold higher than prim ary cultured alveolar epithelial cells (Kobayashi et al., 1995; M atsukawa et al., 1997). This m ay be attributed to differences in: (a) cellular m orphology (A549 being prim arily type H in nature w hile prim ary cultured cells are type-I-like in nature), and (b) R t (A549 cells, 100 - 764 ohms.cm^ vs. prim ary alveolar cells -2 ,0 0 0 ohms.cm^). The A549 lung carcinom a cell line has very noticeable differences from native epithelia. First, they do not express sufGcient phenotypic features o f alveolar cells to allow determ ination o f their origin and do not differentiate into cells th at exhibit type I-like characteristics. Second, A549 cells lack phosphatidyl glycerol and surfactant protein SP-A foimd in rat and hum an alveolar type II cells (Korst et al., 1995; M ason and W illiams, 1980). Third they do not form dom es th at are indicative o f active ion transport (Robinson et al., 1984). Therefore, these differences in m orphological and biochem ical characteristics indicate that the structural and functional 232 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. characteristics o f A 549 ceils are not representative o f native epithelia. M oreover, the two log higher perm eability o f FITC-dextrans, underscores the lack o f functional restriction offered by the cell line in the difiusional translocation o f m acrom olecules. Collectively, these argum ents reiterate and support the notion that prim ary cultured respiratory epithelial cells represent a suitable in vitro m odel for the evaluation o f drug transport. Since the m odel p-blockers exhibit significant pharm acological effects (tracheobronchial constriction) on the airways in vivo, no inform ation is available on the in vivo pulm onary absorption o f p-blockers for direct com parison o f transcellular difiusional perm eation. A num ber o f reports by Schanker and his co-w orkers docum ented that the perm eability o f drugs in the rat lung w as dependent on the partition coefficient o f the drug (Schanker, 1978; Enna and Schanker, 1972; Lanm an e t a i, 1973). Later, Taylor (T aylor, 1990) com piled the available data on the absorption rate o f 21 compounds (cardiac glycosides, antibiotics, organic anions, organic cations and steroids) w ith a w ide range o f lipophilicity in the rat lung in vivo, and foimd that the absorption rate o f lipophilic com pounds was over 100-tim es higher than the hydrophilic compounds. A plot o f the absorption rate versus the lipophilicity o f these compounds revealed a sim ila r trend as foim d for the airw ay (M athias e t a i, 1996) and the alveolar epithelial (Saha et al., 1994) culture m odels that describe th is influence o f lipophilicity on drug absorption. 233 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Evidence fo r transcellular transport o f drugs via specialized m em brane-bound proteins (carriers and receptors) involved in the translocation o f drugs across cell m em branes has been studied (Joris and Q uinton, 1989; C loutier and Lesniak, 1985; Y am ashita et al., 1996). However, adequate in vivo inform ation on drugs that utilize these carrier-proteins is not available to establish a correlation o r trend w ith that found in vitro. In Fig. 5-4, a schem atic representation o f the all transport proteins involved in ion, solute, and m acrom olecular drug transport in the conducting airway epithelium in culture as w ell as in other in vitro m odels is depicted. In the trachea, the apically located ion channels include the Na^- and the C l -channels. The basolaterally located ion transport processes include the Na^K^ATPase, Na^(2C1*)K^ cotransporter, -channel, and the pH regulators Na^/H^ antiport and CI /HCO3 antiport. Transport processes such as the Na^-glucose cotransporter (B asset e t a l, 1988; Saum on e t a l, 1990), and the anion transporter (E nna and Schanker, 1973) m ay m ediate the active transport o f solutes. O ther active solute processes such as the Na^m ucleoside transport processes and H^-coupIed dipeptide transporter (Y am ashita et a l, 1996) have been suggested. Carriers for the internalization o f therapeutic proteins and genes include the polym eric im m unoglobulin receptor (pIgR) (Ferkol et a l, 1995), the album in receptor (A lb) (M a et a l, 1993; Johnson e t a l, 1989) and the transferrin receptor (Tf), may also be present. T ransport processes for w hich evidence is uncertain are denoted by W ith the increasing interest in the respiratory tract as an attractive route o f drug delivery in vivo and the developm ent o f new in vitro epithelial cell culture m odels, it w ould be possible to directly com pare the bioavailablility o f aerosolized drug or 234 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. localization o f inhaled dnigs w ith the extent o f drag absorption o r cellular retention in culture, respectively. The acquisition o f such data, w ould lead to useful inferences on in vitro-in vivo correlation and enable a prediction o f the absorption o f drag candidates intended fo r inhalation delivery. 235 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 00 00 ■ ► oc i a .S i I a I • o g I (w O 1 i s ê 236 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VI. CONCLUSION 237 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A. SUMMARY OF FINDINGS A prim ary cell culture m odel o f isolated tracheal epithelial cells was developed in air- interfaced conditions. Com pared to liquid-covered subm erged cultures (LCC), the air- interfaced cultures (A IC) exhibit m orphological, electrical, and ion transport properties th at mimic the native epithelium . A ir-exposure was foim d to stim ulate grow th and differentiation, ciliogenesis, and im proves the electrical properties o f epithelial m onolayers. A lm ost all o f the baseline Isc was accounted fiar by Na% Cl" and HCO3' m ovem ent across the m em brane w ith air-interfaced cultures displaying a 70% am iloride-sensitive N a^ current and 10% C l current. In contrast, LCC dem onstrated a 50% Na^ current and 35% Cl current. A naerobic cellular respiration (m easured by the am ount o f lactate secreted to the bathing m edia) was 43% higher in LCC than in AIC. Taken together, the im proved differentiation o f the air-interfaced cultures, along w ith ion transport properties and characteristics o f cellular respiration closely resem ble the native epithelium , thus, making it the m odel o f choice for the study o f drug transport processes in the conducting airway. In order to understand the perm eability characteristics o f AIC to exogenous solutes, the influence o f m olecular size and solute lipophilicity was studied. Using m annitol and FIT C -dextrans in the size range o f 4,000 to 70,000 daltons as hydrophilic m arkers the perm eability was found to be inversely related to size up to a m olecular cut o ff lim it o f 20, 000 daltons. This was consistent w ith the existence o f a single 238 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. population o f equivalent pores o f 5 nm radius that restricts the transepithelial diffusion o f m acrom olecules greater than 20,000 daltons. W hen solutes o f the sam e size were studied, perm eability was found to be dictated by the lipophilicity o f the solute. A sigm oidal relationship exists betw een drug perm eability and lipophilicity with a h alf maximal perm eability occurring a t a log partition coefficient o f 2.08. Collectively, these perm eability characteristics w ere sim ilar to that found in the native tissue or other respiratory epithelial models w hich validates the airw ay epithelial cell culture m odel for the study o f drug transport. In the second phase o f the research, the air-interfaced airw ay epithelial cell culture m odel was used to evaluate the nucleoside uptake to establish evidence for a nucleoside transport process, previously not known to exist in the conducting airway. Using uridine as a model nucleoside, nucleoside uptake was rapid, tem perature and direction dependent. Uptake w as strongly dependent o n the presence o f N a \ coupling w ith nucleoside w ith a stoichiom etry o f 1:1. Uridine uptake was saturable (Km = 3.4 pM , Vm ax = 0.7 pm oles/mg protein/30s) and insensitive to the nucleoside transport inhibitor N B T l. A ll physiological nucleosides were effective in com peting with uridine uptake, w here the inhibition w as found to be com petitive in nature w ith a Ki o f 44 pM . Substrate selectivity studies w ith structurally sim ilar nucleoside analogs revealed that base m odified analogs generally interact with the nucleoside transporters, whereas, those analogs w ith substitution on the sugar ring are less likely to in teract An intact ribose ring structure with an unm odified 3 -OH was essential for nucleoside transport 239 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. activity. These structural requirem ents for nucleoside analogs are consistent w ith that reported for Na^-nucleoside transporters in the kidney, brain, and alveolar m acrophages. Based on functional inhibition o f ^H-uridine, ^H-thymidine and ^H-inosine w ith physiological nucleosides, tw o Na^-dependent nucleoside transporters, the pyrim idine selective N 2 and broad substrate selective N 4, as w ell as the ei equilibrative transporter are present on the apical m em brane o f the airway epithelium . W hile on the basolateral m em brane an e i equilibrative transporter m ay m ediate nucleoside tranpsort across the basolateral m em brane. The overall scheme o f nucleoside transport in airw ay epithelia cells involves the rapid internalization o f nucleosides and nucleoside analogs via the Na^-driven N2 and N 4 nucleoside transporter. The equilibrative transporter may also function in nucleoside influx except th eir relative role com es into play only at high concentrations. Once accum ulated w ith the intracellular com partm ent, efflux o f nucleosides may occur dow n a concentration gradient to the basolateral com partm ent or the apical com partm ent via the ei equilibrated transporters located on the apical and basolateral cell m em branes, leading to a steady state in nucleoside flux in and out o f the cell. Therefore, the hom eostasis o f physiological nucleosides and nucleoside analogs that interact w ith the nucleoside transporters w ithin the cell, involves the functioning o f all these transporters in concert w ith each other. In the final stages o f the research, the effect o f virus infection on nucleoside transport function was exam ined. A t first, an infection m odel was developed w ith hum an adenovirus (AdS) infection o f rabbit airway epithelial cells. The infectivity and 240 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. replication was studied by plaque assay, im m unofluorescence staining and the Adenoclone enzym e imm unoassay for release virus antigen. V irus replication exhibited a typical growth curve with a tw o-fold increase in virus yield 48 h r after infection. The bioelectric properties were altered to a m inor extent 48 hr after infection. However, no disruption in barrier properties w as observed. N ucleoside transport function was also found to be slightly altered at early stages o f infection. The uptake o f uridine w as decreased by 30-35% in infected monolayers com pared to controls. Therefore, despite the m inor reduction in transporter activity it is likely that cells w ould still accum ulate nucleoside drugs that are substrates for the nucleoside transporter in the airw ay epithelium . 241 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 3 ( D T3 O Q . C s Q . T3 ( D W W o' 3 O CD 8 3 . ( Q 3 CD C 3 - 3 " CD CD T3 O Q . C S - . o 3 T3 O CD Q . O C T3 CD C / J C / J o' 3 pyrimidines purines and pyrimidines Na+ Nucleoside Na+ Nucleoside Nucleoside Nucleoside Apical Basolateral concentrative ^ equilibrative •>’ passive Fig. 6-1: Schematic representation of nucleoside transport in airway epithelial cell monolayers B. SIGNIFICANCE OF THE FINDINGS T he research in th is project has resulted in the developm ent o f a new cell culture model o f air-interfaced rabbit tracheal epithelial cells. The epithelial cells in this model closely resem bled native tracheal epithelium in structure and function, thus validating the model for the evaluation o f airway epithelial ion and drug tran sp o rt There are three major areas this research has significantly contributed to: First the finding that exposure o f the airw ay epithelium to air stim ulates cellular differentiation, im proves the electrical and ion transport properties, and influences cellular respiration. Secondly, the perm eability characteristics o f the airw ay epithelium to drugs that opt for the paracellular, transcellular and carrier-m ediated pathw ays have been studied and compared, thus laying the foundation o f evaluating the feasibility o f therapeutically active m acromolecule (proteins, peptides and genes) and sm all solutes delivery to the respiratory epithelium . Thirdly, nucleoside transport processes were identified on the airw ay epithelial lum inal surface that can utilized for the site-selective targeting o f nucleoside analogs to the respiratory epithelium . In the first case, our observations underscores the importance o f culture condition in the prim ary culturing o f airw ay epithelial cells (Johnson e t al., 1993; Yam aya et al.. 1992; Mathias e t al., 1995) and im derm ines the effects conventional subm erged culturing creates for airw ay epithelial cells. This air-interfaced cell culture m odel might be useful to understand cellular processes in healthy and diseased states 243 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. & such as altered electrolyte and fluid balance as in cystic fibrosis, viral infection, and inflam m ation (W iddicom be, 1986; Reiss e ta l., 1991; R obison and Kim, 1995), and drug transport characteristics o f drugs intended for inhalation delivery to the respiratory tract (M athias et aL, 1996; Yam ashita et a l, 1996). A n area o f considerable interest is the identification o r cataloguing o f solute transport processes th at could be taken advantage o f to delivery drugs that are structurally sim ilar to the natural substrates. Examples o f this include: nucleobase transporters, dipeptide transporters, m onocarboxylic acid transporters, organic anion and cation transporters. The elucidation o f the perm eability characteristics o f the airway epithelium to drugs o f varying sizes and lipophilicity brings on the next tier o f inquiries that focuses on the evaluation o f therapeutically active m acrom olecules and sm all solutes in need o f alternate routes o f delivery, such as insulin, grow th horm one, and cytokines. These studies would shed light on the feasibility o f protein and peptide drug absorption and m etabolism , thus setting the stage for selecting drug candidates that exhibit favorable perm eability and enzym atic stability in the respiratory tract. A nother application o f enormous potential is gene delivery to the respiratory tract (Lu et a l, 1989; Curiel et a l, 1992; Ferkol et a l, 1995) currently underw ay fo r the correction o f cystic fibrosis and tti-antitrypsin deficiency (Knowles et a l, 1995; Stanley et a l, 1991). New strategies to facilitate gene delivery to cells can be explored including, the use o f specific receptor such as transferrin, the polym eric im m unoglobulin receptor and the surfactant protein class o f receptors, known to be present in the airw ay epithelium (Curiel et a l, 1992; 244 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ferkol e t al., 1995; Baatz et al., 1994). O ther aspects o f drug delivery w here the cell culture model m ay be applied, is for screening the absorption, m etabolism and toxicity o f new chem ical entities (NCE’s). The m odel may be useful for the evaluation o f prodrugs, form ulation strategies and drug transport enhancem ent strategies. Such preliminary evaluation w ith cell culture m odels would allow the selection o f the best candidates to be follow ed up for in vivo studies in sm all anim al m odels, prim ates, and hum an clinical trials. The third area o f significant contribution lies in the dem onstration o f a concentrative and equilibrative nucleoside transporters on the lum inal surface o f airw ay epithelial cells. Physiologically this has w ide im plications and suggests a m echanism for the efScient conservation o f nucleosides secreted or entrapped in the airway surface liquid. Therapeutically nucleoside analogs were shown to interact w ith the apical nucleoside transporters enabling these hydrophilic generally m em brane im perm eable drugs to concentrate in the cytoplasm ic com partm ent by active carrier-m ediated mechanisms. These results can lead to the rational design o f nucleoside analogs that possess the structural requirem ents to interact with epithelial nucleoside transporters, thereby avoiding com plicated chem ical m odifications (prodrugs) or conjugations needed to facilitate delivery to the cytoplasm ic compartment. Therefore, the strategic approach in drug design and delivery to the respiratory tract would be to use drugs th at conform with the m inim al essential requirem ents, ensuring a relatively high absorption. A lternately new nucleoside analogs can be designed that take both the viral enzym e and the uptake structural requirem ents into account. The current studies, have also laid the 245 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. foundation to structurally identify nucleoside transporters in the airw ay epithelium using m olecular techniques and oocyte expression system s. The evaluation o f virus infection in the airw ay epithelial m odel, and their effect on epithelial barrier properties and nucleoside transport function presents another area o f fundam ental im portance in antiviral drug delivery to the infected respiratory epithelium . T he finding that Ad5 reduces nucleoside transport opens interesting possibilities as to w hether transporter function o r its synthesis and / or cellular trafficking is disrupted (G urevitz et aL, 1987; Rhoads et aL, 1989). In addition, viral induced changes in m em brane perm eability also play an im portant role in drug transport, as an increase in membrane leakiness m ay reduce the ability to accum ulate drug (via saturable processes) or im prove the perm eability o f an im perm eable analog (C arrasco, 1978; Carpio et aL, 1981; del C astillo e t aL, 1991). W hether these effects are an outcom e o f viral genetic competency (viral replication) o r viral structural proteins w ould further shed light on viral pathogenesis and enable us to devise an appropriate strategy to counter respiratory viruses. C. FUTURE CONSIDERATIONS Based on the results obtained for the developm ent and characterization o f the air- interfaced culture m odel, nucleoside transport in the airw ay and viral infection studies, the follow ing questions rem ain to be considered. 246 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (1) Culture model (a) Is the culture m odel a good predictor o f drugs transported via active mechanisms? Perm eabili^ studies w ith drugs th at are transported prim arily by passive diffusion either through the paracellular or transcellular pathway are generally in good agreem ent w ith that found in other studies. However, no inform ation is available on saturable solute transport processes in the airway fo r the purpose o f com parison. In this study, evidence for carrier-m ediated nucleoside transport was obtained in culture, therefore, it would be o f interest to note the sim ilar or dissim ilar characteristics in excised epithelia. Solute transport processes that can be com pared include the Na^-glucose transporter, the organic anion transporter and the dipeptide transporter am ong others. (b) Is lactate release m ediated by a organic anion or m onocarboxylic acid transporter? Preliminary results suggested that lactate is preferential efQuxed across the basolateral m embrane. A hypothesis to be tested is that a m onocarboxylic acid transporter or organic anion transporter m ediates the selective efflux o f lactate across airway epithelial basolateral m em branes. To address this issue evidence for such an active transport m echanism selective for endogenous organic anions needs to be ascertained. In addition, the interaction o f structurally siim ilar therapeutically active drugs w ith this transporter can be investigated. This can be studied using ^H-lactic acid or ^H-pyruvate as model substrates for the m onocarboxylic acid transporters M C T l and M CT2. A commonly used substrate for the organic anion transporter is ^H-P-amino hippuric acid (PAH). 247 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (2) Nucleoside Transport (a) W hat is the relative density^ o f concentrative and equilibrative transporters? Since both concentrative and equilibrative Qq)es o f transporters are present in airw ay epithelial cells, knowledge o f the relative populations per cell w ould be useful. U nfortunately, no specific radioligand or photoafGnity label is available for the concentrative transporters or the ei type o f equilibrative transporter. To date, the NBTI (specific for the es type o f transporter) is the only ligand, therefore, until new ligands are discovered other m eans o f quantitation have to be pursued, such as antibody inhibition o f nucleoside uptake o r sem iquantitative techniques such as estim ating band intensities from N orthern blots or RT-PCR products. This would involve the structural identification o f the N 2, N 4 and ei nucleoside transporter isoform s using m olecular biology approaches such as functional expression o f isolated mRNA injected in a Xenpus laevis oocyte expression system. (b) Are the structural requirem ents for concentrative and equilibrative transporter distinct? O n the same lines are the structural requirem ents for the basolateral transporter sim ilar to that o f the apical transporters. The results obtained in this study did not differentiate between uptake inhibition w ith nucleoside analogs via the concentrative or equilibrative transporters. However, for the purpose o f dissecting the structural requirements for equilibrative and concentrative transporters this issue can be addressed by perform ing inhibition studies in Na^-free buffer. (c) W hat is the m echanism o f regulation o f concentrative and equilibrative transporters? 248 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Evidence fo r PKC, purinergic, and horm onal regulation o f nucleoside transporters has been dem onstrated in various cell types (Lee et a i, 1991; Delicado et al.. 1994; Gomez- A ngelats e t al.. 1996). W hether the airw ay epithelial cells respond in a sim ilar m anner would shed light on the cellular m echanism responsible for inducing o r inhibiting nucleoside transport fonction in vivo. Besides, if upregulation o f the transporter num bers o r transporter fonction (Km and Vmax) could be investigated as a possible means o f enhancing the delivery o f nucleoside analogs to the airway epithelium . This can be approached by examining the effect o f m odulators o f that particular pathway on nucleoside transport. For example; stim ulation w ith PM A and inhibition with staurosporine, on PKC regulation o f nucleoside uptake can be investigated. Likewise, purinergic stim ulation w ith UTP or ATP (specific for the P2„ receptor), or horm onal regulation w ith insulin, glucagon and other cytokines w ould further our understanding o f nucleoside transport function in healthy and diseased states. (d) Can nucleosides substrates for the nucleoside transporter be delivered by aerosol to sufScient therapeutic concentrations in vivo. The determ ination o f the in vivo aerosolized delivery o f a nucleoside analog that is a substrate fo r the nucleoside transporter w ould validate the strategy in enhancing the delivery o f nucleoside drugs via the nucleoside transporter. This could be tested by studying th e airw ay association o f radiolabeled nucleoside drug, follow ing intratracheal or aerosolized adm inistration. Com petition o f drug uptake w ith excess unlabeled drug o r uridine w ould shed light on the utility o f endogenous solute transporter process in airway epithelial drug transport. 249 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (e) A re nucleoside transporters present in hum an airw ay epithelium ? If so, w hat relative distribution. Since the eventual outcom e o f this research would be to deliver nucleoside analogs to the hum an airw ays, knowledge o f th e existence o f nucleoside transport process in the lum inal epithelium still rem ains to be ascertained. This can be tested by running a sim ilar set a experim ents as in this study w ith isolated hum an tracheal o r bronchial cells. A convenient prelim inary screening can be perform ed w ith norm al hum an bronchial epithelial cell line (NHBE) using th e NBHE pack available through Clonetics, San D iego, CA. (3) V iru s In fectio n S tudies Issues to be addressed based on the prelim inary results on the effect o f virus on epithelial barrier properties include: (a) Is the m em brane leakiness increased? (b) Is Na^ and CT transport processes affected by infection? (c) A re the barrier properties affected by virus protein o f genetically com petent replicating virus? The issues o f m em brane leakiness can be studied using impermeable analogs such as radiolabeled m annitol, sucrose and inulin. A t 48 hr post-infection in culture no significant increase in m em brane leakiness was observed. However, the alteration o f the m em brane structure in late infection w ould provide an estim ate o f the extent o f 250 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. disruption o f cell membrane structure. The alteration o f ion transport function can be estimated by observing the relative responses o f infected cell layers to Na^ channel modulators (am iloride and benzam il), Na^K^ATPase inhibitor (ouabain), the CT -current modulators (furosem ide, N PA A ). The genetic com petency o f AdS to induce the alteration o f epithelial structure and function can be investigated by determining whether a epithelial changes take place in cells infected w ith heat-inactivated, replication deficient AdS, or cell monolayers inoculated w ith virus shells separated from intact virions in a density gradient. (d) Is the decrease in nucleoside transport activity related to the num ber o f functioning transporters? (e) A re the alteration o f epithelial bioelectric properties a function o f inhibition o f cellular protein synthesis? The im pact o f viral infection o n airway nucleoside transport processes can be determined by estim ating the kinetic properties (K m and V m ax) o f uridine uptake in infected and m ock-infected cells. A change in the Vmax w ill confirm whether the number o f functionally active sights (number o f functional transporters) are decreased due to the adenoviral infection. I f this is a direct outcom e o f disruption o f cellular protein synthesis, a decrease in the translation o f new nucleoside transporters can be tested using the protein synthesis inhibitor, cyclohexim ide. 251 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VII. REFERENCES 252 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Adjei, A. and Garren, J. (1990). Pulm onary delivery o f peptide drugs: effect o f particle size on bioavailability o f leuprolide acetate in healthy m ale hum an volunteers. Pharm. Res. 7, 565-569. Adler, K .B., Cheng, P.W ., and Kim , K.C. (1990). Characterization o f guinea pig tracheal epithelial cell maintained in biphasic o rg a n o ^ ic culture: cellular com position and biochem ical analysis o f released glycoconjugates. Am. J. R espir. C ell Mol. Biol. 2, 145-154. Adson, A ., Raub, T.J., Burton, P.S., Barshun, C .L., Hilgers, A .R ., A udus, K.L., and Ho, N.F. (1994). Quantitative approaches to delineate paracellular difhision in cultured epithelial cell m onolayers. J. Pharm . Sci. 83, 1529-1536. Al-Bazzaz, F.J. and Cheng, E. (1979). Effects o f catecholam ines on the ion transport in dog tracheal epithelium . J. Appl. Physiol. 47, 397-403. Altiere, R.J. and Thompson, D C . (1996). Physiology and pharm acology o f the airways. In Inhalation aerosols: physical and biological basis for therapy. A .J. Hickey, ed. (N ew York: M arcel Dekker), pp. 85-137. Artursson, P. (1990). Epithelial transport o f drugs in cell culture. I: A m odel for studying the passive difhision o f drugs over intestinal absorptive (C aco-2) cells. J. Pharm. Sci. 79,476-482. Baatz, J.E ., Bruno, M.D., Ciraolo, P.J., G lasser, S.W ., Stripp, B .R ., Sm yth, K.L., and Korfhagen, T.R. (1994). U tilization o f m odified surfactant-associated protein B for delivery o f DNA to airway epithelial cells in culture. Proc. N atl. A cad. Sci. USA 91, 2547-2551. Balwierczak, J.L ., Krulan, C M ., Wang, Z.C., Chen, J., and Jeng, A Y. (1989). Effects o f adenosine A2 receptor agonists on nucleoside transport. J. Pharm acol. Exp. Ther. 251, 279-287. Basset, G ., Saum on, G., Bouchonnet, F., and Crone, C. (1988). A pical sodium -sugar transport in pulm onary epithelium in situ. Biochim . Biophys. A cta 942, 11-18. Becker, S., Koren, H.S., and Henke, D C. (1993). Interleukin-8 expression in norm al nasa epithelium and its m odulation by infection w ith respiratory syncytial virus and cytokines tum or necrosis factor, interleukin-1, and interleukin-6. Am . J. Respir. Cell Mol. B iol. 8 , 20-27. Belt, J.A ., M arina, N.M ., Phelps, D.A., and Crawford, C.A. (1993). N ucleoside transport in norm al and neoplastic cells. Adv. Enzyme Res. 33, 235-252. 253 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bender, A ., Schfer, V ., Steffan, A.M ., Royer, C ., Kreuter, J., Rubsam en, W .H., and von B riesen, H. (1994). Inhibition o f HTVin vitro by antiviral drug-targeting using nanoparticles. Res. V irol. 745,215-220. Benos, D . (1982). A m iloride: a m olecular probe o f sodium transport in tissues and cells. Am. J. Physiol. 242, C I31-C 145. Berg, M .M ., Kim, K .J., Lubm an, R.L., and C randall, E.D. (1989). H ydrophilic solute transport across rat alveolar epithelium . J. A ppl. Physiol. 66, 2320-2327. B igby, T .D ., Lee, D .M ., M eslier, N ., and G ruenert, D C . (1989). Leukotriene A 4 hydrolase activity o f hum an airw ay epithelial cells. Biochem. Biophys. Res. Comm. 164, 1-7. B oucher, R. and G a t^ , J. (1983). C haracterization o f sodium transport by excised rabbit trachea. J. A ppl. Physiol. 55, 1877-1883. Boum ah, C.E., H arvey, C.M ., Paterson, A .R .P., Baldwin, S.A., and Young, J.D . (1994). Functional expression o f the nitrobenzylthioinosine-sensitive nucleoside transporter o f hum an choriocarcinom a (BeW o) cells in isolated oocytes o f Xenopus laevis. Biochem . J. 299, 769-773. B razzell, R.K ., Sm ith, R .B ., and K ostenbauder, H.B. (1982). Isolated perfused rabbit lung as a model for intravascular and intratracheal adm inistration o f bronchodilator drugs I: isoproterenol. J. Pharm . Sci. 7 1 ,1268-1274. B rett, C .A ., W ashington, C .B ., O tt, R.J., G utierrez, M.M., and G iacom ini, K.M . (1993). Interaction o f nucleoside analogs w ith sodium -nucleoside transport system in brush border m em brane vesicles from the hum an kidney. Pharm. Res. 1 0 ,423-426. Bronk, J R. and H astew ell, J.G . (1988). The transport and m etabolism o f naturally occurring pyrim idine nucleosides by isolated ra t jejenum . J. Physiol. 395, 349-361. Cabantchick, Z.I. and G insburg, H. (1977). T ransport o f uridine in blood cells. D em onstration o f a sim ple carrier-m ediated m echanism . J. Gen. Physiol. 69, 75-96. C arpio, M .M ., Babiuk, L.A ., M isra, V., and Blum enthal, R.M. (1981). Bovine rotavirus cell interactions: effect o f virus infection on cellular integrity and m acrom olecular synthesis. Virol. 114, 86-97. C arrasco, L. (1978). M em brane leakiness after viral infection and a new approach to the developm ent o f antiviral agents. N ature 272, 694-699. 254 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C ereijido, M., R obbin, E.S., Dolan, C ^ ., Rotunno, C ^ ., and Sabatini, D.D. (1978). Polarized m onolayers form ed by epithelial cells on a perm eable and translucent support. J. CeU Biol. 77, 853-880. C han, T.C.K., Shaffer, L., Redmond, R ., and Pennington, K . (1993). Permeation and m etabolism o f anti-H IV and endogenous nucleosides in hum an immune effector cells. Biochem . Pharm acol. 46, 273-278. C he, M ., N ishida, T ., Gatmaitan, Z , and A rias, I.M . (1992). A nucleoside transporter is functionally linked to ectonucleotidases in rat liver cananicular membrane. J. Biol. Chem . 267,9684-9688. Chediak, A.D. and W anner, A. (1990). The circulation o f th e airw ays: anatomy, physiology and potential role in drug delivery to the respiratory tract. Adv. Drug D eliv. Rev. 5, 11-18. C hu, C.K., Schinazi, R.F., Ahn, M .K ., U llas, G.V., and G u, Z. (1989). Structure- activityrelationships o f pyrim idine nucleosides as antiviral agents for human im m unodeficiency virus type I in peripheral blood m ononuclear cells. J. Med. Chem. 32, 612-617. C lark, A.B., R andell, S.H ., Nettesheim , P., Gray, T.E., B % nell, B., and Ostrowski, L.E. (1995). Regulation o f ciliated cell differentiation in cultures o f rat tracheal epithelial cells. Am. J. R espir. C ell Mol. Biol. 1 2 ,329-338. C loutier, M.M., W ong, D.W ., and O gra, P.L. (1989). R espiratory syncytial virus alters electrophysiologic properties in cotton rat airway epithelium . Pediatr. Pulmonol. 6, 164- 168. C loutier, M.M. and Lesniak, K.M. (1985). p-A m inohippurate transport in canine tracheal epithelium . J. Appl. Physiol. 59, 1585-1589. Cohen, A., U llm an, B., and M artin, D . W. (1979). C haracterization o f a m utant m ouse lym phom a cell w ith deficient transport o f purine and pyrim idine nucleosides. J. B iol. Chem . 254, 112-116. Colem an, D.L., T uet, I.K ., and W iddicom be, J.H . (1984). E lectrical properties o f dog tracheal epithelial cells grown in m onolayer culture. Am. J. Physiol. 246, C355-C359. Connelley, M .C., R obbins, B.L., and Fridland, A. (1993). M echanism o f uptake o f the phosphonate analog (S)-l-(3-hydroxy-2-phosphonylm ethoxy-propyl)cytosine (HPM PC) in V ero cells. Biochem . Pharmacol. 4 6 , 1053-1057. 255 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Crack, B E ., Pollard, C.E., Beukers, M .W ., Roberts, S.M ., Hunt, S.F., Ingall, A .H ., M cKenchinie, K.C.W ., Ijzrerman, A .?., and Lefif, P. (1995). Pharm acological and biochem ical analysis o f FPL 67156, a novel, selective inhibitor o f ecto-A TPase. Br. J. Pharm acol 114,475-481. Crawford, C JL , Ng, C.Y.G., Noel, D .L., and Belt, J.A . (1990). N ucleoside transport in L1210 m urine leukem ia cells; Evidence fo r three transporters. J. Biol. Chem . 265, 9732- 9736. Crawford, C.R. and B elt, J.A. (1991). Sodium -dependent, concentrative nucleoside transport in W alker 256 rat carcinosarcom a cells. Biochem. Biophys. R es. Comm. 175, 846-851. Crom well, O., Ham id, Q ., Corrigan, C .G ., Barkans, J., M eng, Q., C ollins, P.D ., and Kay, A.B. (1992). Expression and generation o f interleukin-8, IL-6 and granulocyte- m acrophage colony stim ulating factor by bronchial epithelial cells and enhancem ent by IL-1 beta and tum or necrosis factor alpha. Immunol. 77, 330-337. C uriel, D .T., Agarwal, S., Romer, M .U ., Wagner, E., Cotten, M., B irsnstiel, M .L., and Boucher, R.C. (1992). Gene transfer to respiratory epithelial cells via th e receptor- m ediated endocytosis pathway. Am. J. Respir. Cell M ol. Biol. 6, 247-252. Das, D.K. and Steinberg, H. (1988). Adenosine transport by the lung. J. A ppl. Physiol. 65, 297-305. De Clercq, E. (1995). Antiviral therapy for human im m unodehciency virus infections. Clin. M icrobiol. Rev. 8, 200-239. de Jong, P., van Sterkenberg, M .A.J.A., Kempenaar, J.A ., Dijkman, J.H ., and Ponec, M. (1993). Serial culturing o f human bronchial epithelial cells derived from biopsies. In V itro Cell. Dev. Biol. 29A, 379-387. de Jong, P., van Sterkenberg, M .A.J.A., Kempenaar, J.A ., Hesseling, S.C ., M ulder, A .A ., M ommaas, A.M ., Dijkman, J.H ., and Ponec, M. (1994). C iliogenesis in the hum an bronchial epithelial cells cultured at the air-liquid interface. Am . J. Respir. Cell M ol. Biol. 1 0 ,271-277. Debs, R.J., M ontgom ery, B.A., Brunette, E.N., DeBruin, M ., and Shanely, J.D . (1988). A erosol administration o f antiviral agents to treat lung infection due to cytom egalovirus. J. Infect. Dis. 157,327-331. D efer, C ., Belin, M .T., Caillet-Boudin, M L., and Boulanger, P. (1990). Hum an adenovirus-host cell interactions: com parative study w ith mem bers o f subgroups B and C. J. V irol. 64, 3661-3673. 256 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. del C astillo, J R ., Ludert, J.E ., Sanchez, A ., Ruiz, M .C., M ichelangeli, F., and Liprandi, F. (1991). R otavirus in& ction alters Na^ and hom eostasis in M A -104 cells. J. Gen. V irol. 72, 541-547. Delicado, E.G ., C asillas, T., Sen, R.P., and M iras-Portugal, M .T. (1994). Evidence that adenine nucleotides m odulate nucleoside transporter function. Characterization o f uridine transport in chrom affin cells and plasm a m em brane vesicles. Eur. J. Biochem. 22 5 ,355-362. Diam ond, J.M . (1979). Osm otic water flow in leaky epithelia. J. M embr. Biol. 51, 195- 216. Doherty, A .J. and Jarvis, S.M . (1993). Na^-dependent and -independent uridine uptake in an established renal epithelial cell line, OK, from the opossum kidney. Biochim . Biophys. A cta 1147, 214-222. Dom in, B.A ., M ahony, W .B., and Zimmerman, T.P. (1993). M embrane perm eation m echanism o f 2 ,3 -dideo:tynucleosides. Biochem. Pharm acol. 46, 725-729. Drusano, G .L., Yuen, G .J., M orse, G., Cooley, T.P., Seidlin, M ., Lam bert, J.S., Liebm an, JA ., V alentine, F T ., and Dolin, R. (1992). Im pact o f bioavailability on determ ination o f m axim al tolerated dose o f 2 ,3 -dideoxyinosine in phase I clinical trials. A ntim icrob. A gents Chem other. 36, 1280-1283. Durbin, R.P., Frank, H ., and Solomon, A.K. (1956). W ater flow through frog gastric m ucosa. J. Gen. Physiol. 39, 531-535. Enna, S.J. and Schanker, L.S. (1972). A bsorption o f drugs from the rat lung. Am. J. Physiol. 223, 1227-1231. Enna, S.J. and Schanker, L.S. (1973). Phenol red absorption from the rat lung: evidence o f carrier transport. Life Sci. 12, 231-239. Fang, X., Parkinson, F.E., Mowles, D.J., Young, J.D ., and Cass, C.E. (1996). Functional expression o f a recom binant sodium-dependent nucleoside transporter w ith selectivity for pyrim idine nucleosides (cN T l) by transient expression in the culture. Biochem. J. 457-465. Ferkol, T., Perales, J.C ., Eckm an, E., Kaetzel, C.S., Hanson, R.W ., and Davis, P.B. (1995). Gene transfer into airway epithelium o f anim als by targeting the polym eric im m unoglobulin receptor. J. Clin. Invest. 9 5 ,493-502. 257 ) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fiedler, M .A., Kaetzel, C.S., and D avis, P.B. (1991). Sustained production o f secretory com ponent by hum an tracheal epithelial cells in prim ary culture. Am . J. Physiol. 261, L255-L261. Fium e, L ., Busi, C., D i-Stefano, G ., and M attioli, A. (1993). C oupling o f antiviral nucleoside analogs to lactosam inated hum an album in by using im idazolides o f their phophoric esters. A nal. Biochem . 2 1 2 ,407-411. Frank, K.B., Chiou, J.F ., and C heng, Y.C. (1985). Interaction o f DNA polym erase and nucleotide analog triphosphates. A dv. Enzym e Regul. 24, 377-384. Fulm er, J.D . and C rystal, R.G. (1976). B iochem istry basis o f pulm onary function. In R .G . C rystal, ed. ( G arcia, C.K ., Brown, M S., Pathak, R .K ., and G oldstein, J.L. (1995). cDNA cloning o f M CT2, a second m onocarboxylate transporter expressed in different cells than M C T l. J. B iol. Chem . 270, 1843-1849. G ati, W .P., M isra, H .K ., Knaus, E .E ., and W ieb, L.I. (1984). Structural m odifications at the 2' and 3'-positions o f some pyrim idine nucleoside as determ inants o f their interaction w ith the m ouse erythrocyte nucleoside transporter. Biochem . Pharm acol. 33, 3325-3331. G entry, S.E., Culp, D .J., Roberts, N .J., M arin, M .G ., Simons, R.L., and Latchney, L.R. (1988). Influenza virus infection o f tracheal gland cells in culture. J. V irol. 62, 1524- 1529. G iacom ini, K.M ., M arkovich, D ., W erner, A ., Biber, J., Wu, X., and M urer, H. (1994). E xpression o f renal N a^-nucleoside cotransport system (N2) in X enopus laevis oocytes. Pflugers Arch. 427, 381-383. G ilbert, B E., W yde, P R ., Am brose, M .W ., W ilson, S.Z., and Knight, V. (1992). Further studies w ith short duration ribavirin aerosol for the treatm ent o f influenza virus infection in m ice and respiratory syncytial virus infection in cotton rats. A ntiviral Res. / 7, 33-42. G insberg, H.S., Horswood, R.L., Chanock, R.M ., and Prince, G.A. (1990). Role o f early genes in pathogenesis o f adenovirus pneum onia. Proc. Natl. Acad. Sci. USA 87, 6191- 6195. G om ez-A ngelats, M ., del Santo, B ., and M ercader, J. (1996). Horm onal regulation o f concentrative nucleoside transport in liver parenchym al cells. Biochem . J. 313,915-920. 2S8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. G onzalez-M ariscal, L. (1992). The relationship betw een structure and function o f tight junctions. In T ight Junctions. M. Cereijido, ed. (Florida: CRC Press), pp. 67-75. Goodchild, J., Porter, R.A., Raper, R.H., Sim , I.S ., Upton, R.M ., Viney, J., and W adsworth, H .J. (1983). Structural requirem ents o f olefinic 5-substituted deoxyuridines for antiherpes activity. J. Med. Chem. 26,1252-1257. Goodman, B .E ., Anderson, J.L., Clemens, J.W ., Kircher, K .J., Storm o, M.L., W altz, J.S., W altz, W .F., and W hite, J.W . (1994). D ifferences in sodium and D-glucose transport betw een ham ster and rat lungs. J. A ppl. Physiol. 76, 2578-2585. G riffith, D.A ., Doherty, A .J., and Jarvis, S.M . (1992). N ucleoside transport in cultured LLC -PK iepithelia. Biochim. Biophys. A cta 1106, 303-310. Gruenert, D C ., Finkbeiner, W .E., and W iddicom be, J.H. (1995). Culture and transform ation o f hum an airway epithelial cells. Am. J. Physiol. 286, L347-L360. Guo, Y., Schneider, L.A., and W angensteen, O .D . (1995). HO Cl effects on tracheal epithelium : conductance and perm eability m easurem ents. J. A ppl. Physiol. 78, 1330- 1338. Gurevitz, M ., Schulze, I T., Swierkosz, E.M ., A ren, M.Q., and Schwarz, K.B. (1987). Effect o f influenza B virus on nutrient transport in cultured epithelial cells. Lab. Invest. 6, 657-664. Gutierrez, M .M ., Brett, C.A., O tt, R J., Hui, A C ., and Giacom ini, K.M . (1992). N ucleoside transport in brush border m em brane vesicles from hum an kidney. Biochim . Biophys. A cta 1105, 1-9. Gutierrez, M .M . and Giacomini, K.M. (1993). Substrate selectivity, potential sensitivity and stoichiom etry o f Na^-nucleoside transport in brush border m em brane vesicles o f hum an kidney. Biochim . Biophys. A cta 1149, 202-208. Gutierrez, M .M . and Giacomini, K.M. (1994). Expression o f a hum an renal sodium nucleoside cotransporter m. Xenopus laevis oocytes. Biochem. Pharm acol. 48,2251- 2253. H all, C.B., M cBride, J.T., W alsh, E.E., B ell, D M ., Gala, C .L., H ildreth, S., TenEycl, L.G., and H all, W .J. (1983). Aerosolized ribavirin treatm ent o f infants w ith respiratory syncytial viral infection, a double blind study. N . Engl. J. M ed. 308, 1443-1447. Hammer, K ., H atlelid, J., Grotli, M., Arukwe, J., Klaveness, J., R ise, F., and Undheim, K. (1996). E ther, cabonate and urthane deoxynucleoside derivatives as prodrugs. A cta Physiol. Scand. 50, 609-622. 259 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hansch, C ., Sammes, P.G., and Taylor, J.B . (1990). Com prehensive m edicinal chem istry: The rational design, m echanistic study and therapeutic application o f chem ical compounds (New York: Pergam on Press). Harkem a, J.R ., M ariassy, A T ., St.George, J., Hyde, D.M ., and Piopper, C.G. (1991). Epithelial cells o f the conducting airway. In The Airway Epithelium . S.G. Farm er and D.W . H ay, eds. (New York: M arcel Dekker), pp. 3-40. Hasegawa, T., Seki, T., Juni, K ., Saneyoshi, M ., and Kawaguchi, T. (1993). Prodrugs o f 2',3'-didehydro-3'-deoxythym idine. J. Pharm . Sci. 82, 1232-1236. He, Y ., Sanderson, I.R., and W alker, W .A. (1994). Uptake, transport and m etabolism o f exogenous nucleoisdes in intestinal epithelial cell cultures. J. N utrition 124, 1942-1994. H ierholzer, J.C. (1992). Adenoviruses in im m unocomprom ised host. Clin. M icrobiol. Rev. 5,262-274. Hosoya, K., Kubo, H., Natsum e, H., Sugibayashi, K., M orim oto, Y., and Yam ashita, S. (1993). The structural barrier o f absorptive muosae: site difference o f the perm eability o f fluorescein isothiocyanate-labeled dextran in rabbits. Biopharm . Drug Dispos. 14, 685-696. Hu, M . (1993). Comparison o f uptake characteristics o f thym idine and zidovudine in a hum an intestinal epithelial m odel system. J. Pharm. Sci. 82, 829-833. Huang, Q ., Harvey, C M ., Paterson, A.R.P., Cass, C.E., and Young, J.D. (1993). Functional expression o f Na^-dependent nuclesoide transport system s o f rat intestinal oocytes o f Xenopus laevis. J. B iol. Chem. 2 6 8 ,20613-20619. Huang, Q ., Yao, S.Y.M., R itzel, M .W .L., Paterson, A .R.P., C ass, C.E., and Young, J.D. (1994). Cloning and functional expression o f a com plem entary DNA encoding a m am m alian nucleoside transport protein. J. Biol. Chem. 269, 17757-17760. Ilow ite, J.S. and Niederman, M .J. (1990). Problems and opportunities in the topical treatm ent o f infectious diseases o f the respiratory tract. A dv. Drug Deliv. Rev. 5, 93- 105. Inoue, S. and Hogg, J.C. (1977). Freeze-etch study o f the tracheal epithelium o f norm al guinea pigs w ith particular reference to the intercellular junctions. J. Ultrastruct. Res. 61, 89-99. Iseki, K ., Sagawara, M ., Fujiw ara, T., N assani, I., Kobayashi, M ., and M iyazaki, K. (1996). Transport m echanism o f nucleoside and the derivative, 6-m ercaptopurine 260 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. riboside across rat intestinal brush-border mem brane vesicles. Biochim . Biophys. A cta 1278, 105-110. Jakobs, E.S., V an Os-Corby, D.J., and Paterson, A .R.P. (1990). Expression o f sodium - linked nucleoside transport activity in m onolayer cultures o f IEC-6 intestinal epithelial cells. J. B iol. Chem. 2 6 5 ,22210-22216. Jakobs, E.S. and Paterson, A.R.P. (1986). Sodium -dependent, concentrative nucleoside transport in cultured intestinal epithelial cells. Biochem . B iophys. Res. Comm. 140, 1028-1035. Jam igan, F., Davis, J.D ., Bromberg, P.A., G atsy, J.T ., and B oucher, R.C. (1983). B ioelectric properties and ion transport o f excised rabbit trachea. J. Appl. Physiol. 55, 1884-1892. Jarvis, S.M ., Hammond, J R ., Paterson, A.R.P., and C lanachan, A .S. (1982). Species differences in nucleoside transport. A study o f uridine transport and nitrobenzylthioinosine binding by mammalian erythrocytes. Biochem . J. 208, 83-88. Jarvis, S.M . (1989). C haracterization o f sodium -dependent nucleoside transport in rabbit intestinal brush-border membrane vesicles. B iochim . Biophys. Acta 979, 132- 138. Jarvis, S.M . and G riffith, D A . (1991). Expression o f th e rabbit intestinal N2 Na^- nucleoside transporter in Xenopus laevis oocytes. Biochem . J. 278, 605-607. Jetten, A .M ., Brody, A .R ., Deas, MA.., Hook, G .E.R., R earick, J.I., and Tacher, S.M . (1987). R etinoic acid and substratum regulate the differentiation o f rabbit tracheal epithelial cells into squam ous and secretory phenotype. Lab. Invest. 56, 654-664. Jiang, C ., Finkbeiner, W .E., W iddicombe, J.H., M cCray Jr., P.B ., and M iller, S.S. (1993). A ltered fluid transport across airway epithelium in cystic fibrosis. Science 262, 424-427. Johnson, L.G ., Cheng, P.W ., and Boucher, R.C. (1989). A lbum in absorption by canine bronchial epithelium . J. A ppl. Physiol. 6 6 ,2772-2777. Johnson, L.G ., Dickman, K.G ., Moore, K.L., M andel, L.J., and Boucher, R.C. (1993). Enhanced Na^ transport in an air-liquid interface culture system . Am. J. Physiol. 264, L560-L565. Jones, K .B ., Holland, J.F ., Foureman, G.L., Bend, J R ., and Fouts, J R. (1983). Induction o f xenobiotic m etabolism in C lara cells and alveolar type-II cells isolated from rat lungs. J. Pharm acol. Exp. Ther. 2 25,316-319. 261 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Jo n s, L . and Q uinton, P.M . (1989). Evidence for electrogenic N a-giucose cotransport in tracheal epithelium . Pflugers Arch. Eu. J. Phsiol. 415, 118-120. Jzerm an, AJ*., K ruidering, M ., van W eert, A ., van B elle, H., and Janssen, C. (1992). [^H]-R7S231 - a new radioligand for the nitroben^lthioinosine sensitive nucleoside tm asport proteins. Characterization o f (+/-)-[^H]-R75231 binding to c a lf lung m em branes, stereospecificity o f its tw o isom ers, and com parison w ith [^H]- nitrobenzylthioinosine binding. N aunyn-Schm iedebergs Arch. Pharm acol. 3 4 5 ,558- 563. K aartinen, L ., N ettesheim , P., Adler, K .B., and Randell, S.H. (1993). R at tracheal epithelial cell differentiation in vitro. In Vitro Cell. Dev. Biol. 29A, 481-492. R am ps, J.A ., Sw art, P.J., M orselt, H.W ., Pauwels, R., D e Bethune, M .P ., D e C lercq, E., M eyer, D .K ., and Scherphof, G.L. (1996). Preparation o f characterization o f conjugates o f (m odified) hum an serum album in and liposomes: drug carriers w ith an intrinsic anti- HIV activity. B iochim . Biophys. A cta 1 278,183-190. K ilbum , K.H . (1974). Functional m orphology o f the distal lung. Int. R ev. C ytol. 37, 153-270. Kim , K .-J., M atsukaw a, Y ., Lee, V .H .L., and Crandall, E D. (1995). C haracteristics o f album in absorption across rat alveolar epithelial cell m onolayers. FA SEB J. 9, A569 Kim , K .C . (1985). Possible requirem ent o f collagen-gel substratum fo r production o f m ucin-like glycoproteins by prim ary rabbit tracheal epithelial cells in culture. In V itro C ell. D ev. B iol. 21, 617-621. BCim, K .J., Cheek, J.M ., and Crandall, E D . (1991). C ontribution o f active N a^ and Cl- fiuxes to net ion transport by alveolar epithelium . Respir. Physiol. 85, 245-256. Kim , K .J. and Crandall, E D. (1983). Heteropore populations o f bullfirog alveolar epithelium . J. A ppl. Physiol. 54, 140-146. Kim , K .J. and C randall, E D. (1995). Specialized alveolar epithelial transport processes. In F luid and solute transport in the airspaces o f the lungs. R M . E ffros and H.K . Chang, eds. (N ew York: M arcel Dekker,Inc ), pp. 219-247. K innula, V .L., A dler, K .B ., Ackley, N .J., and Crapo, J.D . (1992). R elease o f reactive oxygen species by guinea pig tracheal epithelial cells in vitro. Am. J. Physiol. 262, L708-L712. 262 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. K night, V. and G ilbert, B E. (1988). A ntiviral therapy w ith small particle aerosol. Eur. J. C lin. M icrobiol. In fe c t Dis. 7, 721-731. K now les, M .R., C larke, L.L., and Boucher, R.C. (1991). Activation by extracellular nucloetides o f chloride secretion in the airway epithelia o f patients w ith cystic fibrosis. N. Engl. J. M ed. 325, 533-538. K now les, M .R., H ohneker, K.W., 22iou, Z., Oisen, J.C ., Noah, T.L., Hu, P.C ., Leigh, M .W ., Engelhardt, J.F ., Edwards, L.J., and Jones, K.R. (1995). A controlled study o f adenoviral-vector-m ediated gene transfer in the nasal epithelium o f patients with cystic fibrosis. N. Engl. J. M ed. 333, 823-831. K nupp, C.A., Shyu, W .C., Dolin, R ., V alentine, F.T., M cLaren, C., M artin, R., Pittm an, K .A ., and Barbhaiya, R.H. (1991). C lin. Pharm acol. Ther. 4 9 ,523-535. K obayashi, S., K ondo, S., and Juni, K. (1995). Perm eability o f peptides and proteins in hum an cultured alveolar A549 cell m onolayer. Pharm aceutical Research 12, 1115-1119. K om pella, U.B., K im , K .J., and Lee, V.H .L. (1993). A ctive chloride transport in the pigm ented rabbit conjunctiva. Curr. Eye Res. 12, 1041-1048. K ondo, M., Finkbeiner, W .E., and W iddicom be, J.H . (1991). A sim ple technique for culture o f highly differentiated cells firom dog tracheal epithelium. Am. J. Physiol. 261, L106-L117. K ondo, M., Finkbeiner, W .E., and W iddicom be, J.H . (1993). Cultures o f bovine tracheal epithelium w ith differentiated ultrastructure and ion transport. In Vitro Cell. Dev. Biol. 29A, 19-24. K orst, R.J., Bew ig, B ., and Crystal, R.G. (1995). In vitro and in vivo transfer and expression o f hum an surfactant SP-A- and SP-B-associated protein cDNAs m ediated by replication-deficient, recom binant adenoviral vectors. Hum. Gene Ther. 6, 277-287. K ubo, H., Hosoya, K ., Natsume, H., Sugibayashi, K., and Morimoto, Y. (1994). In vitro perm eation o f various m odel drugs across rabbit nasal mucosa. Int. J. Pharm . 103, 27- 36. Kunzelm ann, K ., Y ezzi, M .J., Schwiebert, E.M ., Kuo, W .L., Chin, L., Stanton, B.A., and Gruenert, D C . (1993). An im m ortalized cystic fibrosis tracheal epithelial cell line hom ozygous for th e delta 508 CFTR m utation. Am. J. Respir. Cell M ol. Biol. 8, 522- 529. Landis, E.M. and Pappenheim er, J R. (1963). Exchange o f substances through the capillary wall. In H and Bood o f Physiology; Circulation,Section 2, Vol. 2. W.F. 263 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. H am ilton and P. Dow, eds. (W ashington D.C. Am erican Physiology Society), pp. 961- 1034. Lane, B.P., M iller, S.L., and Drum m ond, E.J. (1976). U se o f tracheal organ culture in toxicity testing. Environ. Health P ersp ect 1 6 ,89-98. Lanman, R.C., G illilan, R.M ., and Schanker, L.S. (1973). Absorption o f cardiac glycosides firom the respiratory tract. J. Pharmacol. Exp. Ther. 187, 105-111. Lechner, J.F., H augen, A ., Autrup, H ., McClendon, LA., Trump, B J . , and Harris, C.G. (1981). Clonal grow th o f epithelial cells firom normal adult human bronchus. Cancer Res. 41, 2294-2304. Lee, C.W ., Sokolosld, J.A ., Sartorelli, A C., and Handschumacher, R.E. (1991). Induction o f the differentiation o f H L-60 cells by phorbol 12-myristate 13-acetate activates a Na^-dependent uridine transport system. Involvm ent o f protein kinase C. Biochem. J. 274, 85-90. Lee, T.C., Wu, R., B rody, A.R., B arrett, J.C ., and N ettesheim , P. (1984). Growth and differentiation o f ham ster tracheal epithelial cells in culture. Exp. Lung Res. 6 , 27-45. Lipson, S.M. (1989). R apid culture confirm ation o f adenovirus isolates using the Adenoclone enzym e imm unoassay (EIA ) test kit. Diagn. M icrobiol. In fect Dis. 12, 361-362. Litterst, C.L., M im naugh, E.G., Reagan, R.L., and Gram, T.E. (1975). Com parison o f in vitro drug m etabolism by lung, liver and kidney o f several common laboratory species. Drug Metab. D isposition 3,2 4 9 Lu, L., Zeitlin, P.L., Guggino, W .B., and Craig, R.W. (1989). Gene transfer by lipofection in rabbit and human secretory epithelial cells. Pflugers Arch. Eu. J. Phsiol. 415, 198-203. M a, J.K .H., Bhat, M ., and Rojanasakul, Y. (1996). D rug m etabolism and enzyme kinetics in the lung. In Inhalation aerosols:physical and biological basis for therapy. A.J. H ickey, ed. (New York: M arcel D ekker), pp. 155-195. M a, T.Y., Hollander, D ., Riga, R ., and Bhalla, D. (1993). Autoradiographic determ ination o f perm eation pathw ay o f perm eability probes across intestinal and tracheal epithelia. J. Lab. Clin. M ed. 122, 590-600. M agnani, M ., C asablanca, A., Fratem ale, A., Brandi, G ., Gessani, S., W illiam s, R., G iovine, M., Dam onte, G ., De Flora, A ., and Benatti, U. (1996). Synthesis and targeted 264 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. delivery o f an azidothym idine hom odinucleotide conferring protection to m acrophages against retroviral infection. Proc. N atl. A cad. Sci. USA 9 3 ,4403-4408. M ahony, W .B., D om in, B.A., and Zim m erm an, T.P. (1991). G anciclovir perm eation o f the hum an erythrocyte membrane. Biochem . Pharmacol. 41, 263-271. M an, S J .P ., H ulbert, W .C., Mok, K., Ryan, T., and Thom son, A.B.R. (1986). Effects o f sulfur dioxide on pore populations o f canine tracheal epithelium . J. Appl. Physiol. 60, 416-426. M arriott, C. (1990). M ucus and m ucociliary clearance in the respiratory tract. Adv. Drug D eliv. Rev. 5, 19-35. M artin, D. and M eckling, G.K.A. (1996). Omega-3 polyunsaturated fatty acids increase purine but not pyrim id in e transport in L1210 leukem ia cells. Biochem. J. 315, 329-333. M ason, R .J. and W illiam s, M.C. (1980). Phospholipid com position and ultrastructure o f A549 cells and other cultured pulm onary epithelial cells o f presum ed ^rpe II cell origin. Biochim . Biophys. A cta 617, 36-50. M athias, N .R ., Robison, T.W ., Kim, K .J., and Lee, V.H.L. (1995). Developm ent and characterization o f rabbit tracheal epithelial cell culture m odels for drug transport studies. Pharm . Res. 12, 1499-1505. M athias, N .R ., Kim , K .J., and Lee, V .H X . (1996). Targeted delivery to the respiratory tract; Solute perm eability o f air-interfaced cultured rabbit tracheal epithelial cell m onolayers. J. Drug Targeting 4, 79-86. M atsukawa, Y., Yam ahara, H., Lee, V .H .L., Crandall, E.D ., and Kim, K.J. (1996). H orseradish peroxidase transport across rat alveolar epithelial cell m onolayers. Pharm . Res. 13, 1311-1335. M atsukawa, Y., Lee, V .H .L., Crandall, E D ., and Kim, K .J. (1997). Size-dependent dextran transport across rat alveolar epithelial cell m onolayers. J. Pharm. Sci. 86, 305- 309. M axwell, D M. (1983). The role o f antibiotics given by inhalation in chronic chest disease. J. A ntim icrob. Chemother. 1 1 ,203-206. M eckling, G.K .A., G uilbert, L., and C ass, C.E. (1993). CSF-1 stim ulates nucleoside transport in S I m acrophages. J. Cell. Physiol. 155, 530-538. 265 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. M ercarder, J., G om ez-A ngelats, M ., Santo, B., Casado, F.J., Felipe, A ., and Pastor- A ngalata, M . (1996). N ucleoside uptake in rat liver parenchym al cells. Biochem. J. 3 1 7, 835-842. M ilton, S.G. and K nutson, V .P. (1990). Comparison o f the function o f the tight junctions o f endothelial cells and epithelial cells in regulating the m ovem ent o f electrolytes and m acrom olecules across the cell monolayer. J. C ell. Physiol. 144,498- 504. M ontgom ery, B .A ., Luce, J.M ., Turner, J., Lin, E.T., Debs, R.J., Corkery, K.J., B runette, E.N., and H opew ell, P C. (1987). Aerosolized pentam idine as sole therapy for pneum ocistis carinii pnem onia in patients w ith acquired immune deficiency syndrom e. Lancet 2 , 480-483. M orim oto, K ., Y am ahara, H ., Lee, V .H .L., and Kim, K.J. (1993). D ipeptide transport across rat alveolar epithelial cell m onolayers. Pharm. Res. 10, 1668-1674. M orim oto, K ., Y am ahara, H ., Lee, V .H .L., and Kim, K.J. (1994). T ransport o f thyrotropin releasing horm one across rat alveolar epithelial cell m onolayers. Life Sci. 54, 2083-2092. M orita, T., Yam am oto, Y ., H ashida, M ., and Sezaki, H. (1993). Effects o f various absorption prom oters o n pulm onary absorption o f drugs w ith different m olecular w eights. B iol. Pharm . B ull. 16, 259-262. M orris, D.J. (1994). A dverse effects and drug interactions o f clinical im portance w ith antiviral drugs. Drug Safety 10, 281-291. M ortell, K.H., M arm orstein, A .D ., and Cramer, E.B. (1993). Fetal bovine serum and other sera used in tissue culture increase epithelial perm eability. In V itro Cell. Dev. B iol. 29A, 235-238. N araw ane, M ., Podder, S.K ., Bundgaard, H., and Lee, V.H.L. (1992). Segm ental differences in drug perm eability, esterase activity and ketone reductase a c tiv i^ in th e albino rabbit intestine. J. D rug Targeting 1, 7-17. N ettesheim , P., Jetten, A .M ., Inayam a, Y., Brody, A .R., George, M .A ., Gilmore, L .B ., G ray, T., and Hook, G .E .R . (1990). Pathways o f differentiation o f airw ay epithelial cells. Environ. H ealth Perspect. 85, 317-329. N ettesheim , P. and B arrett, J.C . (1984). Tracheal epithelial cell transfiarmation: a m odel system for the studies o n neoplastic progression. CRC Crit. Rev. Toxicol. 12,215-239. 266 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. O liver, J.M . and Paterson, A.R.P. (1971). N ucleoside transport. I. A m ediated transport in hum an erythocytes. Can. J. Biochem. 4 9 ,262-270. Palu, G ., Handschum acher, R.E., Parolin, C., Stefanelli, S., and Palatini, P. (1990). E ffect o f herpes sim plex virus type 1 infection on nucleoside transport in H eLa S3 cells. J. G en. Virol. 71, 673-679. Paradiso, A.M ., C heng, E.H., and Boucher, R.C. (1991). Effects o f bradykinin on intracellular calcium regulation in hum an cilated airw ay epithelium . Am. J. Physiol. 261, L63-L69. Paterson, A .R .P., Yang, S., Lau, E.Y., and Cass, C.E. (1979). Low specificity o f the nucleoside transport m echanism o f R P M I6410 cells. M ol. Pharmacol. 16, 900-908. Paterson, A .R .P., K olossa, N., and Cass, C.E. (1981). Transport o f nucleoside drugs in anim al cells. Pharm ac. Ther. 12, 515-536. Plagem ann, P.G .W ., W ohlheuter, R.M ., and Wofifendin, C. (1988). N ucleoside and nucleobase transport in animal cells. Biochim. Biophys. A cta 947,405-443. Plagem ann, P.G .W . (1991). Na^-dependent, concentrative nucleoside transport in rat m acrophages. Biochem . Pharmacol. 42, 247-250. Piopper, C.G., M ariassy, A T., W ilson, D.W ., A lley, J.L ., N ishio, S.J., and Nettesheim , P. (1983). C om parison o f noncUiated tracheal epithelial cells in six m am m alian species: ultrastructure and population densities. Exp. Lung Res. 5, 281-294. Prus, K ., Averett, D R., and Zimmerman, T.P. (1990). Transport and m etabolism o f 9-b- D -arabinofuranosylguanine in a hum an T-lym phoblastoid cell line. N itrobenzylthiinosine-sensitive and -insensitive influx. C ancer Res. 50, 1817-1821. Raeburn, D., Underwood, S.L., and Villam il, M E. (1992). Techniques o f drug delivery to the airways, and the assesm ent o f lung function in anim al m odels. J. Pharmacol. Toxicol. M ethods 27, 143-159. R eiss, T.F., G ruenert, D C ., Nadel, J.A ., and Jacoby, D.B. (1991). Infection o f cultured hum an airway epithelial cells by influenza A virus. Life Sci. 49, 1173-1181. R eiss, T.R., G ruenert, D C ., Nadel, J.A ., and Jacoby, D .J. (1991). Infection o f cultured hum an airway epithelial cells by influenza A virus. Life Sci. 4 9 ,1173-1181. Renkin, E.M. (1954). Filtration, difiusion and m olecular sieving through porous cellulose m em branes. J. Gen. Physiol. 38, 225-243. 267 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rennard, S I., Beckm ann, J.D ., and Robbins, R-A. (1991). Biology o f the airw ay. In The Lung: Scientific Foundations. R.G. Crystal and J.B. W est, eds. (N ew York: Raven Press), pp. 157-168. Rensen, P.C., van D ijk, M .C., Havenaar, E.G., Bijsterbosch, M .K ., K ruijt, K J ., and van B erkel, T.J. (1995). Selective liver targeting o f antivirals by recom binant chylom icrons- a new th e r ^ u tic approach to hepatitis B. N at. M ed. 7,221-225. Rhoads, J.M ., M acleod, R.J., and Ham ilton, J R . (1989). Dim inished brush-border m em brane Na-dependent L-alanine transport in acute viral enteritis in piglets. J. Pediatric Gastroenterol. N utrition 9,225-231. R obinson, P.C., Voelker, D R ., and M ason, R.J. (1984). Isolation and culture o f human alveolar type H epithelial cells. C haracterization o f their phospholipid secretion. Am. Rev. Respir. Dis. 130, 1156-1160. R obison, T.W ., Dorio, R.J., and Kim, K.J. (1993). Form ation o f tight m onolayers o f guinea pig airway epithelial cells cultured in an air-interface: B ioelectric properties. Biotechniques 75,468-473. Robison, T.W. and Kim, K.J. (1994). A ir-interface cultures o f guinea pig airw ay epithelial cells: effects o f active Na^ and Cl- transport inhibitors on bioelectric properties. Exp. Lung Res. 2 0 ,101-117. R obison, T.W. and Kim , K.J. (1995). Dual effect o f nitrogen dioxide on barrier properties o f guinea pig tracheabronchial epithelial m onolayers cultured in an air- interface. J. Toxicol. Envir. H ealth 44, 57-71. Roovers, K.I. and M eckling, G.K.A. (1996). Characterization o f equilibrative and concentrative Na^-dependent (cif) nucleoside transport in acute prom yelocytic leukemia N B 4 cells. J. Cell. Physiol. 166, 593-600. Sabourin, P.J., Tynes, R., Smyser, B., Dahl, H., and Hodgson, E. (1988). The FAD- containing monooxygenase: oxidation o f nitrogen-, sulfur-, and phosphorus-containing com pounds. In Biological reactive interm ediates, m . M olecular and cellular m achanism s o f action in anim al m odels and hum an disease. R. Snyder, ed. (N ew York: Plenum Press), pp. 263-280. Saha, P., Yamahara, H., Kim, K .J., Crandall, E.D ., and Lee, V .H .L. (1994). Influence o f lipophilicity on b-blocker transport across rat alveolar epithelial cell layers. J. Contr. R el. 32, 191-200. Saum on, G., Seigné, E., and Clérici, C. (1990). Evidence for a sodium -dependent sugar transport in rat tracheal epithelium . Biochim . Biophys. Acta 1023, 313-318. 268 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Schafer, J.A . and Barfiis, D.W. (1986). M echanism o f transm em brane transport in isolated cells and their experimental study. In M em brane transport o f anitneoplastic agents. I D. Goldm an, ed. (New York: Pergam on Press), Schanker, L.S. (1978). Drug absorption from the lung. Biochem. Pharm acol. 27, 381- 385. Schneeberger, E.E. (1980). Heterogeniety o f tight junction m orphology in extrapulm onaiy and intrapulm onary airways o f the rat. Anat. Rec. 198, 193-208. Schoenw ald, R.D. and Huang, H.S. (1983). C om eal penetration behavior o f beta blocking agens I: Physicochem cial factors. J. Pharm . Sci. 72,1266-1272. Schulze, K. and Schultheiss, H P. (1995). The role o f the ADP/ATP carrier in the pathogenesis o f viral heart disease. Eur. H eart J. 16, 64-67. Seale, J.P . (1990). Chem ical targeting o f drug therapy in asthm a and airw ay disease. Adv. D rug D eliv. Rev. 5, 73-92. Shen, B .Q ., Finkbeiner, W.E., W ine, J.J., Nfrsny, R .J., and W iddicombe, J.H . (1994). Calu-3: a hum an airw ay epithelial cell line w hich show s cAM P-dependent C f secretion. Am. J. Physiol. 266, L493-L501. Shi, B. and Tien, H.T. (1986). A ction o f calcium chaim el and beta-adrenergic blocking agents in the bilayer lipid membranes. Biochim . Biophys. Acta 859, 125-134. Shi, M .M . and Young, J.D. (1986). [^H]-DipyridamoIe binding to nucleoside transporters from guinea-pig and rat lung. Biochem . J. 240, 879-883. Sim on, E.E. and M cDonald, J.A. (1990). E xtracellular m atrix receptors in the kidney cortex. Am . J. Physiol. 259, F783-F792. Sinko, P .J., Patel, N .R ., and Hu, P. (1994). Int. J. Pharm . 109, 125-133. Sleigh, M .A ., Blake, J R., and Liron, N . (1988). The propulsion o f m ucus by cilia. Am. Rev. R espir. D is. 157, 726-731. Sm ith, L.L. and W yatt, I. (1981). The accum ulation o f putrescine into slices o f rat lung and brain and its relationship to the accum ulation o f paraquat. Biochem . Pharmacol. 50, 1053-1058. 269 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sorbara, L .R ., M aldarelli, F., M itsuya, H., Sim pson, I.A., and Zeichner, S.L. (1996). Hum an im m unodeficiency virus type I infection o f H9 cells induces increased glucose transporter expression. J. V irol. 70, 7275-7279. Spector, R . (1985). Uridine transport and m etabolism in the central nervous system . J. N eurochem . 45, 1411-1418. Speirs, V . and Cutz, E. (1993). A n overview o f culture and isolation m ethods fo r in vitro studies on pulm onary neuroendocrine cells. Anat. Rec. 236, 35-40. Stanley, C ., Rosenberg, M .B ., and Friedmann, T . (1991). G ene transfer into rat airw ay epithelial cells using retroviral vectors. Som atic Cell. M ol. G enetics 1 7 ,185-190. Sterk, P .J. (1993). V irus-induced airways hyperresponsiveness in man. Eu. R espir. 6, 894-902. Stout, S.A . and Derendorf, H . (1987). Local treatm ent o f respiratory infections w ith antibiotics. D rug Intell. C lin. Pharm . 21, 322-329. Stow, R .A . and Bronk, J.R . (1993). Purine nucleoside transport and m etabolism in isolated ra t jejenum . J. Physiol. 46 8 ,311-324. Stutts, M .J., C otton, C.U., Y ankaskas, J R., Cheng, E., K now les, M R., Gatzy, J.T ., and B oucher, R .C . (1985). C hloride uptake in cultured airway epithelial cells from cystic fibrosis patients and norm al individuals. Proc. N atl. Acad. Sci. USA 8 2 ,6677-6681. Stutts, M .J., Schwab, J.H ., Chen, M.G., K now les, M R., and B oucher, R.C. (1986). Effects o f Pseudomonas aeruginosa on bronchial epithelial ion transport. Am . Rev. Respir. D is. 134, 17-21. Svensson, L., Finley, B.B., B ass, D., von B onsdorff, C., and G reenberg, H.B. (1991). Sym m etric infection o f rotavirus on polarized hum an intestinal epithelial C aco-2 cells. J. V irol. 65, 4190-4197. Svensson, U ., Persson, R., and Everitt, E. (1981). V irus-receptor interaction in the adenovirus system : I. Identification o f virion attachm ent proteins o f the HeLa cell plasm a m em brane. J. Virol. 38, 70-81. T avakoli-Saberi, M R. and A udus, K.L. (1989). Physicochem ical factors affecting beta- adrenergic antagonist perm eation across cultured ham ster pouch buccal epithelium . Int. J. Pharm . 5 6 , 135-142. Taylor, G . (1990). The absorption and m etabolism o f xenobiotics in the lung. A dv. Drug D eliv. Rev. 5, 37-61. 270 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Terzaghi, M ., N ettesheim , P., and W illiams, M .L. (1978). Repopulation o f denuded tracheal grafts w ith norm al, preneoplastic, and neoplastic epithelial cell populations. C ancer R es. 55,4546-4553. Thom pson, L.F. (1985). Ecto-5'-nucleotidase can provide the total purine requirem ents o f m itogen-stim ulated hum an T-cells and rapidly dividing hum an lym phoblastoid cells. J. Im m unol. 134, 3794-3797. T orres, M ., Delicado, E.G ., Fideu, M.D., and M iras-Portugal, M .T. (1992). Down- regulation and recycling o f the nitrobenzylthioinosine-sensitive nucleoside transporter in cultured chrom affin cells. Biochim. Biophys. Acta 1105,291-299. Trum p, B .F., Resau, J., and Barrett, L.A. (1980). M ethods o f organ culture fo r hum an bronchus. In M ethods in Cell Biology. C.C. H arris, B.F. Trump, and G.D. Stoner, eds. (N ew Y ork: Academ ic Press), pp. 1-14. Tsuzuki, N ., Hama, T ., Kawada, M., Hasui, A ., Konishi, R., Shiwa, S., O chi, Y., Futaki, S., and Kitagawa, K. (1994). Adamantane as a brain-directed drug carrier fo r poorly absorbed drug. 2. A ZT derivatives conjugated w ith the 1-adam antane m oiety. J. Pharm . Sci. 8 3 ,481-484. Tucker, S.P. and Com pans, R.W . (1992). V irus infection o f polarized epithelial cells. Adv. V irus Res. 42, 187-247. Ulug, E.T., Garry, R .F., and Bose, H R. (1989). The role o f m onovalent cation transport in Sindbis virus m aturation and release. V irol. 172,42-50. U ssing, H.H. and Z ehran, K. (1950). A ctive transport o f sodium as the source o f electric current in the short-circuited isolated frog skin. Acta Physiol. Scand. 23, 110-126. van B ree, J.B ., de B oer, A.G., Danhof, M ., G insel, L.A., and Breim er, D.D. (1988). C haracterization o f an "in vitro" blood-brain barrier: Effects o f m olecular size and lipophilicity on the cerebrovascular endothelial transport rates o f drugs. J. Pharm acol. Exp. Ther. 247, 1233-1239. V an Scott, M .R., Lee, N .P., Yankaskas, J R ., and Boucher, R.C. (1988). E ffect o f horm ones on grow th and function o f cultured canine tracheal epithelila cells. Am. J. Physiol. 255, C237-C245. V an Scott, M R., Cheng, P.W ., Henke, D C ., and Yankaskas, J R. (1991). C ell culture o f airw ay epithelia. In The Airway Epithelium: Physiology, Pathophysiology and Pharm acology. S.G. Farm er and D.W.P. Hay, e& . (New York: M arcel D ekker, Inc.), pp. 135-167. 271 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Vijayalakshm i, D. and Belt, (1988). Sodium -dependent nucleoside transport in m ouse intestinal epithelial cells. J. Biol. Chem. 25, 19419-19423. W achsman, M ., Petty, B.G., Cundy, K .C., Jaffe, H.S., Fisher, P.E., Pastelak, A ., and Lietm an, P S. (1996). Pharm acokinetics, safety and bioavailability o f HPMPC (cidofovir) in hum an im m unodeficiency virus-infected cells. A ntiviral Res. 29, 153-161. W aclawski, A.P. and Sinko, P.J. (1996). Oral absorption o f anti-acquired immune deficiency syndrom e nucleoside analogues.2. Carrier-m ediated intestinal transport o f stavudine in rat and rabbit preparations. J. Pharm . Sci. 8 5 ,478-485. W ang, W., Sasaki, H ., Chien, D .S., and Lee, V.H.L. (1991). Lipophilicity influence on conjunctival drug penetration in the pigm ented rabbit: a com parison w ith com eal penetration. Curr. Eye Res. 10, 571-579. W angensteen, D. and Yankovich, R. (1979). Alveolar epithelium transport o f album in and sucrose: concentration difference effect. J. Appl. Physiol. 47, 846-850. W angensteen, O.D ., Schneider, L .A ., Fahrenkrug, S.C., Brottm an, G.M., and M aynard, R.C . (1993). Tracheal epithelial perm eability to nonelectrolytes: species difference. J. A ppl. Physiol. 75, 1009-1018. W annamaker, V.L. and Nagy, L .E . (1995). Equilibrative adenosine transport in rat hepatocytes after chronic ethanol feeding. Alcoholism C lin. Exp. Res. 19, 735-740. W amick, C.T., M uzik, H., and Paterson, A.R.P. (1972). Interference w ith nucleoside transport in m ouse lymphom a cells proliferating in culture. C ancer Res. 32, 2017-2022. W elsh, M.J. (1987). Electrolyte transport by airway epithelia. Physiol. Rev. 67, 1143- 1184. W hitcutt, M .J., A dler, K.B., and W u, R. (1988). A biphasic cham ber for m aintaining polarity o f differentiation o f cultured respiratory tract epithelial cells. In Vitro Cell. Dev. Biol. 2 4 ,420-428. W iddicombe, J.H ., Coleman, D .L ., Finkbeiner, W.E., and Tuet, LK. (1985). Electrical properties o f m onolayers cultured from cells o f human tracheal epithelium. J. Appl. Physiol. 5 8 ,1729-1735. W iddicombe, J.H . (1986). C ystic fibrosis and beta-adrenergic response o f airw ay epithelial cell cultures. Am. J. Physiol. 251, R818-R822. 272 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. W illiam s, T.C., D oherty, A .J., G riffith, D .A ., and Jarvis, S.M . (1989). Characterization o f sodium -dependent and sodium -independent nucleoside transport system s in rabbit brush-border and basolateral plasma-membrane vesicles firom the renal outer cortex. Biochem . J. 264,2 2 3 W illiam s, T.C. and Jarvis, S.M. (1991). M ultiple sodium -dependent nucleoside transport system s in bovine renal brush-border m em brane vesicles. Biochem . J. 274, 27- 33. W inther, B ., Gw altney, J.M ., and Hendley, J.O. (1990). Respiratory virus infection o f m onolayer cultures o f hum an nasal epithelial cells. A m . Rev. Respir. Dis. 141, 839-845. W ohlheuter, R.M ., M arz, R., Graff, J.C ., and Plagem ann, P.G.W . (1978). A rapid m ixing technique to m easure transport in suspended anim al cells; applications to nucleoside transport in N ovikoff rat hepatom a cells. M ethods Cell Biol. 2 0 ,211-236. Wu, R ., N olan, E., and Turner, C. (1985). Expression o f tracheal differentiated functions in serum -fiee horm one-supplem ented m edium . J. C ell. Physiol. 125, 167-181. Wu, R ., M artin, M R ., Robinson, C.B., St George, J.A ., Piopper, C.G., K urland, G., Last, J.A ., Cross, C .E., M cDonald, R .J., and Boucher, R . (1990). Expression o f mucin synthesis and secretion in human tracheobronchial epithelial cells grow n in culture. Am. J. R espir. C ell M ol. B iol. 3 , 467-478. W u, R. and Sm ith, D. (1982). Continuous m ultiplication o f rabbit tracheal epithelial cells in a defined, horm one supplemented medium. In V itro C ell. Dev. B iol. 1 8 ,800- 812. Wu, X ., Yuan, G., B rett, C.A., Hui, A C ., and G iacom ini, K.M . (1992). N ucleoside transport in the choroid plexus, identification o f a novel Na^-nucleoside transporter w ith broad purine and pyrim idine selectivity. J. Biol. Chem . 267, 8813-8818. Wu, X ., Gutierrez, M .M ., and Giacom ini, K.M . (1994). Further characterization o f the sodium -dependent nucleoside transporter (N3) in the choroid plexus fiom the rabbit. B ioch im . Biophys. A cta IJ91, 190-196. Yam ahara, H., M orim oto, K., Lee, V .H .L., and Kim , K .-J. (1994). Effects o f protease inhibitors on vasopressin transport across rat alveolar epithelial cell m onolayers. Pharm aceutical R esearch II, 1617-1622. Y am ashita, P., Kim, K .J., and Lee, V.H.L. (1996). T ransport characteristics o f L- cam osine in rabbit tracheal epithelial cell m onolayers cultured at an air-interface. Pharm. Res. 13, S249(A bstract) 273 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Y am aya, W ., Finkbeiner, W .E., C hun, S.Y ., and W iddicombe, J.H . (1992). D ifferentiated structure and function o f cultures firom human tracheal epithelium . Am. J. Physiol. 262, L713-L724. Y archoan, R., Pluda, J.M ., Pem o, C F ., M itsuya, H., Thomas, R .V ., W yvill, K.M ., and B roder, S. (1990). In itial clinical experim ents w ith dideoxynucleosides as single agents and in com bination therapy. AIDS: A nti-H IV agents, therapies and vaccines. Ann. NY A cad. Sci. 616, 328-343. Z ietlin, P.L., L oughlin, G.M ., and G uggino, W .B. (1988). Ion transport in cultured fetal and ad ult rabbit tracheal epithelium . Am . J. Physiol. 254, C691-C698. Zim m erm an, T .P., M ahony, W .B., and Prus, K. (1987). 3'-Azido-3'-deoxythym idine, an unusual nucleoside analog that perm eates the m em brane o f hum an erythrocytes and lym phocytes by nonfacilitated diffiision. J. B iol. Chem. 262, 5748-5754. Z ou, F.C ., D utschm an, G.E., De C lercq, E., and Cheng, Y.C. (1984). Differential binding afiinities o f sugar m odified derivatives o f (E)-5-(2-brom ovinyl)-2'-deoxyuridine for herpes sim plex virus-induced and hum an cellular deoxythym idine kinases. Biochem. Pharm acol. 33, 1797-1800. 274 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Basolateral transport of nucleosides in rabbit tracheal epithelial cells: An equilibrative nucleoside transporter outcome

PDF

Drug efflux pumps in rabbit conjunctival epithelial cells

PDF

Drug transport mechanisms in the rabbit corneal epithelium: In vitro-in vivo correlation

PDF

Etk/Bmx activation modulates barrier function in epithelial cells

PDF

Conjunctival epithelial uptake of biodegradable nanoparticles: Mechanism, intracellular distribution, and absorption enhancement

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

Colchicine protects mice from the lethal effect of agonistic anti -Fas antibody

PDF

Effect of lipidization on transport and uptake of peptides across rat alveolar epithelial cell monolayers

PDF

DNA structures associated with the Fragile X triplet repeat sequences

PDF

Cationic cell penetrating peptides: characterization of transport properties in epithelial cells and their utilization as delivery systems for protein and peptide drugs

PDF

Biochemical, molecular biological and computer modeling studies of the conjunctival dipeptide transporter isoforms

PDF

Computer prediction of peptide binding to class I MHC molecules

PDF

Characterization and identification of endogenous factor(s) that enhance insulin transport across primary rat alveolar epithelial cell monolayers

PDF

Cytochrome P450 3A metabolism in the rabbit lacrimal gland and conjunctiva

PDF

Homeostatic regulation of intracellular signaling networks by Etk /Bmx

PDF

The polymeric immunoglobulin receptor in ocular tissues

PDF

A dynamic apical actin cytoskeleton facilitates exocytosis of tear proteins in rabbit lacrimal acinar epithelial cells

PDF

Pharmacokinetic And Cytotoxicity Studies Of Oxazaphophorine Metabolites

PDF

Alteration of the in vitro and in vivo processing of a polypeptide, BBI, through conjugation with palmitic acid

Asset Metadata

Creator Mathias, Neil Raymond (author)

Core Title An air-interface rabbit tracheal epithelial cell culture model for the characterization of nucleoside and nucleoside drug transport

Degree Doctor of Philosophy

Degree Program Pharmaceutical Sciences

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Health Sciences, Pharmacy,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Lee, Vincent (committee chair), Haworth, Ian (committee member), Kim, Kwang-Jin (committee member), Shen, Wei-Chiang (committee member), von Grafenstein, Herman (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c17-305383

Unique identifier UC11350220

Identifier 9816102.pdf (filename),usctheses-c17-305383 (legacy record id)

Legacy Identifier 9816102.pdf

Dmrecord 305383

Document Type Dissertation

Rights Mathias, Neil Raymond

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