Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Basolateral transport of nucleosides in rabbit tracheal epithelial cells: An equilibrative nucleoside transporter outcome
(USC Thesis Other)
Basolateral transport of nucleosides in rabbit tracheal epithelial cells: An equilibrative nucleoside transporter outcome
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
NOTE TO USERS
This reproduction is the best copy available.
®
UMI
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
BASOLATERAL TRANSPORT OF NUCLEOSIDES IN
RABBIT TRACHEAL EPITHELIAL CELLS:
AN EQUILIBRATIVE NUCLEOSIDE TRANSPORTER OUTCOME
by
Sharon Kuang-Hui Wu
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 2002
Copyright 2002 Sharon Kuang-Hui Wu
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3094418
Copyright 2002 by
Wu, Sharon Kuang-Hui
All rights reserved.
®
UMI
UMI Microform 3094418
Copyright 2003 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF SOUTHERN CALIFORNIA
The Graduate School
University Park
LOS ANGELES, CALIFORNIA 90089-1695
This dissertation, wr i t t e n b y
.5 ~ H u ^-_kk_ _ _
Under th e direction o f h . M J C . D issertation
C o m m i t t e e , a n d a p p r o v e d by all its me m b e r s ,
has been p resen ted to an d a ccep ted b y The
Graduate School, in p a rtia l fu lfillm en t o f
r e q u i r e m e n t s for th e degree o f
DOCTOR OF PHILOSOPHY
Dean o f G raduate S tu dies
D ate A u g u st 6 , 2002
DISSERTA TION COMMITTEE
Chairperson
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGMENTS
My deepest gratitude goes to my graduate advisor and coach, Dr. Vincent Lee. His
enthusiasm for science, dedication to mentoring, and advocacy for freedom of scientific
inquiry kept me on the right track throughout my graduate career. It was a great pleasure
to conduct this dissertation under his supervision in a motivating inspiring and caring
environment. I have learned from him to be untiring, focused and confident. It was an
honor to be on his championship team.
I sincerely thank my guidance committee members Drs. Ian Haworth, Kwang-Jin
Kim, and David Ann for their stimulating suggestions and tremendous encouragement.
They challenged me to constantly strive for cutting edge research in my project. A special
note of thanks goes to Drs. David Ann and Helen Lin for indoctrinating me in the
principles and tools in molecular biology—the backbone of my project..
I am grateful to all of my colleagues in the Lee Lab for their help and discussions
and for being the best of friends. Above all, I grateful thank my parents for their selfless
love and support, and to my sister for her care and companionship throughout my
graduate years.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
ACKNOWLEDGMENTS.........................................................................II
LIST OF TABLES................................................................................... IX
LIST OF FIGURES...................................................................................X
LIST OF SCHEMES.............................................................................XIV
LIST OF ABBREVIATIONS ..... XV
ABSTRACT................... XVII
I. INTRODUCTION............. .1
A SUMMARY OF THE EVOLUTION OF RESEARCH IN NUCLEOSIDE TRANSPORT
1. O v e r v ie w
1.1. Physiological role of nucleoside transporters: uptake or efflux?......................2
1.1.1. Salvage pathway.................................................................................. 13
1.1.2. Cell to cell signaling.............................................................................15
1.2. Functional history.......................................................................................... IB
1.3. Our tracheal epithelial cell culture model..................................................... 23
1.3.1. Functionally identified nucleoside transport processes......................23
1.3.2. Usefulness for the evaluation of nucleoside drug transport..................24
2. H e t e r o g e n e it y o f M a m m a l ia n E q u il ib r a t iv e N u c l e o s id e
T r a n s p o r t e r s (E N T s) .....................................................................................................................26
2.1. Molecular history........................................................................................... 26
2.2. Genomic information..................................................................................... 29
2.3. Molecular characteristics............................................................................... 30
2.4. Functional characteristics.............................................................................. 33
2.4.1. Kinetics................................................................................................ 33
2.4.2. Pharmacology...................................................................................... 34
2.5. Distribution of ENTs..................................................................................... 35
2.5.1. Tissue and cell localization......................................................... 35
2.5.2. Subcellular localization........................................................................37
2.6. Stmcture-function relationships......................................................................38
2.6.1. Changing the binding sites...................................................................38
2.6.2. Inhibitors and reactive chemical modifiers......................................... 41
2.6.3. Substrate specificity ........................ 42
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.6.4. Site-directed mutagenesis..................... .....43
2.6.5. Key structural requirements .................................................. 44
2.7. Modulation........................................ 45
2.7.1. Second messengers.............................................................................. 47
2.7.2. Hormones.............. 48
2.7.3. Diet and expression.................................... 49
2.7.4. Injury and inflammation........................ 49
3 . T h e r a p e u t ic im p o r t a n c e o f E N T s a s D r u g T a r g e t s ......... 53
3.1. Antiviral drugs............................................................................................... 53
3.2. Anticancer drugs................................................ 54
3.3. Possible relations of ENTs and purinoceptors............................................... 55
3.4. Transporter inhibitors.................................................................................... 59
3.5. Clinical disorders of purine metabolism in salvage pathway.........................60
I I . S T A T E M E N T O F P R O B L E M ....................... 62
N u c l e o s id e s: A f a m il y o f h y d r o p h il ic m o l e c u l e s im p e r m e a b l e
ACROSS CELL MEMBRANES
1. O b je c t iv e s a n d G o a l s .................... 63
2. C e n t r a l H y p o t h e s is ....................................... 63
3. S p e c if ic A i m s ...................................................................................................................................64
I I I . M E T H O D S ................................... 74
T o o l s f o r m e a s u r in g n u c l e o s id e t r a n s p o r t a n d d e t e c t in g
INVOLVED TRANSPORTERS
1. C e l l C u l t u r e s ...................................... 75
1.1. Rabbit tracheal epithelial cell monolayers (RTEC) for functional
characterization..................................................................................................... 75
1.2. Cell line cultures for transient transfection.....................................................76
1.2.1. Cells grown on cluster plates for substrate uptake...............................76
1.2.2. Cells grown on Clearwells® for asymmetry uptake.............................76
2. F u n c t io n a l C h a r a c t e r iz a t io n o f P r im a r y C u l t u r e d R T E C ..............77
2.1. Uridine metabolism studies ................................................................77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.2.3H-uridine uptake studies.............................................................................. 77
2.2.1. Time course studies.............................................................................77
2.2.2. Effects of uridine concentration.......................................................... 78
2.2.3. Inhibition studies with NBMPR.......................... 79
2.2.4. Substrate selectivity............................................................................. 79
2.3.3H-uridine efflux studies.......................... 79
3. T is s u e D is t r ib u t io n ................................... 80
3.1. Northern blot analysis.................................................................................... 80
3.2. Reverse transcription-polymerase chain reactions (RT-PCR)........................81
3.2.1. Reverse transcription of cDNA from rabbit tracheal RNA..................81
3.2.2. Degenerate polymerase chain reaction (PCR).....................................81
3.3. Ribonuclease protection assay (RPA)............................................................83
4. M o l e c u l a r C l o n in g ................................................................ 84
5. M o l e c u l a r C h a r a c t e r iz a t io n ...................................................................................... 85
5.1. Establishment of heterologous expression systems........................................85
5.1.1. Construction of DNA plasmids............................................................85
5.1.2. Transient transfection.......................................................................... 86
5.2. Detection of recombinant protein expression................................. 87
5.2.1. Western blot analysis...........................................................................87
5.2.2. Membrane orientation..........................................................................88
5.2.3. Membrane localization..................... 89
5.2.4. Cellular colocalization......................................................................... 90
5.3. Functional characterization of transiently transfected cells............................91
5.3.1. Uptake studies in the transfected cells.................................................91
a) Concentration dependency..............................................................91
b) Effect of NBMPR............................................................. 92
c) Substrate selectivity.........................................................................92
5.3.2. Efflux studies of transiently transfected cells......................................92
5.4. Biochemical Characterization.........................................................................93
5.4.1. Lys-C digestion................................................................................... 93
5.4.2. PNGase F treatment.............................................................................93
6. M o d u l a t io n ................... ..9 4
6.1. Pharmacological modulation..........................................................................94
6.2. Transcriptional and post-translational modulation.........................................94
6.2.1. Procedures for nucleoside deprivation.................................................94
6.2.2. Determination of nucleoside levels by high performance liquid
chromatography (HPLC)................................................................................95
6.2.3. Semi-quantitative RT-PCR uptake studies................................... 96
6.2.4. Evaluation of basolateral uridine uptake in nucleoside-deprived
cells.................................................................................................................97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IV. RESULTS 99
D is c o v e r ie s a n d c o n f ir m a t io n s
1. F u n c t io n a l E v id e n c e f o r B a so l a t e r a l N u c l e o s id e T r a n s p o r t
P r o c e s s e s ............................. .1 0 0
1.1. Time course and Na+ -dependency............................................................... 100
1.2. Concentration-dependency.......................................................................... 100
1.3. NBMPR sensitivity ............................................................................... 102
1.4. Substrate selectivity..................................................................................... 102
1.5. Efflux studies............................................................................................... 104
2. M o l e c u l a r a n d F u n c t io n a l C h a r a c t e r is t ic s o f Id e n t if ie d E N T 2
a n d i t s S p l i c e V a r ia n t f r o m R a b b it T r a c h e a ............................................................107
2.1. Molecular characteristics............................................................................ 107
2.2. Tissue distribution....................................................................................... 112
2.3. Expression and detection..............................................................................117
2.3.1. Apparent molecular mass...................................................................117
2.3.2. Membrane localization.......................................................................118
2.3.3. Cellular colocalization........................................................................120
2.4. Functional characteristics.............................................................................126
2.4.1. In HEK heterologous expression system...........................................126
a) Validation of HEK heterologous expression system.....................126
b) Time course...................................................................................131
c) NBMPR, dilazep, or dipyridamole sensitivity...............................131
d) Efflux studies.................................................................................134
e) Concentration dependency............................................................134
f) Substrate specificity......................................................................137
g) Coexpression.................................................................................137
2.4.2. In MDCK heterologous expression system........................................140
a) Validation of MDCK heterologous expression system.................140
b) Effect of epitope tagging on membrane expression and
functionality .........................................................................142
c) Concentration dependency............................................................142
d) NBMPR sensitivity........................................................................145
e) Substrate selectivity.......................................................................145
2.5. Functionality, expression, and subcellular distribution of the
C-terminal truncated rabbit ENT2.......................................................................148
2.5.1. Functionality and expression..............................................................148
2.5.2. Subcellular distribution...................................................................... 150
2.6. Structural validation......................................................................................150
2.6.1. Verification of membrane orientation................................................150
2.6.2. Lys-C digestion..................................................................................155
2.6.3. PNGase F treatment...........................................................................162
with permission of the copyright owner. Further reproduction prohibited without permission.
3 . F u n c t io n a l a n d M o l e c u l a r M o d u l a t io n o f B a s o l a t e r a l E N T s 162
3.1. Pharmacological modulation of basolateral nucleoside transport
activity........................................................................................................ 162
3.2. Transcriptional and post-translational modulation of ENTs............... 168
3.2.1. Extent of depletion on intracellular nucleoside pool following
nucleoside deprivation.......................................................... 168
3.2.2. Time course of rbENT2/2A mRNA levels following
nucleoside deprivation............................ 168
3.2.3. Evaluation of basolateral uridine uptake in nucleoside-
deprived cells............................................................................................... 171
3.2.4. Effect of actinomycin D or cycloheximide on rbENT2/2A
mRNA level and basolateral transporter activity.........................................171
V. DISCUSSION.... ..... 176
R e l a t iv e r o l e a n d c o n t r ib u t io n o f E N T 2/2A in m e d ia t in g
NUCLEOSIDE TRANSPORT ACROSS RABBIT TRACHEAL EPITHELIAL CELLS
1. F u n c t io n a l E v id e n c e f o r B a s o l a t e r a l 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 P r o c e s se s in R T E C M o n o l a y e r s .................. 177
1.1. Basic characteristics of transporters ........................................... 177
1.2. Relative distribution and population................................................. 178
1.3. Drug delivery applications............................................................................180
1.4. Molecular rationale.......................................................................................182
2. M o l e c u l a r , C e l l u l a r , a n d S t r u c t u r a l S im il a r it ie s a n d
D if f e r e n c e s b e t w e e e n Id e n t if ie d R a b b it E N T 2 a n d I t s S p l ic e
V a r ia n t ........................................................................................................................ 183
2.1. Genetic information......................................................................................183
2.2. Tissue distribution........................................................................................183
2.3. Functional characteristics.............................................................................184
2.3.1. In HEK heterologous expression system...................................................184
2.3.2. In MDCK heterologous expression system........................................186
2.3.3. Functionality expression, and subcellular distribution of the
C-terminal truncated rbENT2 protein..........................................................188
2.4. Polarized distribution....................................................................................191
2.5. Subcellular localization................................................................................192
2.6. Putative structural features...........................................................................195
2.6.1. Membrane orientation.................................................. 196
2.6.2. Lys-C digestion ...............................................................................198
2.6.3. PNGase F treatment...........................................................................201
3. M o d u l a t io n o f R a b b it E N T s ...... 201
3.1. Pharmacological modulation of basolateral nucleoside exit of RTEC
monolayers...........................................................................................................201
3.1.1. Modulation by protein kinases.........................................................202
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.1.2. Modulation by cytoskeletal components.............. 203
3.2. Transcriptional modulation of rabbit ENTs............................................... 206
3.2.1. System for metabolic stress.............................................. 207
3.2.2. Effect of starvation and antimetabolites..................... 208
3.2.3. Therapeutic significance........................................................... 212
V I. C O N C L U S IO N ................... 214
Highlights and gaps
1. Summary of Finding s ....................... ...215
1.1. Functional characterization of basolateral nucleoside transport...................215
1.2. Functional characterization and pharmacological classification of
rbENT2A............................................................................................................ 215
1.3. Structural validation..................................................................................... 216
1.4. Physiological regulation of basolateral nucleoside transport processes 218
1.5. Gene regulation of ENTs in homeostasis of nucleosides.............................218
2. S ig n if ic a n c e o f t h e F in d in g s ................... 2 19
2.1. Functional importance..................................................................... 221
2.2. Molecular insights....................................................................................... 222
2.3. Summary of key findings............................................................................. 223
3. F u t u r e C o n s id e r a t io n s ................................................................................................. 224
3.1. What is the physiological role of rbENT2/2A? Are there any
phenotypes associated with rbENT dysfunction?.................................. 225
3.2. New sorting motif via alternative splicing?..................................................226
3.3. New functional characteristics via alternative splicing?..............................227
3.3.1. Implications in modulation by physiological machinery...................228
3.4. What is the endogenous expression and distribution profile of
rbENT2/2A?.................................................................. 229
3.4.1. Development of antibodies for immunodetection..............................230
3.4.2. Additional applications of antibodies.................................................231
3.5. What is the consequence of molecular and structural markers due to alternative
splicing?.............................................................................................................. 232
3.5.1. Biotech approach................................................................................232
3.5.2. In Silico approach...............................................................................233
VII. REFERENCES .......... 234
M o v e r s, s h a k e r s a n d p io n e e r s.
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
Table 1-1. Functional characteristics of nucleoside transporter subclasses.
Table 1-2. Nucleoside transport (NT) processes of human cells.
Table 1-3. Members of the ENT and CNT family of transporters.
Table 1-4. The Km values for permeants of h/rENTl or h/rENT2
recombinant proteins expressed in Xenopus oocyte systems.
Table 1-5. Effect of different substances and experimental conditions on
the expression and activity of the equilibrative nucleoside transporters.
Table 4-1. Na+ -dependency of initial uptake rates, kinetic parameters (Km
and Jm ax), and apparent permeability coefficients (Pa p p ) of H-uridine
measured from apical and basolateral fluids of RTEC monolayers.
Table 4-2. Inhibition of rbENT2-mediated uridine uptake by NBMPR,
dipyridamole, and dilazep.
Table 4-3. Effect of A23187 and forskolin on kinetic properties of
basolateral 3H-uridine uptake in RTEC monolayers.
Table 4-4. Basolateral 3H-uridine uptake for 1 min by tracheocyte
monolayers after pharmacological manipulation.
Table 5-1. Kinetic parameters of uridine uptake
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
P age
Fig. 1-1. Production, release, and metabolism of adenosine. 5
Fig. 1-2. Chemical structures of purine and pyrimidine bases. 7
Fig. 1-3. Drug uptake into infected epithelial cells via basolateral ENTs 9
Fig. 1-4. Intestinal drug transport mechanisms. 12
Fig. 1-5. Role of ENTs in drug efflux in epithelial systems. 13
Fig. 1-6. Chemical structures of endogenous and synthetic nucleosides that
ave been reported to interact with nucleoside transporters from Na+ -dependent
or -independent superfamilies.
2 0
Fig. 1-7. Topological model of mammalian equilibrative nucleoside
transporters.
32
Fig. 1-8. Chemical structures of adenosine receptor agonists with subtype
selectivity.
57
Fig. 1-9. Chemical structures of adenosine receptor antagonists with
subtype selectivity.
58
Fig. 2-1. Central hypothesis. 64
Fig. 4-1. Inhibition of 3H-uridine uptake from basolateral fluid in primary
cultured RTEC as a function of NBMPR concentration, under Na+ -free condition.
103
Fig. 4-2. Effect of nucleosides, and various nucleoside and nucleobase
analogs on 3H-uridine uptake from basolateral and apical fluid in primary
cultured RTEC monolayers.
105
Fig. 4-3. Efflux of 3H-uridine across the basolateral membrane of primary
cultured RTEC monolayers after apical 3H-uridine preloading in BRS.
106
Fig. 4-4. Nucleotide and deduced amino acid sequences of rbENT2 and
rbENT2A cDNA clones.
109
Fig. 4-5. Diagram of rbENT2 primary transcript and the gene products
produced by alternative splicing.
110
Fig. 4-6. Predicted membrane topological model of rbENT2 and
rbENT2A.
113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 4-7. Deduced amino acid sequence of rbENT2 and rbENT2A
compared with the human and rat equilibrative nucleoside transporters.
114
Fig. 4-8. Detection of rbENT2 and rbENT2A transcripts by Northern blot 116
analysis, RT-PCR and RPA.
Fig. 4-9. Detection of recombinant, HA-tagged, rbENT2 and rbENT2A 119
proteins expressed in HEK 293 or MDCK cells by Western blot analysis.
Fig. 4-10. Confocal fluorescence micrographs of MDCK cells grown on 121
permeable supports.
Fig. 4-11. Functional localization of MDCK cells transiently transfected 122
with pSG5.HA/rbENT2, pSG5.HA/rbENT2A constructs and pSG5.HA parent
vector alone.
Fig. 4-12. rbENT2 and rbENT2A colocalize with an ER marker. 124
Fig. 4-13. rbENT2 or rbENT2A does not colocalize with a lysosomal 125
marker.
Fig. 4-14. rbENT2 or rbENT2A does not colocalize with a nucleus 127
marker.
Fig. 4-15. Functional characterization of endogenous nucleoside transport 129
in untransfected HEK293 cells.
Fig. 4-16. Effect of NBMPR on 3H-uridine uptake by HEK293 cells. 130
Fig. 4-17.3 H-uridine uptake by HEK293 cells transiently transfected with 132
either rbENT2/pcDNA3 or rbENT2A/pcDNA3 constructs.
Fig. 4-18. Efflux of 3H-uridine by rbENT2-transfected HEK293 cells after 135
3H-uridine preloading.
•5
Fig. 4-19. Concentration dependency of H-uridine uptake by HEK293 136
cells transiently transfected with rbENT2/pcDNA3 construct.
Fig. 4-20. Substrate specificity of 3H-uridine uptake by HEK293 cells 138
transiently transfected with rbENT2/pcDNA3 construct.
Fig. 4-21. Functional characterization of 3H-uridine uptake by HEK293 139
cells transiently co-transfected with rbENT2/pcDNA3 and rbENT2A/pcDNA3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 4-22. Functional characterization of endogenous equilibrative 141
nucleoside transport in untransfected MDCK cells.
Fig. 4-23. 3H-uridine uptake by MDCK cells transiently transfected with 143
pSG5.HA/rbENT2, pSG5.HA/rbENT2A constructs, and pSG5.HA parent
vector alone.
Fig. 4-24. Concentration dependency of 3H-uridine uptake by MDCK cells 144
transiently transfected with rbENT2/pcDNA3 and rbENT2A/pcDNA3
constructs.
Fig. 4-25. Effect of NBMPR on 3H-uridine uptake by MDCK cells 146
transiently transfected with rbENT2/pcDNA3 and rbENT2A/pcDNA3
constructs.
Fig. 4-26. Substrate selectivity of various nucleosides on 3H-uridine 147
uptake by MDCK cells transiently transfected with rbENT2/pcDNA3 and
rbENT2A/pcDNA3 constructs.
Fig. 4-27. Detection of HA-tagged, truncated rbENT2 proteins expressed 149
in MDCK cells by Western blot and confocal fluorescence microscopy.
Fig. 4-28. The proposed topological models of rbENT2 and rbENT2A. 151
Fig. 4-29. Validation of cell plasma membrane intact nonpermeabilized 153
condition, and permeabilized condition.
Fig. 4-30. Orientation of the N-terminus of rbENT2 and rbENT2A. 154
Fig. 4-31. Orientation of the C-terminus of rbENT2 and rbENT2A. 156
Fig. 4-32. Digestion with Lys-C protease of the HA-tagged rbENT2 157
transiently transfected in MDCK cells.
Fig. 4-33. Digestion with Lys-C protease of the HA-tagged rbENT2A 159
transiently transfected in MDCK cells.
Fig. 4-34. Deglycosylation of the HA-tagged rbENT2 and rbENT2A with 163
PNGase F.
Fig. 4-35. Concentration dependency of 3H-uridine uptake from the 166
basolateral fluid in primary cultured RTEC monolayers in the presence of
A23187 or forskolin.
xii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 4-36. Extent of depletion on intracellular nucleoside pool following 169
nucleoside deprivation.
Fig. 4-37. Time course of rbENT2/2A mRNA levels following nucleoside 170
deprivation by semiquantitative RT-PCR.
Fig. 4-38. Evaluation of basolateral uridine uptake in nucleoside-deprived 172
cells by 3H-uridine uptake studies.
Fig. 4-39. Effect of actinomycin D or cycloheximide on induced 174
rbENT2/2A mRNA level.
Fig. 4-40. Effect of actinomycin D or cycloheximide on induced 175
basolateral transporter activity.
Fig. 5-1. Carrier mediated basolateral nucleoside transport in primary 181
cultured RTEC.
Fig. 5-2. cAMP dependent modulation of ENT activity. 203
Fig. 5-3. Ca2 + dependent modulation of ENT activity. 205
Fig. 5-4. Model of nucleoside starvation. 208
Fig. 6-1. A schematic summary of functional findings and molecular 220
outlook.
xiii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF SCHEMES
Scheme 2-1. Specific aim #1.
Scheme 2-2. Specific aim #2.
Scheme 2-3. Specific aim #3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF ABBREVIATIONS
[Ca2+ ]i. intracellular Ca2 + concentration
2’-dU - 2 ’-deoxyuridine
3’-dA - 3’-deoxyuridine
5-FU - 5-flurouracil
5'-NT - 5 '-nucleotidase
A23187 - a calcium ionophore
ACV - acyclovir
ADA - adenosine deaminase
A TPS - acquired immunodeficiency syndrome
AK - adenosine kinase
AMP - adenosine monophosphate
Ara-A - adenosine arabinoside
Ara-C - cytosine arabinoside
ATP - adenosine triphosphate
AZT - 3 ’ -azidothymidine
BBB - blood brain barrier
BRS - bicarbonated Ringer’s solution
BSA - bovine serum albumin
cAMP - cyclic adenosine 3 ’ ,5 ’ -monophosphate
CF - cystic fibrosis
cib - concentrative NBMRR-insensitive broad-selective
cif- concentrative NBMRR-insensitive formycin-selective
cit - concentrative NBMRR-insensitive thymidine-selective
CNS - central nervous system
CNT - concentrative nucleoside transporter
cs - concentrative NBMRR-sensitive
CSF - cerebrospinal fluid
csg - concentrative NBMRR-sensitive guanosine-selective
ddA - 2’ ,3 ’ -dideoxyadenosine
ddl - 2’ ,3 ’ -dideoxyinosine
DMEM - Dulbecco’s modified eagle medium
DNase I - deoxyribonuclease I
dNTP - deoxynucleotide
ei - equilibrative NBMPR-insensitive
ENT - equilibrative nucleoside transporter
ES - embryonic stem
es - equilibrative NBMPR-sensitive
FBS - fetal bovine serum
GAPDH - glyceraldehyde-3 -phosphate dehydrogenase
GFP - green fluorescent protein
GST - glutathione S-transferase
GTP - guanosine triphosphate
H89 - a PKA inhibitor
HEK - human embryonic kidney
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
HGPRT - hypoxanthine-guanine phosphoribosyltransferase
HtV - human immunodeficiency virus
HPLC - high performance liquid chromatography
EDU - 5-iodo-2’ -deoxyuridine
IFN-y - interferon y
Jm ax - maximal uptake
Km - Michaehs-Menten constant
M-CSF - macrophage-colony stimulating factor
MDCK - Madin Darby Canine Kidney
MEM - minimal essential medium
MPA - phenolic acid
MTX -methotrexate
N1 - Na+ -nucleoside transporter (purine selective)
N2 - Na+ -nucleoside transporter (pyrimidine selective)
N3 - Na+ -nucleoside transporter (broad selective)
N4 - Na+ -nucleoside transporter (broad selective)
NBMPR - nitro-benzyl-mercapto-purine-ribose
NT - nucleoside transport
PAGE - polyacrylamide gel electrophoresis
Papp - apparent permeability coefficient
PBS - phosphate-buffered saline
PCMBS - p-chloromercuriphenyl sulphonate
PCR - polymerase chain reaction
PKA - protein kinase A
PKC - protein kinase C
PRPP - 5-phosphate-a-D-ribosyl-1 -pyrophosphate
RACE - rapid amplification cDNA ends
RPA - ribonuclease protection assay
RT - reverse transcription
RTEC - rabbit tracheal epithelial cells
SAM - S-adenosyl methionine
SDS - sodium dodecyl sulafate
SFR - sodium-free Ringer’s solution
S-MEM - Ca2 + -free minimum essential medium
SPNT - Na+ -purine nucleoside transporter
TLC - thin layer chromatography plate
TMD - transmembrane domain
X V I
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
The objective of this research was to study the functional and molecular characteristics of
the equilibrative nucleoside transporters (ENT) in the basolateral membrane of primary
cultured rabbit tracheal epithelial cells (RTEC). The long-term goal is to use ENT for
improving the delivery of nucleoside drugs.
RTEC was used to study the functional characteristics of basolateral nucleoside
transport with 3H-uridine as a substrate. This process was bidirectional; Na+ -independent;
temperature (85-90%) and concentration dependent (Km = 210.2 pM, V m ax = 356.8
pmol/mg protein/min); and nitrobenzyl-mercaptopurine-ribose (NBMPR) insensitive
(IC25 = 0.2 pM, IC75 = 270 pM); and purine/pyrimidine selective]. Calcium ionophore,
A23187, inhibited 70% basolateral uridine transport in a protein kinase C (PKC)
independent (tamoxifen-irreversible) and microtubule dependent (colchicine-reversible)
manner. In contrast, transport was increased 51% by adenylyl cyclase activator forskolin,
in a protein kinase A (PKA) dependent (H89-reversible) manner. Pharmacological
inhibition of de novo nucleoside biosynthesis showed that basolateral nucleoside
transport in RTEC offers a salvage pathway by up-regulating transport activity.
High capacity (Vm a x = 1.26 nmol/mg protein/5 min) and low affinity (Km = 272.8
pM) nucleoside carrier (kinetics determined in HEK293 and MDCK cells) rbENT2 had a
novel alternatively spliced variant, rbENT2A, isolated by RT-PCR approach on RNA
prepared from rabbit trachea. rbENT2A (not rbENT2) showed no nucleoside transport
activity when expressed in HEK293 cells, but was active when expressed in MDCK cells
with low capacity (Vm a x = 0.6 nmol/mg protein/5 min) and high affinity (Km = 41.6 pM).
When rbENT2A was co-expressed with rbENT2 in HEK293 cells, the combination
xvii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
inhibited functional nucleoside transport in a dose-dependent manner. In addition,
rbENT2 and rbENT2A highly differ from each other in their size, secondary structure and
membrane-spanning pattern, and subcellular localization, although they are present at a
constant ratio in each tissue of rabbit epithelial tracheocytes, kidney cortex, kidney
medulla, and intestine.
In summary, an equilibrative nucleoside transport process is demonstrated in the
basolateral aspect of the rabbit tracheal epithelial cells. The alternative splicing variant,
rbENT2A possesses a markedly different kinetics and distinct pharmacological profile,
implicating that rbENT2A acts in conjunction with the rbENT2 fine-tuning equilibrative
nucleoside transport function in rabbit trachea.
xviii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I. INTRODUCTION
A SUMMARY OF THE EVOLUTION OF RESEARCH IN NUCLEOSIDE TRANSPORT.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1. O v er v iew
1.1. Physiological role of n u cleo sid e tran sp orters: u p ta k e o r efflu x?
A fundamental definition of the role of all types of transporters in cell function is a
changing and growing model. Essentially, it is important to define the role of transporter
proteins in generally all cell types, in order to offer methods that employ them in drug
design and delivery. For example, in the most classic studies when pioneering scientists
were first investigating the functional properties of ionic channels in excitable
membranes, they came to a general consensus that in virtually all cases membrane
transporters function in a response to counteract any changes from resting state/potential
of the host cell (Hille, 1992). Cell volume changes are known to activate Cl' channels
from the large CIC-family. Cell membrane potential changes can activate potassium
channels, since the electrochemical gradient of this ion is the one that largely controls this
characteristic (Hille, 1992). Finally, changes in the availability of basic cell nutrients can
affect solute transporters. The basic niche of transporters in cells is their role in the
maintenance of homeostasis. Therefore, transporter proteins are the most fundamental
components of cells that can potentially act as mediators of molecular drug transport.
Their role is limited to select drug candidates that happen to be substrates that are
otherwise largely impermeable across cell membranes.
Nucleosides and nucleotides play a crucial role in nearly all types of biochemical
processes, ranging from nucleic acid synthesis (e.g., DNA and RNA) to metabolic
regulators (e.g., cyclic AMP) in biological systems. Most cells are equipped with
biosynthetic machinery necessary to synthesize purine and pyrimidine nucleosides via de
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
novo biosynthetic pathways. Cells that lack, or have impaired de novo pathways can
derive preformed nucleosides or nucleobases from dietary or endogenous nucleic acids
via salvage pathways. These cells include bone marrow cells, enterocytes and brain cells
(Murray, 1971).
Due to their hydrophilic nature, nucleosides do not readily cross biological
membranes by passive diffusion, and specialized plasma membrane nucleoside
transporters are required for inward salvage and outward efflux of physiological
nucleosides from and into extracellular space in the overall nucleoside homeostasis.
Hence, pharmacological manipulation of extracellular (purinergic receptor agonists) and
cytoplasmic (anticancer and antiviral nucleoside analogs) nucleosides can be achieved by
an understanding of functional and molecular mechanisms of nucleoside transport
processes.
The uptake of nucleosides from the intestinal lumen is perhaps the most abundant
exogenous source of these molecules. Breakdown of nutrients in the gastrointestinal tract
releases nucleosides, which are then absorbed across the intestinal epithelium. Plasma
membrane transporters play a key role in nutrient absorption and whole-body
homeostasis. The activity of various transport systems, particularly those involved in
amino acid and carbohydrate uptake, is highly sensitive to food intake and nutrient
composition (Kilberg et al., 1993; Thomson and Wild, 1997). Nutrient transporters in the
small intestine adapt to food supply by modifying their translocation activity, mostly in
response to changes in the number of transporters at the plasma membrane of the
enterocyte (Mailliard et al., 1995; Dyer et al., 1997).
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Dietary nucleosides as precursors for nucleic acid synthesis are important for
rapidly dividing cells since gut epithelial cells have limited capacity for de novo purine
and pyrimidine synthesis, especially during growth and tissue repair (Uauy et al., 1990;
Uauy et al., 1994). For example, semi-purified diets lacking nucleotides can decrease
protein synthesis in the liver and small intestine (Lopez-Navarro et al., 1995; Lopez-
Navarro et al., 1996). Nucleosides are not only degraded but also incorporated into both
organs (Witte et al., 1991). In summary, adaptive modulation of plasma membrane
transporters appears to be necessary for regulation of nutrient uptake and processing in
most organs, especially in intestine and liver. Nucleoside transporters have their own
niche within these specialized tissues. The uptake of nucleosides is not considered as a
regular source of food derived energy, and is secondary to that of carbohydrates, and
amino acids. Specialized epithelial cells conventionally use these molecules as building
blocks for genetic materials, or as modified signaling molecules.
The production, release, and metabolism of adenosine depend on several cellular
factors (Fig. 1-1), but ultimately they are determined by the ratios of relevant enzyme
activities (Brundege and Dunwiddie, 1997). The intracellular concentration of adenosine
depends not only on adenosine-forming enzymes, but also on enzymes that metabolize
adenosine. Thus, the net release or uptake of adenosine depends on the relative activities
of at least four relevant metabolizing enzymes: adenosine kinase (AK), which is involved
in phosphorylation by way of the salvage pathway; adenosine deaminase (ADA), which
deaminates adenosine to inosine; both intracellular and surface-located forms of 5'-
nucleotidase (5’-NT), which produce adenosine from adenosine monophosphate (AMP);
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
^-rndmmarn adenosine
kinase
SAH admmim ■ — inosln©
Sa h a adenosine
T deaminase
intracellular
I n i nucleoside extracellular
©etoS’ - transporter
ATP ^ nucleotidase ■ f
's'-AMP — * ► adenosine
Fig. 1-1. Production, release, and metabolism of adenosine,
and S-adenosylhomocysteine (SAH) hydrolase, which provides a possible alternative
route to adenosine. The family of nucleoside transporters has to act in concert with these
regulatory enzymes. Ultimately, if this pool of adenosine and/or other nucleosides is to
leave the intracellular compartment in which the synthetic enzymes are present in, the
mediation of nucleoside transporters is required. All nucleosides are too hydrophilic to
cross phospholipid bilayers that compartmentalize areas within mammalian cells, and
ultimately divide the extracellular environment from the cytoplasm.
Extracellular nucleosides, not coming from nutrients, have essential roles (Fig. 1-
1). Their origins are specialized tissues and organs that produce and expel them into
blood/plasma via elaborate mechanisms by which most are encompassed within the few
families of nucleoside transporters. It is important to briefly revisit some of the tissues
and organs known to drive out nucleosides or nucleotide precursors.
Adenosine nucleotides are present in bile, which suggests that hepatocytes may
release nucleotides into the canaliculus where they are promptly degraded into adenosine
by ecto-ATPase and 5'-NT, which have been identified in the canalicular plasma
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
membrane but not in rat liver sinusoidal membrane vesicles (Che et al., 1992). Adenosine
is then transported into hepatocytes by a Na+ -dependent nucleoside transporter that is
present in the canalicular plasma membrane. Study of its expression during development
and other physiological circumstances suggests that the transporter may play an
important role in maintaining hepatic purine levels that are essential for the liver to serve
as a major source of purines for tissues (i.e., brain, muscle) that lack pathways for de
novo purine biosynthesis (Che et al., 1997).
Reasons for the initial secretion of nucleotides/nucleosides are two fold:
derivations of regulatory and signal transducing events, or for maintaining homeostasis.
Moreover, in mammals the liver is the major site of de novo purine nucleotide synthesis
and provides purine bases and nucleosides to be salvaged and used by tissues incapable
of de novo synthesis. Hence, the liver may play a partial role in the maintenance of
interorgan nucleoside equilibrium. For example, the brain has a low level of PRP
amidotransferase and partially depends on exogenous purines for proper cellular
functions. Erythrocytes and polymorphonuclear leukocytes cannot synthesize 5’-
phsphoribosylamine, hence must use exogenous purines for biosynthesis of purine
nucleotides. Other cells, such as bone marrow, intestinal mucosal cells, and erythrocytes
are also deficient in de novo purine biosynthesis and depend on preformed purines to be
delivered through the circulatory systems.
The importance of nucleoside transporters in their role of delivering building
blocks for host cells to use in synthesis of genetic material is clear. The process can be
regulated by a number of mechanisms, likely to happen on multiple levels. Purine
nucleosides are particularly sensitive to a unique mechanism. For example, a tightly
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
regulated system helps to maintain proper intra-abdominal extracellular adenosine
concentrations. Damage to tissues that lack a de novo purine biosynthetic pathway may
generate a signal that enhances hepatic purine synthesis, release into plasma, and
recapture from bile. All these steps are closely regulated at the transcriptional or
translational levels. Conversely, severe acute or chronic liver damage may contribute to
pathology in extrahepatic tissues such as brain, muscle, and others that depend on the
liver for purines. Nucleoside transporters are the most elemental cellular units that play a
limiting role in the release or absorption of substrates in the tissues introduced above.
a) Uptake
The most abundant physiological
purines and pyrimidines within
Purine Pyrimidine
oligonucleotides like DNA or RNA is Fig. 1-2. Chemical structures of purine
bases are biosynthesized at comparatively faster rates. The rationale for this observation
is multifold, and can be summarized in a nutshell by the importance of high demand for
molecules like adenosine triphosphate (ATP), guanosine triphosphate (GTP) or S-
adenosyl methionine (SAM). These derivatives of adenosine or guanosine are extremely
important as the basic units of energy currency (ATP, the most abundant molecule in
cells), important signaling molecules (GTP), and the only and major donor of methyl
groups (SAM, is the second most abundant molecule in cells after ATP). Hence the de
1
nucleosides comprise the purine bases
(Fig. 1-2). Generally, the abundance of
relatively uniform. However, purine and pyrimidine bases.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
novo biosynthesis of the purine rings is relatively more weighed upon in most cell types.
Once combined with the ribose sugar moiety, nucleosides are essentially the molecular
form in which interorganelle or interorgan homeostasis of adenosine, guanosine, etc., is
maintained.
Nucleoside transporters are the most fundamental units that play a regulatory role
in shuttling of these abundant, endogenous, physiological substrates across cell
membranes. The presence of nucleoside transporters in multiple types of epithelial cells
(i.e. intestine, kidney, lung, and placenta) facilitates the salvage and recycling of their
substrates from nutrients or bodily waste fluids. The uptake of nucleosides in non-
epithelial systems (i.e. heart smooth muscles or brain astrocytes) is also important for
their proper function.
Functional transport characteristics of nucleosides and nucleobases have been
studied in a number of cells types. More work remains to be done, but the current
understanding and existing knowledge about these transporters allows for the initiation in
development of nucleoside transporter specific drug delivery systems. The role of
equilibrative nucleoside transporters in uptake of various analogous drugs has not been
widely explored. There are a number of biological and practical reasons for this; first,
these family of transporters are primarily located in cellular domains that are usually
facing away from non-invasive routes of drug delivery, and second, their affinity for
substrates is not as high compared to sodium coupled nucleoside transporter family
members. Nevertheless, the delivery of various nucleoside analogue drugs targeted
against epithelial cell infections or disorders can be achieved after systemic
administration of these agents (intravenous (IV) injection or oral bioavailability). For
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
example, to combat various respiratory infections or airway epithelial cell dysfunctions
the required pharmacological effect of nucleoside analogs can be achieved after they
enter these epithelia via facilitated uptake from the blood/serosal side by equilibrative
nucleoside transporters (ENTs) (Fig. 1-3).
A passive carrier is an energy-independent system that can only transport its substrate
down a concentration gradient. A passive transport system is most effective when the cell
is exposed to a fairly constant level of the substrate. For example, the primary function of
the facilitative glucose carriers is to mediate the exchange of glucose between the blood
and the cytoplasm of cells. The facilitative nucleoside transporters essentially form
b) Efflux
Apical
Transport to and from intracellular storage sites
Drug
Basolateral
Efflux
Drug
Drug administered by
oral or TV injection
BLOOD FLOW
Fig. 1-3. Drug uptake into infected epithelial cells via basolateral ENTs.
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
selective pathways between the major pools of purine and/or pyrimidine base
nucleosides, i.e., the cytoplasm of cells which actively de novo biosynthesize these
molecules, and the blood or extracellular fluids.
Nucleosides could be controlled by the tissue-specific expression and regulation
of several nucleoside transport isoforms with distinct functional properties (e.g. kinetics,
substrate selectivity, pharmacology, and spatial distribution). Nucleoside transporter
mediated efflux of substrates appears to account at least in part for the mechanism by
which nucleosides are properly distributed among the various cells and tissues in
mammalian systems. Additionally, the mechanism most commonly implicated in
adenosine release in the central nervous system (CNS), where this nucleoside acts as a
neurotransmitter, involves bi-directional nucleoside transporters (Latini and Pedata,
2001).
The homeostasis of nucleosides in the body under different metabolic or
pathological conditions is primary governed by de novo biosynthesis of purine and
pyrimidine bases in hepatic cells, and their sinusoidal release or efflux. The ability to
transport nucleosides across the plasma membrane is a common feature to nearly all cells,
from the simple bacterium to the highly specialized mammalian neuron. For example,
studies have clearly demonstrated that a significant proportion of labeled adenosine
uptake and release in rat brain cells is mediated by equilibrative nucleoside transporters
(Gu et al., 1995; Foga et al., 1996).
The release of adenosine into the extracellular space is further dependent on
adenosine membrane transporters, which work via a bidirectional equilibrative
nucleoside transport mechanism (Griffith and Jarvis, 1996). The nucleoside
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
transporters are primarily classified according to either sodium dependency or
sensitivity to inhibition by nitro-benzyl-mercapto-purine-ribose (NBMPR) (Griffith
and Jarvis, 1996). Two sodium-independent nucleoside transporters have been cloned
recently and named rENTl and rENT2 (for rat equilibrative nucleoside transporter 1
and 2) (Yao et al., 1997). The subtype rENTl, which is sensitive to NBMPR
inhibition, is widely distributed in cells and tissues and serves as a target for coronary
vascular drugs such as dipyridamole and dilazep (Yao et al, 1997). The subtype
rENT2, which is resistant to NBMPR inhibition, is less broadly distributed in tissues.
Adenosine flux across the membrane follows the concentration gradient, so
changes in the intracellular concentration strongly affect the extracellular
concentration. Although nucleoside transporters are more obscure as important
mediators in the release of adenosine into the extracellular space, very little is known
about the regulation of these nucleoside transporters during relevant
pathophysiological situations discussed earlier, i.e. hypoxia.
The function of nucleoside transporters in drug efflux has not been evaluated.
Largely, there exists a high imbalance of available information when it comes to
evaluating the exit of mucosally/apically absorbed nucleoside analogs from epithelial
cells into the circulation. This is a mechanistic step where primarily the efflux
characteristics transporters are of interest. Current model depictions usually omit the
serosal/basolateral exit rate from estimations of absorption and/or bioavailability of
various therapeutics. In more recent and limited cases, where the exit step of apically
absorbed drugs is given larger attention, the models usually assume similar properties
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to apical absorption for basolateral exit mainly due to lack of relevant data. Figure 1-4
gives a gross depiction of the imbalance in knowledge that exists when comparing
active routes of drug absorption in the intestine.
Intestinal EpHherial Cell
Brush-Border
Membrane
©ssclateral
Membrane
He*/ Ammo Acid
-Amino Add
lauml
O-Fructose
H+ / Lactic Acid
Hi SCFA
HCOaV
Monocarboxylic Acid
H N icctinte Acid
HCOa /
Nicotinic Acid
O tt'/Follc Acid
Choline
Antiporter
Fig. 1-4. Intestinal drug transport mechanisms. Adopted from Tsuji and
Tamai, 1996.
In evaluating the transepithelial absorption of nucleosides, for example, across
airway epithelial cells, it is important to model their exit mechanism across the basal-
lateral membrane after apical entry. Kinetic measurements of unidirectional fluxes of
nucleosides have shown that the flux in the absorptive direction in these cells is
significantly higher than that in the secretory direction. Therefore it is of value to
kinetically characterize the basolateral efflux mechanism, by either measuring the
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
efflux rate of apically pre-loaded radioisotope tagged substrates into basolateral fluid,
or measuring the same rate via metabolic pre-labeling followed by chromatographic
analysis. Figure 1-5 demonstrates the effectiveness of ENT mediated efflux of drugs,
a possible model of drug absorption examined in the tracheal upper-airway.
Drug administered by
pulmonary route
Mucous layer (lumen)
Na+ Drug
Apical
Basolateral
CNT
Uptake
Drug
ENT
Efflux
Fig. 1-5. Role of ENTs in drug efflux in epithelial systems.
1.1.1. Salvage pathway
Recycling of purine or pyrimidine bases involves degradation of nucleic acids that can
occur intracellularly (through the turnover of unstable messenger RNA species or through
DNA repair pathways), as a result of cell death, or through digestion of ingested nucleic
acids in the diet. In animals, the extracellular hydrolysis of ingested nucleic acids
represents the major route by which bases and nucleosides become available. Cleavage
processes begin at internal linkages - in this case, phosphodiester bonds. Catalysis occurs
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
via endonucleases, such as pancreatic ribonuclease or deoxyribonuclease, which function
to digest nucleic acids in the small intestine. Dietary sources of nucleotides may be
conditionally essential within nutrients. Rapidly growing tissues such as the intestinal
epithelium and lymphoid cells lack significant capacity for de novo synthesis of
nucleotides and require exogenous sources of purine and pyrimidine bases. For example,
dietary purines are not significantly incorporated into hepatic nucleic acids, but
pyrimidines are. Both are taken up by intestinal cells with excess purines converted to
uric acid. Nucleotides are important for normal development, maturation and repair of the
gastrointestinal tract (Uauy et al, 1990; Uauy et al, 1994). Purine destruction in tissues
like the placenta or gastrointestinal tract seem to override any potential economy of
purine salvage, and may represent biochemical adaptation to nucleic acid breakdown
occurring in the context of dietary digestion or extensive programmed cell death.
It is clear that the primary way to reabsorb nucleosides from these sources is
solely through the mediation of nucleoside transporters. The role of nucleoside
transporters in the salvage pathways of purine or pyrimidine nucleobases is rather
unconventional, nevertheless is becoming more accepted. The differential regulation of
nucleoside transporter expression or re-sorting from pre-existing intracellular pools
towards domains facing a source for these molecules has been reported in unique
situations when de novo biosynthesis of these molecules is compromised, or in cases of
starvation. Such evidence directly implicates nucleoside transporters in a role that at least
partially contributes to the salvage of nucleosides from extracellular sources. Moreover,
this type of behavior could be advantageous when trying to prevent the uptake of
exogenous nucleosides to pathophysiological tissues/cells where they are needed for
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
disease growth or proliferation. The toxicity of certain drugs designed to stop the
biosynthesis of nucleosides in preventing the growth of tumors can be enhanced by
inhibitors of nucleoside transport.
In addition to mediating the uptake of natural nucleosides, the nucleoside
transport system has been shown to handle a diverse array of nucleoside analogs that are
being clinically used in cancer therapy including the purine analogues cladribine and
fludarabine and the pyrimidine analogues gemcitabine and cytarabine. The antiviral
nucleoside analogues in the treatment of acquired immunodeficiency syndrome (AIDS)
patients include the purine analogue didanosine and the pyrimidine analogues
lamivudine, zalcitabine, and zidovudine.
1.1.2. Cell to cell signaling
Adenosine, a ubiquitous purine nucleoside, is a physiological regulator of various cellular
activities such as cell growth, differentiation, and cell death (Abbracchio and Bumstock,
1998). It is released into the extracellular environment from metabolically active or
stressed cells and in order to re-enter the cells, it binds to selective G-protein-associated
Ai, A2A , A2B , and A3 membranal receptors (Stiles, 1992; Linden, 2001). Specific surface
receptors for adenosine are found in nearly all cells, and almost every organ system in the
body is regulated by its local release. In the heart, adenosine is known to induce a
cardioprotective effect by regulating electrophysiological properties and to protect
cardiac tissue through an ischemic preconditioning process. In the central nervous
system, it acts as a neuroprotective agent via the suppression of neurotransmitter release
and modulation of dopaminergic motor activity. Furthermore, it affects the immune
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
system by exerting anti-inflammatory activity through the inhibition of cytokine release
and platelet aggregation, induction of erythropoietin production and modulation of
lymphocyte function (Gilbertsen, 1987; Soderback et al, 1991; Bouma et al, 1994; Von
Lubitz et al, 1995).
The release and re-uptake of adenosine into and from the environment where it
acts as an agonist for all the described physiological events is exclusively mediated by
nucleoside transporters. Controlling the activity of these transporters is the most direct
way to modulate the intracellular and extracellular concentrations of nucleosides. The
pharmacological manipulation of nucleoside transporters that are the primary regulators
of adenosine levels in therapeutically relevant bodily compartments, is of great medical
interest. The fast facilitative and equilibrative type of nucleoside transporters has the
kinetic features that are most suitable in regulation of extracellular adenosine activity in
these systems.
a) Level of adenosine in body fluids
Extracellular nucleosides have a signaling role in the nervous system, and the modulation
of their levels is achieved by multiple physiological factors (i.e. degradation or
reabsorption). In mammalian systems, the levels of adenosine range from 0.02 to 0.4 /rM
in plasma and cerebrospinal fluid (CSF). There is no significant change in arterial plasma
levels when multiple variables are considered, i.e. age, species, etc. However, in
pathological situations, such as subjects with ischemic heart disease, higher arterial levels
of adenosine are observed. It has been speculated that in various similar disease cases, an
altered turnover of adenosine concentration detected in plasma is the cause of the
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pathophysiology observed. On the other hand, the CSF level of adenosine is at least 2-3
times lower than levels reported in plasma. Lower levels of adenosine nucleosides in CSF
suggest a major difference in the control of extracellular adenosine in the brain (Sollevi,
1986).
b) Level of adenosine in tissues
Tissue levels of adenosine have been determined in human heart and skeletal muscle
(Sollevi, 1986). The purine nucleotide ATP and its related nucleoside, adenosine,
exert pronounced electrophysiologic, inotropic, lusitropic and metabolic effects in the
mammalian heart. These wide spectrums of effects are the result of direct actions of
these compounds on cardiac myocytes and endothelial cells, mediated by a unique
family of cell surface receptors. In addition, ATP and adenosine can stimulate neural
elements inside and outside the heart and thereby modulate neural control of cardiac
function. Heart smooth muscle cells and vascular cells readily form, transport, and
metabolize the endogenous adenosine molecules, and act mechanistically to regulate
both interstitial and plasma adenosine concentrations. Cardiovascular cells also have
membrane adenosine receptors. Cell and tissue distributions, signal transduction
pathways, and pharmacology of each of the four subtypes of adenosine receptors are
subjects of intense investigation. The role of nucleoside transporter in the regulation
of the signal mediated by these receptors is primary.
Nucleosides and nucleotides have several important actions in the lung
(Fozard and Hannon, 1999), among which are contractile and relaxant effects on
airway smooth muscle, stimulation of mucus and surfactant secretion and stimulation
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of ciliary-beating activity. Nucleosides and nucleotides activate or modulate a number
of inflammatory cells that are involved in lung disease, including mast cells
(Marquardt et al, 1994), neutrophils (Cronstein, 1994), macrophages (Hasko et al,
1996), and eosinophils (Knight et al., 1997). Despite these critical roles, the
contribution of nucleoside transporters to the modulation of intracellular or
extracellular levels of substrates is relatively unexplored. Adenosine and adenine
nucleotides have multiple effects as extracellular mediators in every organ system
(Feoktistov and Biaggioni, 1997) and initiate or modulate cellular responses via cell
surface receptors. It has been shown that adenosine can form in both intracellular and
extracellular spaces and that both sources are important in establishing extracellular
concentrations that determine interaction with the receptors (Brundege and
Dunwiddie, 1997).
1.2. Functional history
Equilibrative nucleoside transporters transport nucleosides down their concentration
gradient, whereas concentrative nucleoside transporters couple uphill substrate transport
with downhill sodium transport. The concentrative transporters are further categorized
based on their interactions with endogenous purine and pyrimidine nucleosides; they are
purine-, pyrimidine-, or broadly-selective. At least 7 kinetically distinct nucleoside
transport processes have been characterized from functional studies of fluxes of
radiolabeled nucleosides in various mammalian cells and tissues (Cass et al., 1999; Hyde
et al., 2001; Ritzel et al., 2001b). They have been categorized into two groups on the
basis of their transport mechanisms. Na+ -independent, equilibrative nucleoside transport
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
systems mediate the bi-directional, concentration gradient-driven transport of substrates,
and Na+ -dependent concentrative nucleoside transport systems mediate the uphill flux of
nucleosides coupled to a downhill Na+ gradient that exists across cell membranes
(extracellular to cytoplasmic) as the driving force (Hyde et al, 2001; Ritzel et al., 2001a).
The equilibrative transport systems with bi Odder range of selectivity and
substrates that can permeate through them are ubiquitously found in most cell types.
Whereas, the sodium coupled concentrative nucleoside transport systems with relatively
narrow substrate selectivity are primarily described in specialized epithelia such as
intestine, liver, kidney, and choroid plexus (Hyde et al, 2001; Ritzel et al, 2001a).
Figure 1-6 shows a broad range of endogenous and synthetic nucleosides that have been
reported to interact with nucleoside transporters from Na+ -dependent or Na+ -independent
superfamilies. Their selectivity depends on multiple factors, including both modifications
on the purine or pyrimidine rings, as well as the sugar moiety.
Nucleoside transport mediated by Na+ -dependent and Na+ -independent
transporters is unique in the fact that it is able to move substrates against a concentration
gradient. Uridine at fiM concentrations, similar to other substrates, is able to induce
inward currents in sodium uptake buffers in various CNT-expressing systems (Mackey et
al., 1999; Dresser et al., 2000; Lostao et al., 2000; Ritzel et al., 2001b). The affinity for
sodium in epithelial tracheocytes is 13-15 mM, with an estimated coupling ratio of 1:1.
This stoichiometry allows for the liberation of enough free energy from the downhill
entrance of Na+ ions into the cytosol, to pump nucleosides into an existing high
intracellular pool. Electrogenic nucleoside uptake currents are largely sodium-dependent.
However, very small nucleoside-induced currents have been reported in sodium-free
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
HO,
" C o
OHOH
OHOH
o
mJL Y %
H c X l
W
Adenosine Guanosine
WSvn
Inosine
$
N H L NHU
H° w
OHOH
V f
OHOH
V " T
OHOH
H M ^ V
1 J
Q H
Thymidine
mn
■ 0
o
.
Cytidine 5-Azacytidine 6-Azacytidine
OHOH
Ribavirin
H j^ ) 7 ^ )
HN H N
K
OH
- g
OH
. y
Uridine 2’-deoxyuridine 5-fluoro-2’deoxyuridine 5-chloro-2’deoxyuridine
OH
H 0 ^
O H O H OHOH OH Q H
3-deazauridine 4-thiouridine 5-bromo-2’deoxyuridine 5-iodo-2’deoxyuridine
‘• V r
HN
A
OHOH QH
5-fluoro-5’-deoxyuridine 5-hydroxymethyI-2’-deoxyuridine
x X ™ HocrSr
0
Hjjl
■ itr'
OHOH
6-azauridine
Fig. 1-6. Chemical stractures of endogenous and synthetic nucleosides reported to
interact with nucleoside transporters from Na+ -dependent or Na+ -independent
superfamilies.
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
choline chloride buffers in several studies. Hence the current hypothesis stands that
nucleoside-induced currents most likely mediated by CNTs, are voltage-dependent and
increase several-fold upon hyperpolarization of cell membranes. This suggests that
transport of substrates by CNTs could be voltage-dependent in addition to Na+ -
dependency (Dresser et al, 2000). Table 1-1 summarizes the current understanding of
functional characteristics for known nucleoside transporter systems.
Table 1-1. Functional characteristics of nucleoside transporter subclasses.
Equilibrative Concentrative
Acronym es ei cif cit cib cs csg
Numerical N l N2 N4 N3 N5 N6
Na+ -dependent
—
+ + + + + +
Na+ : nucleoside 1:1 1:1 1:1 2:1 NDa NDa
Inhibited by
NBMPR
■ - - - - - + +
Dipyridamole + + - - - - + +
Dilazep + +
- - - - + +
“ MU, not determined.
The concentrative transport systems are composed of several subtypes based on
their permeant selectivities and NBMPR sensitivity. The cif, cit, and cib transport
systems are insensitive to NBMPR, but can be differentiated by their permeant
selectivities. The cif-type systems (subtype N l) exhibit preference for purines and
uridine. The cz'r-type systems (subtype N2) exhibit preference for pyrimidines and
adenosine. The cib-type (subtype N3) and another czY -type (subtype N4) systems are
broadly selective for both purines and pyrimidines. The cs- and csg-type systems differ
from other concentrative transport processes in their high sensitivity to low
concentrations of NBMPR, however, their substrate selectivities have not been well
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
defined yet. A few of these systems have been discovered and functionally characterized
in human cells. Table 1-2 lists a summary of these transporters.
Table 1-2. Nucleoside transport (NT) processes of human cells1 .
NT Process NT Process Substrate NT Reference
(acronym) (numerical) Specificity Protein
Equilibrative nucleoside transporter
es NA pur & pyr hENTl Griffiths et al., 1997a
ei NA pur & pyr hENT2 Griffiths et al., 1997b
NA NA ND hENT3 Hyde et al., 2001
Concentrative nucleoside transporter
cif N l pur, urd hCNT22 Wang e ta l, 1997;
Ritzel et a l, 1998
cit N2 pyr, ado hCNTl Ritzel et al., 1997
cit N4 pyr, ado, guo ND Gutierrez and Giacomini, 1993
cib N3 pur & pyr hCNT3 Ritzel et al., 2001a;
Ritzel et al., 2001b
cs N5 ado analogs ND Dagnino et al., 1990;
Dagnino et al., 1991
csg N6 guo ND Flanagan and Meckling-Gill, 1997
*NA, not applicable; ND, not determined; pur, purine; pyr, pyrimidine; ado, adenosine; guo, guanosine;
urd, uridine.
2Also known as hSPNT.
Two types (es-type and ei-type) of the equilibrative nucleoside transport processes
have been characterized by sensitivity to classic nucleoside transport inhibitor NBMPR in
considerable detail (Hyde et al., 2001). The es-types (equilibrative, sensitive to NBMPR)
nucleoside transport systems are sensitive to inhibition by nanomolar concentrations of
NBMPR, whereas the ei-type (equilibrative, insensitive to NBMPR) systems are
relatively insensitive to this inhibitor. In human, both types of processes are potently
inhibited by the coronary vasodilator drugs dipyridamole, dilazep, and draflazine,
although the es-type process is relatively more sensitive to these drugs (Hyde et al.,
2001). In contrast, the ei- and es-type transport processes in rat are both insensitive to
these vasodilators (Hyde et al., 2001).
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.3. Our tracheal epithelial cell culture model
1.3.1. Functionally identified nucleoside transport processes
The trachea is suited for the efficient uptake of nucleoside drugs from the blood. For most
drugs, uptake into tracheocytes across the basolateral membrane occurs via passive
diffusion, with minimal reliance on carrier-mediated transport systems. For highly polar
molecules, passive diffusion no longer represents an efficient mode of tracheocellular
uptake and there is an increased reliance on carder-mediated transport systems. For these
compounds, membrane transport may dictate the overall efficiency of hepatic
elimination. An increased understanding of the factors influencing the intracellular
concentrations of nucleoside analogues in the trachea will improve our ability to
specifically treat upper airway disorders, such as viral infections or defects in the
mucociliary clearance apparatus.
We have provided evidence for the functional existence of apical, Na+ -dependent,
concentrative (N2 and N4 types), and Na+ -independent equilibrative (ez'-type) nucleoside
transport processes in the rabbit tracheal epithelial cells (RTEC). Structurally, nucleoside
transporters tolerate nucleoside substrates with modifications on the nucleobase, but
require an intact ribose moiety, and a 3’-OF! group in particular, for recognition. As a
group, these nucleoside transport processes may play a role in the homeostasis of
physiological nucleosides and the selective cellular absorption of nucleoside analogs in
the airway epithelium, thus optimizing the delivery potential of therapeutically active
nucleoside drugs to the respiratory epithelial lining is possible.
Additionally, we have functionally identified and characterized two facilitated,
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Na+ -independent nucleoside transport processes in basolateral membranes of primary
cultured RTEC, linking our preliminary findings regarding apical Na+ -dependent
nucleoside uptake to the hypothesis of vectorial transepithelial absorption for these
molecules. The exit of intracellularly accumulated nucleosides into basolateral fluid of
our in vitro model appears to be the rate-limiting step in transcellular transport.
Therefore, the putative basolateral equilibrative nucleoside transporter(s) in primary
cultured RTEC can be instrumental in governing the accessibility of substrates to
systemic circulation, or local availability in pulmonary therapeutics.
1.3.2. Usefulness of the evaluation of nucleoside d r u g transport
Tracheal tissues that line the airways provide a first line of defense between an organism
and the environment. Trachea is the uppermost segment of the airways that has the most
immediate contact with inhaled materials; hence exploration of transport processes across
this barrier can have immense implications in today’s ever-growing field of inhaled
therapeutics. Drug distribution into the trachea is strictly regulated by the presence of the
epithelial cells that are connected to each other by tight junctions. Therefore, hydrophilic
compounds cannot cross the barrier in the absence of specific mechanisms such as
membrane transporters or endocytosis.
The trachea monolayer is equipped with membrane transport systems to accept
drug molecules and transport them into trachea. The presence of the transport systems by
functional identification and transcript expression includes transporters for amino acids
(Sloan and Mager, 1999), bicarbonate (Parker et al., 2001), organic cations (Tamai et al.,
1997), glucose (Charon et al., 1994), and nucleosides (Ritzel et al., 2001a). Most of these
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
transporters function in the direction of influx from lumen to blood; the presence of
efflux transporters has also been demonstrated, including P-glycoprotein (Abdullah et al,
1997), MRPs (Renaud et al., 1994), and other unknown transporters. These efflux
transporters seem to be functional for detoxification and/or prevention of nonessential
compounds from entering the trachea. Various drugs are likely to be transported out of
the trachea via such efflux transporters, resulting in the decrease of cytotoxicity for drugs
that have pharmacological targets in peripheral tissues or in the reduction of drug efficacy
because of the lower delivery by efflux transport. Identification of the transporters
functionally present in tracheal epithelium and examination of the possible mechanism
involved in drug transport will provide a rational basis for controlling drug distribution to
the trachea.
Electrolyte transport and osmotic water movement in pulmonary epithelium have
been widely studied from trachea, main-stem bronchi, bronchioles to aveoli. A variety of
electrolyte transporters localized to apical or basolateral membranes operate in different
combinations to produce absorption or secretion. Na+ absorption occurs in trachea by
apical Na+ channels that share a sensitivity to amiloride, a potent inhibitor of epithelial
Na+ channels. Cl' secretion is regulated by intracellular second messengers, osmolarity,
and voltage. Electroneutral carriers, such as NaCl(K) cotransport, C 17H C C > 3~ exchange,
and Na+ /HC0 3 ' exchange, also play in concert during absorption and secretion in
airways. Abnormal ion transport in airways may cause cystic fibrosis (CF). In CF
patients, the basic defect is due to a reduced CF conductance and an increased Na+
conductance (Liedtke, 1992).
Transport processes present in the mucosal and serosal membranes of tracheal
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
epithelial cells continue to define the physiological role of this tissue. Select subtypes of
transporters designated for shuttling of impermeable molecules across epithelial
membranes have been identified in the trachea. Evidence for nucleoside transporters in
the trachea and their function on a molecular level is becoming readily available.
Drug inhalation to the mammalian lung can be used to provide a high therapeutic
index for agents targeted to specific respiratory diseases, ranging from common bacterial
or viral infections, to pathophysiological defects such as cystic fibrosis. Following local
administration of drugs in the rabbit lung, comparisons of concentrations in plasma and
the appearance of free-drugs in urine have demonstrated a more rapid systemic
absorption and elimination due to greater accessibility to the systemic circulation
(Meisner et al., 1989). Nevertheless, the usage of the airway as a portal of noninvasive
entry for other agents targeted to systemic diseases has emerged to be insignificant.
2. H e te r o g e n e it y o f M a m m a l ia n E q u ilib r a t iv e N u c le o sid e
Tr a n spo r ter s (E N T s)
2.1. Molecular history
Understanding the relationships among mammalian nucleoside transport processes has
been greatly advanced by the recent molecular cloning and functional expression of
cDNAs encoding nucleoside transporter proteins (Ritzel et al., 2001a). Nothing was
known about the nucleoside transporter protein until the first mammalian member
(designated rCNTl) of the CNT family was identified by functional expression screening
of a rat intestinal library in Xenopus oocytes (Huang et al., 1994). The functional
characteristics of the recombinant rCNTl are typical of the ch-mediated (subtype N2)
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
transport processes. Two related cDNAs encoding membrane proteins with dt-type
activities were isolated from human kidney by reverse transcription-polymerase chain
reaction (RT-PCR) amplification strategies (Ritzel et al, 1997; Felipe et al, 1998). They
have been termed hCNTla and hCNTlb with minor differences in amino acid sequence
attributed to genetic polymorphisms and/or RT-PCR errors. hCNTl is 83% identical to
rCNTl in amino acid sequence with a similar predicted topology of 13 transmembrane
domains (TMDs) (Ritzel et al., 1997; Felipe et al, 1998).
The subsequent molecular cloning has isolated a number of cDNAs encoding a
member of the CNT family from rat liver (designated SPNT (Che et al, 1995)) and
intestine (designated rCNT2 (Yao et al., 1996)) responsible for the cij-mediated (subtype
Nl) nucleoside transport processes. The predicted SPNT and rCNT2 proteins are
identical except for two amino acid residues at 419 (glycine substituted by alanine) and
522 (valine substituted by isoleucine) of rCNT2. rCNT2 is 64% identical to rCNTl and
appears to exhibit a broader tissue distribution than rCNTl. A polymorphism was also
observed between two human cz/-type transporters identified by molecular cloning of
cDNAs from kidney (termed hSPNTl (Wang et al, 1997)) and intestine (termed hCNT2
(Ritzel et al, 1998)) using RT-PCR homology cloning strategies. hSPNTl/hCNT2
exhibits 81% and 72% identity to rCNT2 and hCNTl, respectively. Like CNT1
homologues, CNT2 proteins have similar predicted membrane topology.
Very recently, a broadly selective transporter that mediates Na+ -dependent cib-
type uptake of both purines and pyrimidines has been identified from human mammary
gland (designated hCNT3) and mouse liver (designated mCNT3) (Ritzel et al., 2001a;
Ritzel et al, 2001b). hCNT3 is 48% and 47% identical to hCNTl and hCNT2 with 13
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
putative TMDs. Following a search for additional member of the mammalian CNT
family, it has been difficult to identify the proteins responsible for cs and csg activity
with broad permeant selectivity and high sensitivity to NBMPR. This is because of the
low levels of activity in the cell types in which these transport processes were
characterized and because of the coexistence of other nucleoside transport processes that
may share overlapping functional characteristics. The identity of the proteins responsible
for the cit (subtype N4), cs and csg processes thus remains to be determined.
Isolation of a cDNA encoding the first mammalian member (hENTl) of the ENT
family was characterized in human placenta at the molecular level in 1997 (Griffiths et
al., 1997a). The identification of hENTl was rapidly followed by the cloning of cDNAs
encoding a human ei-type transporter. This protein, designated hENT2 (Griffiths et al.,
1997b), was identified by PCR amplification of a human placental cDNA library using
primers based on the sequence of human HNP36, a delayed early response protein with
sequence similarity to hENTl. The rat homologue rENT2 has also been cloned from
jejunum (Yao et al., 1997). Table 1-3 lists significant discoveries in nucleoside
transporters.
Although a large number of plasma-membrane nutrient transport proteins have
been cloned in the last ten years, much remains to be learned about their structure-
function relationships, membrane topology, posttranslational regulation, and
bioenergetics of transport. Nucleoside transporters occupy an important niche within this
framework, in part due to the nature of their substrates - important building blocks for
Table 1-3. Members of the ENT and CNT family of transporters.
2g
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Transporter Species Residues GenBank™
Accession No.
Reference
hENTl Homo sapiens 456 AAC51103 Griffiths et al., 1997a;
Ward et a l, 2000
rENTl Rattus norvegicus 457 AAB 88049 Yao e ta l, 1997;
mENTl.1 Mus musculus 460 AAF76429 Choi et al., 2000
hENT2 Homo sapiens 456 AF029358 Griffiths et al., 1997b;
Ward et al., 2000
rENT2 Rattus norvegicus 456 AAB88050 Yao etal., 1997
mENT2 Mus musculus 456 AAF76431 Kiss et al., 2000
hENT3 Homo sapiens 475 AF326987 Hyde et al., 2001
mENT3 Mus musculus 475 AF326986 Hyde et al., 2001
hCNTl Homo sapiens 650 U62968 Ritzel et al., 1997
rCNTl Rattus norvegicus 648 U10279 Huang et al., 1994
pkCNTl Sus scrofa 647 AF006973 Pajor, 1998
hCNT2 Homo sapiens 658 AF036109 Wang eta l., 1997;
Ritzel et a l, 1998
rCNT2 Rattus norvegicus 659 U25055 Che eta l., 1995;
Yao etal., 1996
mCNT2 Mus musculus 660 AF079853 Patel et a l, 2000
hCNT3 Homo sapiens 691 AF305210 Ritzel et al., 2001a;
Ritzel et al., 2001b
mCNT3 Mus musculus 703 AF305211 Ritzel et al., 2001a;
Ritzel et al., 2001b
cell function, regulation, growth, and differentiation. Moreover, their substrate selectivity
profiles make them attractive candidates for drug delivery and targeting of a large class of
therapeutics.
2.2. Genomic information
The genomic structure of the hENTl and mENTl genes are located on chromosome 6
(Coe et ah, 1997) and 17 (Choi et al., 2000), respectively. The genes for the sodium-
dependent, concentrative nucleoside transporter, hCNTl (Ritzel et ah, 1997) and hCNT2
(Ritzel et ah, 1998), are mapped to chromosome 15 whereas the newly cloned hCNT3
gene, which contains an upstream phorbol myristate acetate (PMA) response element, is
mapped to chromosome 9 (Ritzel et ah, 2001b). Despite their similar substrates, hENTs
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and hCNTs appear to be members of evolutionarily distinct families since the genes are
located in disparate areas and there is no apparent homology between the cDNA
sequences.
2.3. Molecular characteristics
The encoded 456-residue protein, hENTl, proved to be the first example of a novel group
of transporters that have been designated the ENT family (Baldwin et al., 1999; Cass et
al., 1999). The rat homologue of hENTl, rENTl, a 457-residue protein, is 78% identical
in sequence to hENTl (Yao et al., 1997). Investigation of ENT 1 mRNA and protein
levels in rat, mouse and human tissues has shown that this transporter isoform is very
widely expressed, but is particularly abundant in erythrocytes, placenta, liver, heart, lung,
spleen, testis, kidney, colon and brain (Griffiths et al., 1997a; Choi et al., 2000).
The identification of hENTl was rapidly followed by the cloning of cDNAs
encoding a human ei-type transporter. This protein, designated hENT2, was identified by
PCR amplification of a human placental cDNA library using primers based on the
sequence of human HNP36, a putative protein with sequence similarity to hENTl
(Griffiths et al., 1997b). The same cDNA was independently cloned from a HeLa cell
cDNA library by functional complementation of nucleoside transport deficiency in a
subline of CEM human leukemia cells (Crawford et al., 1998). The rat homologue rENT2
has also been cloned from jejunum (Yao et al., 1997). The 456 residue proteins encoded
by the ENT2 cDNAs are -50% identical in sequence to their ENT1 counterparts.
Investigation of ENT2 mRNA in rat and human tissues indicates that, like ENT1, this
transporter isoform is expressed in a range of tissues including brain, heart, placenta,
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
thymus, pancreas, prostate and kidney, and is particularly abundant in skeletal muscle
(Griffiths et al., 1997b; Yao et al., 1997).
Furthermore, identification of a 419-bp 5’EST sequence (GenBank™ Accession
No. AW226773) from a mouse kidney cDNA clone (I.M.A.G.E. Consortium Clone ID
2285938) was homologous to, but differed from, published ENT1 and ENT2 sequences
(Hyde et al., 2001). DNA sequencing of the clone, revealed that it contained a 1425-bp
open reading frame, flanked by 59-bp and 618-bp 5’and 3’untranslated regions
(GenBank™ Accession No. AF326986). The 474-residue protein encoded by the clone,
which we have designated mENT3, exhibited between 30-33% sequence identity to the
mouse, rat and human ENT1 and ENT2 transporter isoforms (Hyde et al., 2001). The
putative 475 residue protein encoded by this cDNA (GenBank™ Accession No.
AF326987), which has been designated hENT3, was 73% identical in sequence to
mENT3 and 31-33% identical to the mouse, rat and human ENT1 and ENT2 transporter
isoforms (Hyde et al., 2001). Examination of the EST database revealed that ENT3 is
expressed in a range of mouse and human tissues in addition to kidney and placenta,
including adult breast, colon, testis, fetal liver and spleen, and in a number of neoplastic
tissues (Hyde et al., 2001). Their possession of characteristic sequence motifs indicates
that these proteins are genuine members of the ENT family.
Each of the mammalian ENTs is predicted by computer algorithms, such as the
hidden Markov model procedure (Sonnhammer et al., 1998), to possess 11
transmembrane helices, with cytoplasmic N-terminal and central hydrophilic loop regions
and an extracellular C-terminus (Fig. 1-7). Several lines of experimental evidence support
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
nh2
Fig. 1-7. Topological model of mammalian equilibrative nucleoside transporters,
this topology. For example, both hENTl and hENT2 are known to be glycoproteins that
bear N-linked oligosaccharides (Kwong et al., 1993; Ward and Tse, 1999).
The most striking difference of ENT3 from the other mammalian members of the
family is the presence of a long (51 residues), hydrophilic N-terminal region preceding
TMD1. In both mENT3 and hENT3, this region possesses two pairs of adjacent leucine
residues, the first of which is preceded by acidic residues at the 2’ and 4’ positions. Such
sequences resemble motifs involved in the sorting of membrane proteins at the plasma
membrane, endosomes and trans-Golgi network (Sandoval et al., 1994). The N-terminal
hydrophilic region of the yeast ENT FUN26, which is known to be an intracellular
transporter, is also extensive (75 residues) and rich in acidic residues (Vickers et al.,
2000). It is, therefore, possible that the ENT3 proteins reside not at the cell surface but in
some intracellular compartment. The authors are currently raising antibodies against them
in order to investigate this possibility.
2.4. Functional characteristics
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.4.1. Kinetics
Kinetic analysis has confirmed that basolateral membrane preparations consists of two
components that comprise overall uptake of nucleosides: carrier-mediated and passive
diffusion. Overall, more than 90% of uptake is dominated by the carrier-mediated,
equilibrative, and bidirectional component. The affinity of equilibrative nucleoside
transporters for permeants is at least 2-fold lower than that of transporters from the CNT
family (Km of >100 pM is consistent with affinities reported for sodium-independent
uridine transport processes in mammalian cells (Cass et al., 1999)). Kinetic parameters
calculated for both saturable CNT and ENT types of nucleoside uptake indicate that
concentrative uptake is likely to be a high-affinity, low-capacity process, whereas a low-
affinity, high-capacity transporter may comprise the equilibrative side of the family.
Other facilitative transporters also exhibit a lower affinity towards substrates, e.g.,
facilitated glucose transporters (Takata, 1996) and peptide transporters (Terada et al.,
1999). Similarly, dual transport mechanisms for nucleosides have been demonstrated in
rabbit kidney (Williams et al., 1989), and human intestine (Chandrasena et al, 1997),
where a Na+ -dependent secondary active nucleoside transport process was found on the
mucosal surface, and the permeation across serosal cell membranes was mediated by a
Na+ -independent facilitated transporter. In such polarized epithelial systems, molecules
of nutritional value (such as glucose, peptides as well as nucleosides) must cross
basolateral membranes of absorptive epithelia in longitudinal transport from lumen to
systemic circulation (Takata, 1996; Terada et al 1999). Table 1-4 summarizes the Km
values obtained previously for permeants of h/rENTl (Griffiths et al., 1997a; Yao et al.,
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1997) or h/rENT2 (Griffiths et al., 1997b; Yao et al., 1997) recombinant proteins
expressed in Xenopus oocyte systems.
Table l-4.The Km values for permeants of h/rENTl or h/rENT2 recombinant proteins
expressed in Xenopus oocyte systems.
Transporter Species/Tissue Km (mM) Reference
hENTl
hFNT?
Human/Placenta
Hnman/Plfirwita
0.24 ±0.03
0 90 + ft 03
(Griffiths et al., 1997a)
(Griffiths pt nl 1 Q Q 7h)
rENTl
rENT2
Rat/Jejunum
Rat/Jejunum
0.15 ±0.02
0.30 ± 0.04
(Yao etal, 1997)
(Yao et al., 1997)
2.4.2. Pharmacology
The human and rat ENT1 proteins (hENTl, rENTl) and ENT2 proteins (hENT2, rENT2)
are functionally distinguished by sensitivity to inhibition by NBMPR; ENT2 proteins are
relatively insensitive. They also differ in sensitivity to inhibition by the vasoactive drugs
dipyridamole, dilazep and draflazine (hENTl > hENT2 > rENTl = rENT2). Besides
NBMPR, lidoflazine, mioflazine, soluflazine and R73-335 were investigated to be
selective and potent inhibitors of equilibrative nucleoside transporters in rabbit and rat
erythrocytes (Griffith et al., 1990). With hENTl being more sensitive, there is a 7000-
fold and 71-fold difference in sensitivity to NBMPR (IC5 0, 0.4 ± 0.1 nM versus 2.8 ± 0.3
jiM) and dipyridamole (IC50, 5.0 ± 0.9 nM versus 356 ± 13 nM), respectively.
Radiolabeled NBMPR binds to ENT1 cells with a high affinity in the nM range, and each
ENT1 cell has approximately 34,000 transporters with a turnover number of 46
molecules/sec for uridine.
Compounds within this series represented over a 1000-fold range of affinities
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
for the es/ENTl and ez/ENT2 transporters with subtype selectivities (ENT1/ENT2)
ranging from 370 for R70527 to 0.17 for soluflazine (Hammond, 2000). Five other
analogues were identified, in addition to soluflazine, which had significantly higher
affinity for the ei/ENT2 transporter compared with es/ENTl. Structure activity analysis
of these data identified the requirement of a hydrophobic group connected to a 2-
aminocarbonyl piperazine by a 5-carbon chain for high-affinity interactions with
es/ENTl (Hammond, 2000). This hydrophobic moiety was not as important for ez'/ENT2
affinity and in contrast to es/ENTl, a shorter alkyl chain enhanced binding to ezYENT2
(Hammond, 2000). These draflazine analogues also varied in their differential affinities
for mouse vs. human es/ENTl transporters, and the degree of species discrimination was
strongly dependent on the position of the aminocarbonyl group on the piperazine ring
(Hammond, 2000). This information, combined with structural data derived from
molecular studies with ENT1 and ENT2 recombinant proteins, should guide further
development of subtype-selective inhibitors of the equilibrative nucleoside transporters
(Hammond, 2000). Further studies are required to clearly understand the molecular
mechanisms of pharmacological inhibitor interaction with various isotypes of
equilibrative nucleoside transporters.
2.5. Distribution of ENTs
2.5.1. Tissue and cell localization
Equilibrative nucleoside transporters appear to be ubiquitously distributed among cells
and tissues. By high-stringency Northern blot analysis with a placental hENTl probe,
hENT-like mRNA has been observed in many normal human tissues, including fetal
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
brain, liver, spleen, adult adipose tissue, aortic endothelial cells, brain, breast, colon,
heart, lung, ovary, placenta, prostate and uterus (see TIGR Human Gene Index). It was
also detected in cultured human erythroleukemia (K562) cells, as well as in other human
cancer cell lines (A459, G361, HeLa, HL-60, Molt-4, Raji, SW480) (Boleti et al., 1997).
Choi et al, (2000) reported that 2.6 kb ENT1 mRNA exists in all mouse tissues except
skeletal muscle, with highest levels in liver, heart, testis, spleen, lung, kidney, and brain.
ENT2 was also found to express in a number of tissues, and the level of expression was
variable among tissues. A message of about 2.6 kb was observed in most tissues, but a
larger transcript (4 kb) was also detected in thymus, prostate, heart, brain, lung, skeletal
muscle, and pancreas. The highest level of ENT2 expression was in skeletal muscle, with
the 2.6-kb message predominating (Crawford et al, 1998).
At the cellular level, antibody studies (using anti-human erythrocyte nucleoside
transporter (Barros et al., 1995)) showed that expression of nucleoside transporters was
most abundant in lumenal surfaces of fetal capillaries, small placental vessels and
umbilical vein, but absent in the basal membranes of human syncytiotrophoblast, which
was confirmed by confocal immunofluorescence microscopy. Antibodies specific for
hENTl and hENT2 were produced against fragments of the transporter proteins and used
for immunoblot analysis of enriched membrane fractions prepared from several regions
of the human brain. While hENTl was most prevalent in the frontal and parietal lobes of
the cerebral cortex, thalamus, midbrain and basal ganglia, hENT2 was concentrated in the
cerebellum and brainstem regions, particularly the pons. The apparent reciprocal
distribution of hENTl and hENT2 in human brain suggests that these nucleoside
transporter proteins are produced in distinct regions of the CNS where they function in
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
nucleoside salvage and/or regulation of exogenous adenosine.
Within the brain regions that were investigated, the pattern of hENTl distribution
correlated well with adenosine Ai receptor abundance. The regional co-localization of
hENTl and Ai receptor protein suggests an important role of hENTl-mediated transport
process in the control of neuromodulatory actions mediated by adenosine Ai receptors in
human brain (Jennings et al., 2001). Additionally, in independent studies from northern
blot analysis results showed that hENTl mRNA is widely distributed in adult human
brain. 3 5 S-labeled sense and antisense riboprobes, transcribed from a 153-bp segment of
rENTl, were hybridized to fresh frozen coronal sections from adult rat brain and revealed
widespread rENTl mRNA in pyramidal and striatal neurons. Furthermore, regional
localization in rat brain was confirmed by RT-PCR (Anderson et al., 1999b). These
findings detected rENT2 transcript in several brain regions including hippocampus,
cortex, striatum and cerebellum. Results indicate a wide cellular and regional distribution
of ENT2 in rat brain, very similar to ENT1 patterns (Anderson et al., 1999a).
2.5.2. Subcellular localization
Mitochondria, lysosomes and nuclear envelopes are known to have nucleoside transport
activities. Mitochondria isolated from rat liver exhibited 2’-deoxyguanosine transport
which was not blocked by NBMPR (Watkins and Lewis, 1987), whereas mitochondria
isolated from rat testis possess high-affinity binding of NBMPR (Camins et al., 1996).
These results suggested the existence of a nucleoside-selective transport system in the
mitochondria responsible for the movement of nucleosides between the cytosol and the
inner mitochondrial matrix that contains enzymes for nucleoside and nucleotide
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
metabolism.
The lysosomal nucleoside transport process in human fibroblasts exhibited broad
permeant selectivity and NBMPR sensitivity (Pisoni and Thoene, 1989). It is thought to
predominantly function as an exporter of degraded nucleic acids from lysosomes to
cytoplasm because lysosomes contain enzymatic activities capable of degrading nucleic
acids to their constituent nucleosides.
Further evidence for the occurrence of nucleoside transport processes in
intracellular membranes of mammalian cells was demonstrated in cultured human
choriocarcinoma (BeWo) cells using proteoliposome reconstitution (Mani et al., 1998).
The data suggested that the presence of es-mediated (82-85%) and ez-mediated (15-18%)
transport in nuclear envelopes may play an important role in the translocation of
nucleosides between the cytosol and the luminal compartments of nuclear membranes for
DNA and RNA synthesis. hENTl and hENT2 transport activities have been
demonstrated to be present both at cell-surface membranes (Boumah et al., 1992) and
organellar membranes including mitochondria (Camins et al., 1996), nuclear envelopes
(Mani et al., 1998), and lysosomes (Pisoni and Thoene, 1989).
2.6. Structure-function relationships
2.6.1. Changing the binding sites
Sundaram et al. (1998) used the difference in vasodilator sensitivity between hENTl
(vasodilator-sensitive) and rENTl (vasodilator-resistant) as the rationale to construct
chimeras between the two proteins to identify transporter regions responsible for
vasoactive drug binding. It was established that vasoactive drug inhibition involved two
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
domains in the amino-terminal half of hENTl (TMs 1-2 and TMDs 3-6), with TMDs 3-6
being the major site of interaction.
Functional studies suggest that vasodilators and NBMPR compete with substrates
for binding to common or overlapping exofacial sites within the substrate translocation
channel of the transporter (Jarvis et al., 1982; Jarvis et al, 1983; Jarvis and Young, 1986;
Griffith and Jarvis, 1996). However, the general stmctures of vasoactive compounds are
significantly different from the nucleoside derivative NBMPR, and they inhibit
nucleoside transport with lower apparent affinities than NBMPR (e.g., K { for
dipyridamole binding is about 200 nM for recombinant hENTl) (Sundaram et al., 1998).
More significantly, a number of studies have reported that dipyridamole inhibits
dissociation of bound NBMPR, a finding inconsistent with simple single-site models of
ligand binding (Griffith and Jarvis, 1996). Employing similar chimeric strategies to
advance the understanding of vasodilator interactions with novel ENT molecules from
other systems is a valid and reliable method of investigation of transporter interactions
with inhibitors that may bare therapeutic leads.
In addition, the rat transporters, rENTl and rENT2, were employed to determine
NBMPR binding domains free from any structural elements involved in vasodilator
interactions, both of which are resistant to inhibition by vasoactive drugs (Yao et al.,
1997). The major protein sequence differences between rENTl and rENT2 lie in the large
extracellular loop between TMDs 1 and 2, and within the putative central cytoplasmic
loop between TMDs 6 and 7. In addition, the extracellular loop of rENT2 between TMDs
1 and 2 contains 14 fewer amino acids than rENTl, while its central cytoplasmic loop
linking TMDs 6 and 7 has 14 more residues than rENTl. It might be anticipated,
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
therefore, that these loops could be involved in the different NBMPR sensitivities
displayed by the two transporters.
Marked difference in NBMPR sensitivity between the two transporters (Yao et
al., 1997), allows for a series of rENTl/rENT2 chimeric molecules to be generated to
identify regions of rENTl responsible for NBMPR recognition. Results suggest instead
that the structural requirements for NBMPR sensitivity reside within TMDs 3-6 (residues
100-231 of rENTl), the same region previously implicated in vasodilator binding
(Sundaram et al., 1998). The consistency between these two types of molecules binding
to topologically identical regions on ENTs is very interesting, in light of the marked
differences in chemical structures of the molecules at hand. The amino terminus up to the
end of TMD 2 (including the large extracellular loop) also contributes to vasodilator
sensitivity (Sundaram et al., 1998), but, plays no apparent role in the inhibitory action of
NBMPR.
Reported results provide structural evidence that NBMPR and vasodilators
occupy overlapping, but not identical, sites on equilibrative nucleoside transporters.
Analogous findings are consistent with a large body of previous kinetic and ligand
binding studies of native es-type transporters, but are at odds with reports that
dipyridamole inhibits dissociation of bound NBMPR (Griffith and Jarvis, 1996). In their
review of this topic, Griffith and Jarvis (Griffith and Jarvis, 1996) proposed that the latter
effect results from binding of dipyridamole to a secondary low-affinity allosteric site on
the transporter.
Finally, certain findings provide a structural basis to interpret previous NBMPR
photoaffinity labeling studies of the purified human erythrocyte es (hENTl) nucleoside
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
transporter (Kwong et al., 1993). These experiments used enzymatic cleavage to establish
that the site of covalent attachment of NBMPR was within 16 kDa of the site of N-linked
glycosylation (Kwong et al., 1993), now identified as Asn4 8 in the loop between TMDs 1
and 2 (Vickers et al., 2000). Usage of endoprotease cleavage as a tool for identification of
putative inhibitor binding domains is a very useful method of molecular, biochemical
characterization of transporter proteins.
2.6.2. Inhibitors and reactive chemical modifiers
Pretreatment with the impermeable thiol-reactive reagent p-chloromercuriphenyl
sulphonate (PCMBS) caused reversible inhibition of rENT2-mediated cellular uridine
uptake, but had no effect on hENTl, hENT2 or rENTl (Yao et al., 2001b). This
difference correlated with the presence of a unique Cys1 4 0 residue in rENT2 that
was absent in the other ENTs (Yao et al., 2001b). This experimental observation was
further investigated by the technique of site-directed mutagenesis, with the mutated
transporter expressed in Xenopus oocytes (Yao et al., 2001b). Mutation of Cys1 4 0 to
serine produced a functional protein (rENT2/C140S), but was insensitive to inhibition by
PCMBS. Thus, Yao et al, (2001b) identified Cys1 4 0 as the exofacial cystein residue in
rENT2 responsible for sensitivity to inhibition by extracellular PCMBS. Moreover,
uridine protected wild-type rENT2 against PCMBS inhibition, suggesting that Cys1 4 0 in
TMD4 lies within, or is closely adjacent to the substrate-translocation channel of the
transporter. TMD4 has been shown previously to be within a structural domain (TMDs 3-
6) responsible for interactions with NBMPR, vasoactive drugs and nucleobases (Yao et
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
al., 2001a).
2.6.3. Substrate specificity
Na+ -independent ENTs exhibit a broader substrate specificity for purines, pyrimidines,
and a wide range of analogous xenobiotics. Specifically, the ei-type transporters have
also been shown to transport nucleobases such as hypoxanthine (Hyde et al., 2001).
Hence, ENTs are thought to be prospective drug targets for a variety of antitumor and
antiviral nucleoside analogs due to their relatively broad substrate specificity (Pastor-
Anglada et al., 1998).
The need for carrier mediated nucleoside analogue transport can be seen by
evaluation of the vast range of drugs that fit into this category, and have limited
activity due to difficulties in entering target cells. The other aspect that further stresses
the importance of understanding carrier mediated nucleobase and analogue uptake is the
requirement for multiple, usually very large doses of therapeutics in order to achieve the
desired endpoints. For example, current understanding explains that popular antiviral
drug zidovudine appears to cross membranes, like the human placenta, by simple
diffusion (Dancis et al., 1993). The antiviral agents ganciclovir and acyclovir are
transferred across the placenta by simple diffusion as well, at least at therapeutic levels
(Gilstrap et al., 1994) that are considered significant.
Although equilibrative nucleoside transporters are broadly selective, hENT2 is a
generally low affinity nucleoside transporter with 2.6-, 2.8-, 7. 7-, and 19.3-fold lower
affinity than hENTl for thymidine, adenosine, cytidine, and guanosine, respectively
(Ward et al., 2000). In contrast, the affinity of hENT2 for inosine is 4-fold higher than
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
hENTl (Ward et al., 2000). The nucleobase hypoxanthine inhibits radiolabeled uridine
uptake by KENT2, but has minimal effect on hENTl (Ward et al., 2000). Taken together,
these results suggest that hENT2 might be important in transporting adenosine and its
metabolites (inosine and hypoxanthine) in tissues such as skeletal muscle where ENT2 is
predominantly expressed (Ward et al., 2000). These subtle differences between multiple
isoforms of ENTs need to be clearly characterized, along with their biological
distributions, in order to provide clear strategies for targeting selective nucleoside
analogue drugs to specific tissues and organs.
2.6.4. Site-directed mutagenesis
Using site-directed mutagenesis combined with oocyte expression and endoglycosidase
digestion, Asn4 8 was identified as the single site of glycosylation in hENTl (Vickers et
al., 2000). This residue lies in the predicted extracellular loop connecting TMDs 1 and 2,
which also contains potential sites of N-linked glycosylation in the ENT2 and ENT3
isoforms. Using a functional, a-glyco mutant of hENTl (Asn — > Gin) as a template,
glycosylation sites were introduced into the C-terminus or into each of the predicted
exofacial loops of hENTl, confirming the predicted topology (Sundaram et al, 2001a;
Sundaram et al., 2001b). Moreover, Vickers et al. (1999) reported that the conservative
conversion of a specific Asn residue into Gin at position 48 of hENTl and/or the loss of
N-linked glycosylation capability could significantly alter the binding characteristics of
this specific subtype transporter for NBMPR, dilazep and dipyridamole.
Besides investigating the possibility of specific amino acid residues being
involved in substrate or inhibitor binding, site-directed mutagenesis is a direct method of
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
studying key molecular features of ENTs (as well as other transporters). Briefly, as
mentioned before, this approach is required to clearly identify specific locations on ENTs
predicted to be involved in post-translational modifications, like glycosylation or
phosphorylation.
2.6.5. Key structural requirements
The inhibitory profiles of the nucleosides for the es intestinal transporter (Lum et al.,
2000) are virtually identical, where available, to that of the human erythrocyte es
transporter (Paul et al., 1975). These findings suggest that these two transporters are
functionally similar and there is a lot of conservation within their genes through
evolution. In addition, the results of multiple systematic inhibitory profile studies have
drawn several important conclusions concerning possible structural determinants required
for interaction with the recombinant human intestinal es type of transporters. Firstly, the
ribose moiety is the primary (but not exclusive) critical determinant in maintaining the
inhibitory capacity of a nucleoside. Secondly, the 2’ and the 3’ oxygens of ribose
synergistically impart inhibitory characteristics to the nucleosides. Thirdly, modifications
of the nucleoside base do not affect the capacity of the nucleoside to inhibit uridine
uptake by the es transporter except for those that result in a charged molecule, such as the
modifications in 3-deazauridine and 6-azauridine. To extend these findings a little
further; a few simple deductions can be made based upon the general consistencies
reported. For example, these critical structural determinants predict that gemcitabine
(2 ’ ,2 ’ -difluorodeoxycytidine) should be an excellent permeant of the human es
transporter. Indeed, as predicted, gemcitabine is a high-affinity substrate of the es
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
transporter and this transporter activity is required for manifestation of its toxicity in vitro
(Mackey et al., 1998).
Collectively, these findings on the es transporter are both similar and different
from those obtained for the Na+ -dependent intestinal N1 and N2 transporters (Patil et al.,
2000). To retain high inhibitory capacity, the presence of the 3’ oxygen is critical to the
nucleosides for both the N1 and the N2 transporters. With respect to substitutions on the
base, the N1 transporter is much less tolerant than the N2 transporter. In this respect, the
N2 transporter is more similar to the es transporter described here than is the N1
transporter. It is important to compile such valuable structural data in computer-aided
extrapolations that can help predict the structure-functional-activities of novel nucleoside
analogues, specifically in terms of being permeants for ENTs.
2.7. Modulation
Equilibrative nucleoside transporters are highly regulated in mammalian cells (Table 1-
5). They are not constitutively expressed. Moreover, each cell type shows a particular
transporter pattern, probably designed to fulfill general substrate needs but also tissue-
specific requirements, since a single effector can differentially regulate the same
transporter isoform in each cell type. As discussed elsewhere within the context of this
very issue, these transporters can also carry nucleoside-derived anticancer and antiviral
drugs. The cloning of nucleoside transporters has fostered the analysis of their specific
pharmacological profile and the generation of the appropriate tools, essentially
antibodies, to determine their response to a variety of stimuli. Most of the specific
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 1-5. Effect of different substances and experimental conditions on the expression
and activity of the equilibrative nucleoside transporters.
Substance/
condition
Cell type s Transporter Activit
y
Protei
n
mRN
A
Reference
Cell cycle
progression
Rat hepatoma-derived
FAO cells
rENTl ND ND NE (del Santo et al.,
1998)
Cell cycle
progression
Rat hepatoma-derived
FAO cells
rENT2 ND ND T (del Santo et al.,
1998)
PMA Human B lymphocytes hENTl 4 NE 4 (Soler et al.,
1998)
LPS Human B lymphocytes hENTl 4 NE 4 (Soler et al.,
1998)
TNF-a Human B lymphocytes hENTl 4 NE 4 (Soler et al.,
1998)
Nitric oxide Human B lymphocytes hENTl 4 ND NE (Soler et al.,
2000)
M-CSF Murine bone marrow
macrophages
ENT1 T NE t (Soler et al.,
2001b)
M-CSF Murine bone marrow
macrophages
ENT2 NE NE NE (Soler et al.,
2001b)
IFN-y
Murine bone marrow
macrophages
ENT1 NE ND ND (Soler et al.,
2001b)
IFN-y
Murine bone marrow
macrophages
ENT2 NE ND ND (Soler et al.,
2001b)
IFN-y Proliferating macrophages
by M-CSF
ENT1 4 4 4 (Soler et al.,
2001b)
IFN-y
Proliferating macrophages
by M-CSF
ENT2 NE NE NE (Soler et al.,
2001b)
Forskolin Neuroblastoma cells
(Neuro-2A)
es 4 ND ND (Sen etal., 1999)
PDBu or
A23187
Neuroblastoma cells
(Neuro-2A)
es 4 ND ND (Sen et al., 1999)
PMA T84 intestinal epithelial
cells
es 4 ND ND (Mun et al,
1998)
Forskolin Cultured chromaffin cells es 4 ND ND (Sen etal., 1998)
Carbachol Cultured chromaffin cells es 4
ND ND (Sen et al., 1998)
Cell
differentiation
Human neurobalstoma
cells (LA-N-2)
ei t ND ND (Jones et al,
1994)
Triiodo-L-
thyronine
Cultured chromaffin cells es T ND ND (Fideu and Miras-
Portugal, 1992)
PMA, phorbol 12-myristate 13-acetate; LPS, lipopolysaccharide; TNF-a, tumor necrosis factor-oc; M-CSF,
macrophage colony stimulating factor; IFN-y, interferon-y; PDBu, phorbol 12,13-dibutyrate; ND, not
determined; NE, no effect.
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
modifiers of ENT transporter expression remain to be determined, since the lack of
molecular tools has hindered the study of regulation in cell types co-expressing several
transport systems with overlapping specificities.
2.7.1. Second messengers
Further understanding of the posttranslational modulation of nucleoside transport has
pharmacological advantages to improve the transportability of antitumor and antiviral
nucleoside analogs. Regulation of transporters by second messengers has been reported in
carriers such as the glucose transporter (Braiman et al, 1999) and peptide transporter
(Wenzel et al, 1999), however, their regulatory mechanisms vary greatly. In
undifferentiated Neuro-2A cells, the Na+ -dependent adenosine transport was not
modulated by any of known second messenger (like Ca2 + , cAMP), but the es-type
facilitated diffusion component was inhibited by forskolin and A23187 (Sen et al., 1999).
Moreover, Soler et al. (1998) reported that both Na+ - dependent and Na+ -independent
nucleoside transport systems can be regulated by PKC in human B-lymphocytes.
Remarkably, in neuroblastoma cells, only the Na+ -independent nucleoside transport
component was modulated by protein kinase activation, but not the Na+ -dependent one
(Sen et al., 1999). It is likely that cell-specific modulation by protein kinase activation
results in marked differences in nucleoside transporter activity. Cell types coexpressing
different transporter isoforms can exhibit diverging modulation patterns, probably
evolved to fulfill cell-specific requirements. These signaling pathways, if clearly
understood, can be used as tools for controlled and specific modulation of activity in
distinct nucleoside transporters, to help achieve various therapeutic endpoints.
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.7.2. Hormones
Na+ -dependent uridine uptake is stimulated in isolated rat liver parenchymal cells by
glucagon (Gomez-Angelats et al., 1996). This effect is transient, reaches maximum levels
of stimulation 10 min after hormone addition, and is dose-dependent. Glucagon action
can be mimicked by agents that are able to hyperpolarize the plasma membrane (e.g.
monensin) and by dibutyryl cyclic AMP (Gomez-Angelats et al., 1996). The effects
triggered by glucagon, monensin and dibutyryl cyclic AMP are not additive, suggesting a
common mechanism of action. 8-(4-Chloro-phenylthio) adenosine 3',5'-cyclic
monophosphate (PCT), a cyclic AMP analogue, but also a nucleoside analogue, markedly
stimulates Na+ -dependent uridine uptake in an additive manner to that triggered by
monensin, similarly to the effect described for NBMPR. Considering the roles reported
for nucleosides in liver metabolism, the use of PCT as a cyclic AMP analogue should be
precluded.
Insulin is also about to up-regulate Na+ -dependent uridine uptake by a mechanism
which involves a stable induction of this transport activity at the plasma-membrane level.
This is consistent with a mechanism involving synthesis and insertion of more carriers
into the plasma membrane. Recently characterized hepatic concentrative nucleoside
transporter is under short-term hormonal regulation by glucagon, through mechanisms
that involve membrane hyperpolarization, and under long-term control by insulin.
Reports showing hormonal modulation of the nucleoside transporters are of limited
nature, and further studies are required to clearly characterize such regulatory
mechanisms in applicable physiological situations.
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.7.3. Diet and expression
Recent findings have brought forward further evidence of the nutrient regulation of
nucleoside transporters in epithelial cells (Valdes et al., 2000). The effect of nutrient
availability in vivo on nucleoside transporter expression was analyzed in two epithelia,
liver parenchyma and jejunum. Following 48 hr starvation, rats showed an increase in
CNT1 protein levels in jejunum and a slight decrease in hepatic CNT1. Therefore, CNT1-
like activity in brush border membrane vesicles is much higher in fasted rats than in fed
controls. This may not be due to the endocrine changes associated with starvation, which
can be mimicked by feeding on nucleotide-free diets (Valdes et al., 2000). On the
contrary, this suggests that the modulation of CNT1 expression by nutrient availability
varies according to the tissue, thus supporting the hypothesis of tissue-specific regulation
as a novel approach to nucleoside transporter regulation. If this were true, it would imply
that the liver responds to extracellular nucleot(s)ide availability by changes in CNT
transporter expression, which suggests a pivotal role for this organ. This agrees with the
homeostatic function of the liver, which regulates the amino acid and glucose levels of
the whole body.
2.7.4. Injury and inflammation
Adenosine is a ubiquitous signaling nucleoside that is rapidly generated as a net result of
ATP catabolism that occurs in situations of cellular stress or damage. The expression of
cell surface receptors for adenosine is widely distributed among tissues, and this signaling
pathway has been implicated to exert important physiological effects in the
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
cardiovascular, neurological, renal, and immune systems. Additionally, adenosine has
been implicated to play a role in inflammatory lung diseases such as asthma and chronic
obstructive pulmonary disease. Asthmatics have elevated lung adenosine levels, and
adenosine receptor transcripts are increased in inflamed lungs. Further evidence linking
adenosine signaling to asthma comes from reports demonstrating that asthmatics
challenged with adenosine elicit a pronounced bronchoconstriction, whereas
nonasthmatics show no response to such challenges. Collectively, this evidence suggests
that adenosine signaling plays a role in the regulation of pulmonary inflammation and
physiology (Chunn et al., 2001).
The differential cardioprotective effects of inhibitors of adenosine deamination
but not transport, and NBMPR, a selective blocker of adenosine and inosine transport, are
becoming better understood. Superior myocardial protection is reported with inhibition of
nucleoside transport by NBMPR alone or in combination with inhibition of ADA.
Selective blockade of nucleoside transport by NBMPR is more cardioprotective than
inhibition of ADA alone in attenuating myocardial injury (Abd-Elfattah et al., 1998). The
same research group has previously demonstrated the role of nucleoside transport and
purine release in post-ischemic reperfusion injury in several canine models of ischemia.
Interestingly, since rabbits are known to be deficient of xanthine oxidase, they have
attempted to characterize the benefit of selective blockade of purine release in a rabbit
model of coronary artery damage/injury. Conclusions outline that EHNA/NBMPR
induced site-specific entrapment of adenosine of nucleoside transport in the rabbit heart,
in vivo, is a useful application in cardiac injury situations (Abd-Elfattah et al, 1998).
In different studies, Kobayashi et al. (1998) investigated the mechanisms of
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
adenosine production, release, and uptake during chronic hypoxia in rat PC12 cells.
Chronic hypoxia down-regulates the gene expression of rENTl but not that of
rENT2. Alterations in the levels of mRNA are a result of changes in gene
transcription and/or mRNA stability. It appears that chronic hypoxia differentially
modulates these regulatory mechanisms of rENTl and rENT2 gene expression.
Furthermore, adenosine uptake into PC 12 cells is significantly decreased after chronic
hypoxia. Since adenosine uptake was measured in a short incubation time (15 second
intervals), it was relatively independent of changes in the AK activity or levels of
endogenous adenosine (Parkinson and Geiger, 1996). These results suggest an
inhibition of nucleoside transporter function after chronic hypoxia. This response
mechanism could be an important factor in cell to cell signaling during oxidative
stress situations, like the one described above. The implication of nucleoside
transporters in regulation of adenosine signaling is one of great physiological as well
as pharmacological importance.
It is generally thought that sodium independent nucleoside transporters work
bidirectionally. Therefore, it is important to discuss the discrepancy between enhanced
adenosine release and down-regulation of adenosine transporter during chronic hypoxia.
First, hypoxia-evoked release of adenosine does not occur via rENTl, and only uptake
occurs because chronic hypoxia reduces the expression of rENTl. Second, adenosine is
produced extracellularly by breakdown of nucleotides released independently of the
nucleoside carrier, and its reuptake is prevented. The first mechanism is supported by
previous studies indicating that the nucleoside transporters primarily mediate uptake, not
efflux, of adenosine. It has been shown that inhibitors of nucleoside transporters,
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
dipyridamole and NBMPR, cause an increase in extracellular adenosine under various
conditions, including hypoxia (Newby, 1986; Phillis et al., 1989; Ballarin et al., 1991;
Craig and White, 1993). This is in agreement with their findings that intracellular
production of adenosine is enhanced during chronic hypoxia (down-regulation of AK and
ADA, up-regulation of cytoplasmic 5’-NT, whereas expression of rENTl is inhibited by
hypoxia. The second mechanism might also participate in enhanced release of adenosine
during chronic hypoxia. Several previous articles have reported release of nucleotides
during hypoxia in brain (Fowler, 1993; Wallman-Johansson and Fredholm, 1994). This is
also supported by further findings that the membrane-bound form of 5'-NT is
predominant in PC12 cells and the 5'-NT activity of this form is significantly increased
after chronic hypoxia. As the effect of chronic hypoxia on adenosine uptake was
relatively small in reported studies, the regulatory mechanisms for adenosine metabolism
may be more important in controlling extracellular levels of adenosine during chronic
hypoxia. Nevertheless, with more information becoming available about the regulation of
new isotypes of nucleoside transporters, this type of hypothesis is debatable.
In summary, there is a coordinated regulation of nucleoside biosynthesis and
transport within the greater context of cell to cell signaling during stress situations.
Examples used herein clearly showed alterations in adenosine-producing and metabolic
activities during chronic hypoxia in PC12 cells. For instance, the decrease in enzymatic
activities of ADA and AK, accompanied by an increase in both cytoplasmic and ecto-5'-
NT forms, results in an increased capacity to form adenosine both intra- and
extracellularly. This increased potential to generate adenosine and the decreased capacity
to use adenosine indicate that specific cell types can shift toward an adenosine producer
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
phenotype during chronic hypoxia. Furthermore, the reduced function of nucleoside
transporters during chronic hypoxia is likely to be involved in controlling the adenosine
levels. These alterations in enzymes and nucleoside transporters would expose cells to a
higher extracellular concentration of adenosine. Since adenosine is an important
modulator of cellular function in numerous cells during hypoxia (Kobayashi et al, 1998),
it follows that the enhanced release of adenosine and the autocrine/paracrine feedback
regulation of cellular response may have a significant role in modulating cellular
excitability to chronic hypoxia. Manipulation of extracellular adenosine concentrations
and the effect on cell-surface receptors have been investigated, and promise to have great
therapeutic consequences: cell proliferation (Brown et al., 2000), cardiac vasodilation
(Rongen et al., 1994), lipolysis (Schwartzman and Raz, 1982), neuromodulation
(Bumstock, 1990), and platelet aggregation (Puri and Colman, 1998) are only the current
areas being investigated.
3. T h e r a p e u tic Im po r t a n c e o f E N T s a s D r u g T a r g e ts
3.1. Antiviral drugs
Numerous nucleoside analogs have been clinically used in the treatment of viral
infections. For example, zidovudine (Retrovir®) exhibit potent antiviral activity following
intracellular conversion to its active form, inhibits viral RNA synthesis and replication by
inhibiting the enzyme DNA polymerase (reverse transcriptase) in the treatment of human
immunodeficiency virus (HIV) infection (AIDS). Like zidovudine that must be
phosphorylated to be active, didanosine (Videx®) is used for patients who are intolerant
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to zidovudine therapy or who have had previous zidovudine therapy. Zalcitabine (Hivid®)
or lamivudine (Epivir®) is used in combination with zidovudine in the management of
HTV infection (AIDS). Unlike other antiretrovirals, zalcitabine is actively transported
inside the cell whereas AZT and ddl enter cells via passive diffusion. Vibavirin
(Virazole®), a synthetic nucleoside antiviral agent for oral inhalation therapy, possesses
activity against respiratory syncytial virus (RSV), influenza virus, and HSV. Although
the precise mechanism of action is unknown, vibavirin may act as a competitive inhibitor
of cellular metabolites that act on guanosine and xanthosine.
The role of ENTs in the absorption and kinetics of metabolism of such nucleoside
analogues is characterized in a limited context. Antiviral drugs that appear to fit the
general requirements for being permeants of ENTs, will have a significant
pharmaceutical advantage over ones that rely simply on passive diffusion for absorption
and bioavailability.
3.2. Anticancer drugs
There are also nucleoside analogs exhibiting antineoplastic activity. Cladribine
(Leustatin®) is structurally related to fludarabine (Fludara®), but has a different
mechanism of action. It is phosphorylated to its corresponding nucleotide CdATP, which
accumulates and is incorporated into the DNA of cells such as lymphocytes, leading to
DNA strand breaks, inhibition of DNA synthesis, and cell death. Gemcitabine (Gemzar®,
difluorodeoxycytidine, dFdC) is a deoxycytidine antimetabolite, closely related to
cytarabine (Cytosar®). Although cytarabine and gemcitabine are similar in structure and
54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
metabolic pathway, they differ greatly in spectrum of antitumor activity. While
cytarabine is used almost exclusively for acute leukemias, gemcitabine is active against a
variety of solid tumors.
Many of the nucleoside analogs currently used in clinics can cause toxic
responses that are thought to arise from their lack of selectivity for infected or cancerous
cells in some patients. Furthermore, many nucleoside analogs are poorly available after
administration. Therefore, nucleoside transporters become potential drug targets for the
design and synthesis of new nucleoside analogs with the goal of discovering more
selective drugs with optimal membrane permeability.
3.3. Possible relations of ENTs and purinoceptors
As discussed earlier, extracellular purines (adenosine, adenosine diphosphate (ADP), and
ATP) and pyrimidines (uridine diphosphate (UDP) and uridine triphosphate (UTP)) are
important signaling molecules that mediate diverse biological effects via cell-surface
receptors termed purine receptors. Insight into the physiological roles of extracellular
purines and pyrimidines comes from studies of their biological sources and the stimuli for
their release. In this respect, an important line of research stemmed from studies showing
that adenosine is released from the heart during hypoxia to play an important role in
reactive hyperemia.
The general hypothesis of coupling of purine release to metabolic demands via
local regulation of blood flow has been applied to other tissues and includes the release of
adenine nucleotides, particularly ATP, from skeletal muscle during contraction. Analogs
with greater stability than adenosine are produced by modification of the N6 and C2
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
positions of the adenine ring and the 5'-position of the ribose moiety of adenosine, and
have been used extensively in the characterization of adenosine/Pl receptors. NECA, N-
[2-(4-aminophenyl)ethyl] adenosine (APNEA), and N6-(3-[1 2 5 I]iodo-4-aminobenzyl)-5'-
N-methylcarboxamidoadenosine (1 2 5 I-AB-MECA) do not discriminate between
adenosine receptor subtypes. Agonists with subtype selectivity are detailed on individual
adenosine receptor subtypes and the chemical structures of some of these are illustrated
in Figure 1-8.
Xanthines and xanthine derivatives, including the natural derivatives theophylline
and caffeine, are non-selective adenosine/Pl receptor antagonists. They are not universal
adenosine/Pl receptor antagonists; xanthine-resistant relaxations to adenosine and its
analogs were observed in guinea-pig aorta, rat aorta, guinea-pig trachea, porcine coronary
artery, and guinea-pig taenia cecum. It would be interesting to see if these responses can
be blocked by inhibitors of mast cell degranulation. A number of xanthines and non
xanthines identified as adenosine receptor antagonists with reasonable subtype selectivity
are illustrated in Figure 1-9.
It is unclear whether the agonists or antagonists described above are also
substrates for the endogenous nucleoside transporters in the tissues of interest. Since the
endogenous ligand for these receptors is primarily adenosine, one might expect some
overlap in transport characteristic as well, however further studies are required to clearly
characterize this area. Nevertheless, it is clear from previous reports that manipulations
that increase extracellular adenosine, such as adenosine uptake inhibition, inhibition of
adenosine metabolizing enzymes, or activation of Ai receptors by exogenous Ai agonists
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Non-selective
m 2
6 1 «
1
w ' >r \
i f ) 8
o,
HO OH
adenosine
A -j selective
< x
H O H ^
CPA
NH,
w
NECA
a.
&
■ » * , o j
HO OH
CHA
HO OH
CCPA
.P
CH,— CH
••HI
A2A selective
PIA
NH,
ll-Q ^fctya N H - l * 5 .N jC N ^
C H jC H ^N H C P ^O ^
W
C G S216S0
HO OH
APEC
A3 selective
N H C H a — ^ n
U £ >
CHjNHCOv -f'" ' ' v
w
HO OH
IB-MECA
p - v 5
CH5NHCO / O - v J
r i
CHB-MECA
Fig. 1-8. Chemical structures of adenosine receptor agonists with subtype
selectivity.
NECA, N-ethylcar-boxamidoadenosine; CPA, N 6 -cyclopentyladenosine; CHA,
N 6 -cyclopentyladenosine; CCPA, 2-chloro-CPA; R-PIA, (R)N 6 -
phenylisopropyladenosine; CGS21680, 2-[p-(2-carbonyl-ethyl)-
phenylethylamino]-59-N-ethylcarboxamidoadenosine; APEC, 2-[(2-
aminoethylamino) carbonylethylphenylethylamino]-59-N-
ethylcarboxamidoadenosine; IB-MECA, N 6 -(3-iodo-benzyl)-59-(N-
methylcarbamoyl)adenosine; 2CI-IB-MECA, 2-chloro-N 6 -(3~iodobenzyl)-59-
(N-methylcarbamoyl)adenosine.
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Non-selective
° H
K > -
i
c h 3
SPT
A-j selective
■OCHjCONH(CH2)2NH2
A2 A selective
NHj
CH;
HO
C H : > c s c ZM 241385
A3 selective
CH;
c o 2c h 3
L-249313
NH; C l
CH-
L-268605
MRS 1067
MRS 1097
Fig. 1-9. Chemical structures of adenosine receptor antagonists with subtype
selectivity.
8-SPT, 8-(p-sulfophenyl)theophylline; DPCPX, l,3-dipropyl-8-
cyclopentylxanthine; XAC, xanthine amine congener; CSC, 8-(3-
chlorostyry)caffeine; ZM 241385, 4-(2-[7-amino-2-(2-furyl)]l,2,4-
triazolo[2,3-a][l,3,5]triazin-5-ylamino]ethyl)phenol; L-249313, 6-
carboxymethyl-5,9-dihydro-9-methyl-2-phenyl-[l,2,4]-triazolo[5,l-
a][2,7]naphthyridine; L-268605, 3-(4-methoxyphenyl)-5-amino-7-oxo-
thiazolo[3,2]pyrimidine; MRS 1067, 3,6-dichloro-29-isopropyloxy-49-
methylflavone; MRS 1097, 3,5-diethyl 2-methyl-6-phenyl-4-(trans-2-
phenylvinyl)-1,4(i?, S)-dihydropyridine-3,5-dicarboxylate.
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
are of pharmacological significance. Conversely, antagonism of adenosine receptors
and/or the elevated metabolism of extracellular adenosine will have similar
pharmacological modulation effects.
3.4. Transporter inhibitors
Co-administration of 9-|3-D-arabinofuranosyladenine with the ENT inhibitor,
dipyridamole, blocks ENT-mediated efflux, but not CNT-mediated transport, therefore,
enhancing the intracellular accumulation and cytotoxicity of nucleoside drugs in L1210
mouse leukemia cells that contain both ENTs and CNTs (Dagnino and Paterson, 1990).
In a similar strategy to potentiate the action of lometrexol, an inhibitor of glycinamide
ribonucleotide formyltransferase, dipyridamole is combined with hypoxanthine to relieve
the increased toxicity towards normal cells by inhibition of hypoxanthine salvage (Turner
et al., 1997). Beside the above two applications to chemotherapy, coronary vasodilators,
known to be inhibitors of ENTs, block the uptake of extracellular adenosine by binding to
ENTs and thus prolong the effects of adenosine on cell-surface receptors for its
beneficial, cardio-protective effects. Therefore, extensive studies to explore the important
existence of ENTs have been attempted because (i) equilibrative nucleoside transport
processes are normally found as minor transport components, and frequently present with
the concentrative nucleoside transport processes, (ii) ENT-specific transport inhibitors
would allow an enhanced, intracellular accumulation of nucleoside drugs in cells that also
possess CNTs. Consequently, equilibrative nucleoside transporters present a potential
target for the development of improved therapies by the design of more selective
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
inhibitors. Such molecules can be the key to modulating intra and extra-cellular levels of
nucleosides and analogous therapeutics.
3.5. Clinical disorders of purine metabolism in salvage p a th w a y
Kidney stones and symptoms of gout result from deficiency of the salvage enzyme
hypoxanthine-guanine phosphoribosyltransferase (HGPRT). In de novo purine
biosynthesis pathway, several specific enzymatic defects can also lead to excessive
purine synthesis, such as the elevated level of 5-phosphate-a-D-riosyl-1 -pyrophosphate
(PRPP) synthetase and mutations in PRPP amidotransferase. They may result in
excessive accumulation of uric acid or impaired uric acid excretion, resulting in an
elevated level of uric acid and its urate salts. Allopurinol, an inhibitor of xanthine
oxidase, is used to successfully treat gout patients to diminish uric acid synthesis, thereby
preventing the joint pain and neuropathy associated with accumulation of sodium urate
crystals.
In more severe cases, deficiency of HGPRT may cause Lesch-Nyhan syndrome.
Patients with this inherent metabolic disease display severe mental retardation, behavioral
disorders, neurological malfunction and self-mutilation. A reasonable explanation is that
when HGPRT levels are reduced, the levels of free bases hypoxanthine and guanine
increase, leading to an increase in uric acid. The absence of nucleotides will also result in
increases in de novo synthesis of bases. To date, there is no direct treatment for the
defects in HGPRT (located on the X chromosome) underlying enzymatic deficiency
responsible for Lesch-Nyhan syndrome. Allopurinol is clinically used to alleviate only
acute gouty arthritis but not the devastating neurological symptoms of Lesch-Nyhan
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
syndrome. New therapies are being tested such as the neurotransmitter gapapentin that
alleviated self-injurious behavior in a Lesch-Nyhan patient (McManaman and Tam,
1999) and a precursor of purine synthesis, aminoimidazole carboxamide riboside that was
recently approved by NIH to test the effects on Lesch-Nyhan patients. It is hoped that
patients might be treated by gene therapy, perhaps by introducing the HGPRT gene into
brain cells.
The role of ENTs in the regulation of levels of free nucleobases, like
hypoxanthine and guanine, implicated in these disorders is not known. It will be
interesting to see whether or not possible pharmacological stimulants of ENTs can
potentiate the efflux of accumulated nucleobases from systems afflicted with these
diseases.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
II. STATEMENT OF PROBLEM
N u c le o s id e s : A f a m ily o f h y d r o p h ilic m o le c u le s im p e r m e a b le
ACROSS CELL MEMBRANES.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1. O b j e c t iv e s a n d G o a l s
The objective of this Ph.D. dissertation was to gain critical insights into functional and
molecular characteristics of equilibrative nucleoside transport processes in the basolateral
membrane of the rabbit tracheal epithelial cells. Polarized cellular distribution,
pharmacological manipulation and transcriptional modulation were studied. This enables
the use of these transporters as potential targets for the delivery of nucleoside analogs via
the pulmonary route.
The long-term goal of this dissertation has therapeutic relevance. Investigation of
structure-function relationships of ENTs and their substrates will aid in the design of
drugs targeted specifically to these transporters, providing improved membrane
permeability. Specialized nucleoside transport inhibitors with higher potency and
specificity can also be developed offering a second mechanism of clinical treatment.
2. C e n t r a l H y p o t h e s is
Nucleoside transport processes on the basolateral membrane of rabbit tracheal epithelial
cells are mediated by a facilitative-diffusion type carrier system, playing a significant role
in the overall process of transepithelial transport (Fig. 2-1)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Lumen \^ _ J Blood
Na+
Pyrimidine
Na+
Purine
Fig. 2-1. Central hypothesis.
3. S p e c if ic A im s
3.1. To elucidate functional mechanisms of the exit step governing the vectorial
transport of nucleosides across the rabbit tracheal epithelium.
Using primary cultured RTEC grown on permeable supports as a model, uptake studies
from apical fluid revealed that uridine crosses the apical surface of primary cultured
RTEC monolayers by a Na+ -dependent pathway (Mathias et al., 1996a). Since uridine
itself is relatively hydrophilic and impermeable across lipid bilayers, the presence of
nucleoside carriers functionally expressed in the basolateral domain of epithelial cell
membranes is required to complete transepithelial nucleoside transport from lumen to
systemic circulation. Hence, it was hypothesized that basolateral nucleoside transport
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
comprises the exit mechanism of uridine, playing a rate-limiting role in vectorial
absorption of nucleosides and nucleoside analogs from the upper airway.
To elucidate functional mechanisms of nucleoside transport present in the
basolateral membrane of primary cultured RTEC monolayers, Na+ dependency, kinetic
properties, NBMPR sensitivity and selectivity for natural nucleosides of 3H-uridine
uptake were characterized. Interaction of a wide spectrum of nucleobases and various
nucleoside analogs with the basolateral nucleoside transport processes was investigated
by inhibition studies of 3H-uridine uptake to determine their involvement in the cellular
uptake of therapeutic nucleoside analogs in primary cultured RTEC. Understanding the
functional mechanisms associated with basolateral nucleoside transport would improve
the benefits of, and therapeutic impact on, pulmonary delivery of certain antitumor and
antiviral molecules that are the subjects of intense interest.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Specific Aim #1
Functional mechanism of nucleoside transport
system across basolateral membrane of RTEC
Na+ -dependency
< $ > Saturability
< $ ► NBMPR sensitivity
< $ > Substrate selectivity
< $ > Bidirectionality
Scheme 2-1. Specific aim#l.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.2. To clone functionally relevant ENTs from rabbit trachea and delineate their
functional and molecular contribution.
Functional evidence from the initial aim suggests at least two subtypes of Na+ -
independent ENT to be involved in shuttling of molecules across the basolateral
membrane of the rabbit tracheal epithelium. To dissect multiple mechanisms of this rate-
limiting step involved in vectorial transport of nucleosides, it was clear that elucidation of
molecular mechanisms of individual ENTs was needed.
In efforts to provide molecular evidence for the functionality characterized
equilibrative nucleoside transport processes from primary cultured RTEC, isolation of
RNA, reverse-transcribed to cDNA to amplify genes of interest using degenerate primers
designed from the conserved regions among all the already cloned ENTs (GenBank™
Accession No.: AAC39526, AAB97834, AAC62495, 054698, Q99808, CAA60380, and
JC4195), was performed. Using RT-PCR and rapid amplification of cDNA ends (RACE),
a full-length cDNA encoding a novel equilibrative nucleoside transporter (rbENT2) was
identified, as well as its alternatively spliced variant, rbENT2A. The second phase
involved employing newly available bioinformatics computational tools to make
predictions about rbENT2 and rbENT2A. The primary amino acid sequences of rbENT2
and rbENT2A were deduced from the nucleotide sequences beginning at the ATG start
codon. Their secondary topological structures were predicted by hydropathy analysis
using an available computer program TMPred™. Potential N-glycosylation sites and
putative phosphorylation sites, thought to be crucial for transporter function or regulation,
were analyzed by PPsearch™. To examine whether or not this novel splicing event was
unique to the trachea or that both of these messages coexist in other rabbit tissues, we
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
assessed the tissue distribution pattern of these genes in a variety of rabbit tissues by RT-
PCR and ribonuclease protection assay (RPA).
In order to compare and contrast rbENT2 and its splice variant, rbENT2A, in
terms of functional characteristics and spatial distribution within cells, this aim was
formulated. To establish mammalian heterologous expression systems for production of
recombinant nucleoside transporter proteins, rbENT2 and rbETN2A subcloned within a
mammalian expression vector were transiently transfected into non-polarized transformed
human embryonic kidney (HEK) cells, and in separate studies into polarized Madin
Darby Canine Kidney (MDCK) cells. To elucidate functional mechanisms of nucleoside
transport introduced into the heterologous expression systems, Na+ dependency, kinetic
properties, NBMPR sensitivity and selectivity for supplementary nucleosides in terms of
competition for 3H-uridine uptake were done. Following the development of heterologous
expression systems expressing recombinant transporter proteins encoded by the cloned
tracheal rbENT2/2A cDNAs, additional studies were carried out to define their expected
polarized distribution and intracellular distribution by immunohistochemical and confocal
analyses.
Transporter protein identification requires direct characterization of the
polypeptide molecule as well. Therefore, the computational approach was used to design
a number of enzymatic cleavage assays functioning as probes for molecular structure.
Lys-C endoprotease digestion was applied to probe for the presence and accessibility of
key lysine residues. PNGase F treatment identified the presence of predicted
glycosylation sites on the putative extracellular loops of both proteins. Furthermore,
using recombinant techniques, HA-epitopes against a specific antibody were inserted at
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the N- and C-terminal ends of both rbENT2 and rbENT2A to elucidate their membrane
orientation. Both of the transporters were found to have an intracellular N-terminus and
an extracellular C-terminus.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Specific Aim #2
Molecular identification of
rabbit rbENT2 & rbENT2A
Molecular
characl eristics
Functional
characteristics
Localization
Computer-aided prediction of
< $ > Primary amino acid sequence
< $ ► Putative topology
< $ > Potential post-translational
modification sites
< § > Tissue distribution
Membrane localization
< S > Cellular colocalization
2° Structure
validation
< $ > Na+-dependency
< $ > Saturability
NBMPR sensitivity
< $ > Substrate selectivity
< $ > Membrane orientation
Lys-C digestion
< $ > PNGase F treatment
Scheme 2-2. Specific aim #2.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.3. To gain a better understanding of cellular regulatory mechanisms of cloned
transporter rbENT2 and its spliced variant, rbENT2A.
The modulation of cloned rbENT2 and rbENT2A in their native environment were
evaluated in two dimensions. Possible modulation mechanisms involved in basolateral
nucleoside transport activity was evaluated under the influence of well defined signaling
molecules (i.e. Ca2 + , 3 \5 ’-cAMP, and protein kinases). RNA level studies were carried
out, utilizing semi-quantitative PCR techniques to evaluate nutritional triggers that
modulated the constitutive expression of the equilibrative nucleoside transporter gene.
Basolateral nucleoside transport activity was assessed by H-uridine uptake in
primary cultured RTEC monolayers incubated with pharmacological agents that are
known to act along signaling pathways leading to protein kinase C (PKC) and/or protein
kinase A (PKA) activation. The involvement of microtubules in modulation of basolateral
nucleoside transport was also evaluated. Resultant information would allow for the
understanding of regulatory mechanisms of an exclusive exit mechanism of nucleosides
accumulated within ciliated tracheocytes.
To gain a better understanding of the role of the cloned transporter rbENT2 and
its spliced variant, rbENT2A as modulators of intracellular levels of nucleosides in the
state of nucleoside deprivation may be relevant to nucleoside-derived drug absorption and
efficacy via the pulmonary route. Three antimetabolites (MTC, MPA, and 5-FU) known
to inhibit de novo biosynthesis of nucleosides were applied in a three-day treatment
starting from day-4 cultured rabbit tracheal epithelial cells. In addition, the growth
medium was deprived of key amino acids (Asn, Asp, Glu, Gin, and Gly) required for de
novo biosynthesis. We evaluated the extent of depletion on the intracellular nucleoside
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pool by HPLC following nucleoside deprivation. The effects of nucleoside deprivation on
gene regulation of two rbENT2 isoforms and on basolateral transporter function were
assessed by semiquantitative RT-PCR and by 3H-uridine uptake, respectively. To address
the possible mechanism of induction of rbENT2/2A gene levels in response to nucleoside
deprivation, we examined the effects of inhibitors acting at the transcriptional and
translational levels.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Specific Aim #3
Modulation of rabbit rbENT2 & rbENT2A
Pharmacological
modulation of function
i
cAMP effect
^ Ca2 + effect
Microtubule effect
Transcriptional
modulation of transcripts
i
< $ > mRNA levels
❖ T ransporter function
Scheme 2-3. Specific aim #3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
III. METHODS
T o o l s f o r m e a s u r in g n u c l e o s i d e t r a n s p o r t a n d d e t e c t i n g
INVOLVED TRANSPORTERS.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1. Cell Cultures
1.1. Rabbit tracheal epithelial cell (RTEC) monolayers for functional
characterization
Detailed procedures have already been reported for growing rabbit tracheal epithelial cell
monolayers at an air-interface on a permeable support (Mathias, et al., 1995). Briefly,
trachea was excised following euthanasia of pigmented rabbits with an overdose of
sodium pentobarbital solution (85 mg/kg), and incubated in a protease solution (0 .2 %
bacterial protease type XTV in Ca2 + -free minimum essential medium (S-MEM)) for 60
min at 37°C in a humidified incubator with 5% CO2 / 95% air atmosphere. The mucosal
surface was gently scraped with a sterile surgical scalpel. The detached epithelial cells
were mixed with a (deoxyribonuclease I) DNase I solution (0.5 mg/ml DNase I and 10%
FBS in S-MEM) at 37°C, and centrifuged at 210xg for 10 min at room temperature (same
settings apply for all steps below to pellet cells from suspension). Cells were washed
twice with S-MEM containing 10% fetal bovine serum (FBS) by filtration through a 70
pm cell strainer in between washes. The final cell pellet was resuspended in a PC-1
medium modified with 2% (v/v) PC-1 supplement, 100 units/ml penicillin, 100 fig/ml
streptomycin, 50 pg/ml gentamicin, 1 pg/ml fungizone, and 2 mM L-glutamine. Cells
were seeded at a density of 1.3xl06 cells/cm2 on Clearwells®, pre-coated with a S-MEM
solution containing 30 pg/ml collagen (rat tail type I), 10 pg/ml fibronectin (human), and
10 pg/ml bovine serum albumin (BSA). After 24 hrs in culture, the cells were grown at
an air-interface, exposing them to air on their apical surfaces with basolateral volume of
0.3 ml.
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.2. Cell line cultures for transient transfection
Two cell lines were chosen as heterologous expression systems to determine functionality
of individual cloned nucleoside transporters. These cell lines included one non-polarized
kidney cell line (Human Embryonic Kidney (HEK), ATCC CRL-1573), and one
polarized kidney epithelial cell line (gift from Dr. Wei-Chiang Shen, Madin-Darby
canine kidney (MDCK), high resistance).
1.2.1. Cells grown on cluster plates for substrate uptake
HEK and MDCK cells (type I) were seeded and grown on Lab-Tek® II Chamber Slide™
system (Nalge Nunc International, Naperville, IL) or tissue culture-treated cluster plates
(Falcon, Becton Dickinson Labware, Franklin Lakes, NJ) with low-glucose and high-
glucose Dulbecco’s Modified Eagle Medium (DMEM, Life Technologies, Inc.),
respectively. The medium was supplemented with 10 % FBS, Life Technologies, Inc.),
100 units/ml penicillin and 100 /xg/ml streptomycin solution (life Technologies, Inc.).
The growth medium was changed daily. Cells were seeded at a density of 4.0 x 105
cells/well in 6-well 1 day prior to transfection. Experiments were performed following
transient transfection (see Methods 5.1.2.) for functional characterization.
1.2.2. Cells grown on Clearwells®for asymmetry uptake
MDCK cells were seeded and grown on 12-mm Clearwells® (0.4 /xm; Coming Costar,
Cambridge, MA) as described above (see Methods 1.2.1.). Cells were seeded at a density
of 1x10s cells/cm2 on Clearwell® 1 day prior to transfection. Experiments were
performed following transient transfection (see Methods 5.1.2.) for membrane
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
localization.
2. F u n c t io n a l C h a r a c t e r iz a t io n o f P r im a r y C u lt u r e d RTEC
2.1. Uridine metabolism studies
Primary cultured RTEC were dosed with 10 ftCi/ml 3H-uridine. At predetermined time
points, monolayers exposed to 3H-uridine were taken out, washed, and further treated
with 0.5 ml 15% trichloroacetic acid on ice for 1 hr. After centrifuging at 3,000xg for 10
min at 4°C, the supernatant was collected and neutralized with concentrated NaOH,
followed by evaporation under N2. The concentrated sample at a 5 pi volume was spotted
onto a thin layer chromatography plate (TLC; cellulose coated plastic plates of 0.1 mm
thickness that pre-absorbed F254 fluorescent indicator (EM Science)). TLC plates were
washed with water for 2 hours (Jakobs and Paterson, 1986). After air-drying the plates, 1
cm2 strips were made from each cutout lane and extracted by incubating them overnight
with 0.5 ml solution of 0.02 M Tris/HCl containing 0.7 M MgCl2. With 5 ml Econosafe
added to each sample vial containing the strips, its radioactivity was estimated using a LS
1801 System. For the assessment of uridine present in bathing fluids, the incubation
buffers were pooled together after efflux studies were over. These collected fluids were
analyzed as above without the acid-precipitation step.
2.2.3H-uridine uptake studies
2.2.1. Time course studies
Tracheal epithelial cell monolayers on culture days 3-4 (exhibiting transepithelial
electrical resistance of -1.2 k£2»cm 2 and potential difference of -70 mV (apical
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
negative)), were washed with either bicarbonated Ringer’s solution (BRS contains 116.4
mM NaCl, 5.4 mM KC1, 5.6 mM glucose, 0.8 mM NaH2P0 4 , 0.8 mM MgS0 4 , 1.8 mM
C aC fe^ O ,15 mMHEPES (4-(2-Hydroxyethyl)piperazine-1 -ethanesulfonic acid), 25
mM NaHCOs) or sodium-free Ringer’s solution (SFR solution, where NaCl was replaced
with an equimolar concentration of choline chloride, NaHyPCU with KH2PO4, and
NaHC0 3 with choline bicarbonate salt), and equilibrated at 5% CO2 / 95% air, 37°C, in a
humidified incubation chamber. In studies of uptake from apical fluid, the apical buffer
was removed and replaced with 0.1 ml SFR solution or BRS containing 10 pCi/ml 3H-
uridine. Basolateral fluid contained 0.3 ml of the solutions in uptake measurements from
this compartment. Both studies had an appropriate amount of the corresponding buffer
present in the opposite compartment to dosing for maintaining hydrostatic balance.
Uptake of uridine was terminated at 20, 30,45, 60, 90, and 120 seconds by removal of
dosing solutions and washing of tracheocyte monolayers on Clearwells® consecutively,
by dipping them in ice-cold, fresh SFR solutions contained in three 100 ml beakers.
Washed cell monolayers were cut-out and placed in 0.5 ml of 0.5% Triton X-100
contained in scintillation vials, for 30 min, at room temperature. Following cell lysis, 20
pi samples were taken for protein assay (DC protein kit, Bio-Rad Laboratories, Hercules,
CA) and the rest was mixed with 5 ml Econosafe® for assessment of
accumulated intracellular radioactivity using the Beckman LS 1801 System (Beckman
Instruments, Inc., Irvine, CA).
2.2.2. Effects of uridine concentration
The Michaelis-Menten parameters of uridine uptake, Km and J m a x , were estimated by
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
measuring 1 min 3H-uridine (10 ptCi/ml) uptake at various unlabeled uridine
concentrations (0, 25, 50,100, 200, 400, and 800 juM) in BRS or SFR solution dosed
from the apical or basolateral compartments, using the same procedure as above for
radioactivity assessment.
2.2.3. Inhibition studies with NBMPR
Different concentrations (1 nM, 100 nM, 1 ptM, 10 piM, 100 jttM , 1 mM, and 3.5 mM) of
NBMPR were added to the SFR dosing solution containing 10 /tCi/ml 3H-uridine from
the basolateral side, in an attempt to determine the sensitivity of basolateral uridine
uptake process to NBMPR. Uptake was performed for 1 min, terminated by dipping the
tracheocyte monolayers in three consecutive washes with 100 ml ice-cold, fresh SFR
solutions. Radioactivity in tracheocytes was assessed as above. The dose-response curve
was analyzed with GraphPad Prism® version 3.00 for Windows (GraphPad Software, San
Diego, CA) and IC50 value for NBMPR was estimated.
2.2.4. Substrate selectivity
For these studies, 500 jaM unlabeled purines and pyrimidines were coadministered from
the apical or basolateral side in SFR fluid containing 10 jtiCi/ml 3H-uridine. Basolateral
uridine uptake studies (from 0 to 1 min) were then proceeded as above.
2.3.3 H-uridine efflux studies
Tracheocyte monolayers grown on Clearwells® were washed and allowed to equilibrate
in BRS, at 37°C, in 5% CO2 / 95% air humidified incubator. Cells were loaded with
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
uridine from the apical fluid (0.2 ml) in BRS buffer at a final concentration of 10 juCi/ml
3H-uridine for 10, 30, and 60 min. Basolateral BRS volume was 0.8 ml during loading.
Uridine loading was terminated by aspirating both the apical and basolateral fluids, and
washing both sides of the cell monolayers at 37°C, three consecutive times with ample
BRS. Basolateral efflux of uridine was measured after transferring the washed
monolayers to a new cluster plate, followed by the addition of 0 .2 ml and 0 .8 ml of fresh
SFR solution to the apical and basolateral reservoirs, respectively. Basolateral fluid was
collected at predetermined time points in 2 0 0 p.1 aliquots for assay, and replaced with an
equal volume of fresh SFR solution to keep the reservoir volume constant. Five milliliters
of Econosafe® was added to each sample and the radioactivity assayed with the Beckman
LS 1801 System.
3. T is s u e D is t r ib u t io n
3.1. Northern blot analysis
Freshly isolated tracheal epithelial cells (5xl06 cells/200 p .1 ) were lysed with 1 ml of
TRIzol® and repetitive pipetting. Tracheal RNA was extracted per manufacturer’s
direction. The purity and integrity of the isolated RNA were verified by the ratio of
absorbances observed at 260 and 280 nm and also by 1.5% agrose gel electrophoresis. Up
to 30 jag of total RNA isolated from tracheal cells was fractionated by 1.5% agarose gel
electrophoresis and transferred to Biotrans nylon membranes by capillary action in 20x
SSC. RNA was cross-linked to the membrane by irradiation with UV light. Blots were
prehybridized in RNADectector™ formamide hybridization buffer in the presence of 100
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
jug/ml salmon sperm DNA for 1 hr at 42 °C. Hybridization step was carried out overnight
in the same buffer containing 50 ng/ml of a biotinylated rbENT2 gene-specific probe
made using PCR DNA Biotinylation Kit™ . The blots were then washed twice for 15 min
in 2x SSPE/0.5% sodium dodecyl sulfate (SDS) at room temperature, followed by two
washes for 30 min in 0.2x SSPE/0.5% SDS at 55°C. Using the RNADetector™ Northern
Blotting Kit, visualization of the signal was carried out on BioMax™ RS X-ray films. To
determine the levels of RNA loading, the biotinylated probe was stripped off by washing
in 0.5% SDS for 10 min at 95°C, and then re-probed with a biotinylated glyceraldehyde-
3-phosphate dehydrogenase (GAPDH) probe as described above. Normalization of the
signals was performed with respect to the GAPDH level.
3.2. Reverse transcription-polymerase chain reactions (RT-PCR)
3.2.1. Reverse transcription of cDNA from rabbit tracheal RNA
One microliter of oligo (dT)ig (500 jug/ml) was annealed to 5 /rg/11 /rl of rabbit tracheal
DEPC-treated water by incubation for 50 min at 42°C in 20 [i\ total volume comprised of
4 ju .1 of 5x first strand buffer, 10 mM dithiothreitol, 0.5 mM deoxynucleotides (dNTP)
mix, and 1 fil (200 units) of Superscript™ H The synthesis reaction was inactivated by
heating at 70°C for 15 min. Two units of RNase H were then added to remove RNA
complementary to cDNA and incubated at 37°C for 20 min.
3.2.2. Degenerate polymerase chain reaction (PCR)
A degenerate primer pair was chemically synthesized, corresponding to the highly
conserved peptide sequences SGQGLAG and DWLGRSLT as revealed by multiple
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
sequence alignment (in ExPASy (Expert Protein Analysis System) proteomics server at
the Swiss Institute of Bioinformatics (SIB)) of several existing equilibrative nucleoside
transporter family members including ENT1, ENT2, and HNP36 from human and rat.
The nucleotide sequences of degenerate primers were: sense, 5 ’ -AGYGGCCAGGGC
CTRGCWGG-3’; and antisense, 5’-GTWAGGCTCCGKCCYARCCARTC-3\ where K
represents T+G; R, A+G; W, A+C+G, and Y, C+T. The PCR reaction mixture (50 ju .1 )
contained 5% of the cDNA obtained from reverse transcription, 20 mM Tris-HCl (pH
8.4), 50 mM KC1, 2.5 units of Taq polymerase, 0.5 mM mixed dNTPs, and 20 juM
degenerate primers. The PCR reaction conditions were as follows: 94°C for 4 min, one
cycle; denaturation at 94°C for 1 min, annealing at 65°C for 1 min, and elongation at
65°C for 2 min, 30 cycles; and 72°C for 7 min, one cycle. Thirty microliters of the PCR
reaction mixture were electrophoresed in a 1% agarose gel containing 0.5 pg/ml ethidium
bromide, and visualized by UV light. The DNA band corresponding to the predicted size
(-513 bp) was cut out, and extracted using the QIAquick Gel Extraction kit. The resultant
DNA fragment was ligated into a TOPO™ TA cloning® vector following the
manufacturer’s directions. The ligated cDNA was expanded by transformation of E. coli
DH5a competent cells. More than 50 clones were analyzed by endonuclease (EcoRT)
restriction for selection of insert-containing colonies. The resultant plasmid was
sequenced by infrared fluorescent dye-labeled M13 primers (GeneMed Synthesis Inc.,
South San Francisco, CA).
Oligonucleotides for the simultaneous detection of rbENT2 and rbENT2A
transcripts were designed using the Mac Vector® software (Oxford Molecular group,
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Oxford, United Kingdom). The nucleotide sequences of the ENT primers were: sense, 5’-
CGTGGGCATCGTCCTGTCC-3 ’; and antisense, 5 ’ -GCAGCAGATGGGGTTGA
AG-3’. The PCR reaction mixture (50 pi) contained 5% of the cDNA reaction product
from reverse transcription process, 20 mM Tris-HCl (pH 8.4), 50 mM KC1,0.5 mM
dNTP mixture, and 2.5 units of Taq polymerase in the presence of 20 pM primers. The
PCR reaction conditions were as follows: 94°C for 4 min, one cycle; denaturation at 94°C
for 1 min, annealing at 55°C for 1 min, and elongation at 72°C for 1 min, 27 cycles; and
72°C for 7 min, one cycle. PCR products were separated on a 2% agarose gel containing
0.5 pg/ml ethidium bromide, and visualized by UV light.
3.3. Ribonuclease protection assays (RPA)
A PCR fragment of rbENT2 cDNA (nucleotide positions from 619 to 1016 in rbENT2
(GenBank™ Accession No. AF323951)) was amplified by RT-PCR with a pair of
primers designed according to the sequence upstream and downstream of the alternative
splicing domain. The PCR product was ligated into the pGEM®-T Easy Vector System.
An antisense RNA probe was synthesized from the linearized plasmid (cut by Sphl
located at 5’ end of insert) by in vitro transcription with SP6 RNA polymerase
(MAXIscript™ T7/SP6 kit). The RNA probe was biotin-labeled using the BrightStar™
Psoralen-Biotin nonisotopic labeling kit. The RPA was performed using a RPA III™ kit
according to the manufacturer’s protocol. A rbENT2 cRNA (600 pg) was also included in
the RPA as a control and size indicator. After RNase digestion, the protected RNA
fragments were precipitated, separated on a 5% polyacrylamide/8 M urea gel, blotted
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
onto a BrightStar®-Plus™ positively charged nylon membrane, and immobilized by UV
cross-linking. The protected mRNA bands corresponding to the rbENT2 or rbENT2A
were detected using the BrightStar™ nonisotopic RNA detection system, followed by
visualization of the signal on BioMax™ RS X-ray films.
4 . M o l e c u l a r C l o n in g
Double-stranded cDNAs were reverse transcribed from rabbit tracheal RNA, and then
ligated to the adaptor provided with the Marathon™ cDNA Amplification kit. The
antisense primer (5 ’ -GCAGCAGATGGGGTTGAAGAACTC-3 ’) or the sense primer
(5’-GGCAGCCTGTTTGGGCAGCTGGG-3’) with the adaptor primer API (5’-
CCATCCTAATACGACTCACTATAGGGC-3’), respectively, were used to obtain the
remaining 5’- or 3’-ends of cDNAs. Nested 5’- or 3’-RACE PCR was performed using
the antisense primer (5 ’ -GGCGGGGAAGACCGACAGGGTGA-3 ’) or the sense primer
(5’- CACCCTCTTCCTCAGCGGCCAGG-3’), respectively, with the adaptor primer
AP2 (5’- ACTC ACT AT AGGGCTCG AGCGGC-3 ’). The PCR reaction conditions were
as follows: denaturation at 94°C for 4 min, one cycle; 94°C for 1 min, 3 min elongation at
72 °C, five cycles; 94°C for 1 min, 3 min elongation at 70 °C, five cycles; 94°C for 1
min, 3 min elongation at 6 8 °C, 30 cycles; and 72°C for 7 min, one cycle. The PCR
product was separated by 1 % agarose gel electrophoresis, followed by elution from the
gel for ligation into the TOPO™ TA cloning® vector that was used to transform E. coli
DH5a. Multiple clones were selected and analyzed for the sequence of each RACE
product. The full-length cDNAs for rbENT2 and rbENT2A were generated by primers
84
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
complementary to 5’- and 3’-end sequences of the cDNA obtained from 5’- and 3’-
RACE under the PCR condition of 94°C for 4 min, one cycle; denaturation at 94°C for 1
min, annealing at 60°C for 1 min, elongation at 72°C for 1 min, 30 cycles; and 72°C for 7
min, one cycle. The final amplicon was subcloned into the TOPO™ TA cloning® vector
and sequenced by infrared fluorescent dye-labeled M13 primers.
5. M o l e c u l a r C h a r a c t e r iz a t io n
5.1. Establishment of heterologous expression systems
5.1.1. Construction of DNA plasmids
Full-length cDNAs encoding rbENT2 (GenBank™ Accession No. AF323951) and
rbENT2A (GenBank™ Accession No. AF323952) were subcloned in a sense orientation
into the mammalian expression vector pcDNA3 (Invitrogen), pSG5-HA (pSG5 vector
tagged with N-terminal HA), and pMH (C-terminal HA vector, Roche Molecular
Biochemicals, Indianopolis, IN). pcDNA3 vector (Invitrogen, Carlsbad, CA) was used as
a mammalian expression vector for investigating whether the addition of HA tag would
affect the functional activity of epitope-tagged versions of rbENT2 and rbENT2A
isoforms. A forward primer 5’-GGGGTACCGCCATGGCGCGAGGAGACGCC -3’ and
a reverse primer 5 ’ -CGGAATTCGAGCAGGGCCTTGAAGAG-3 ’ were used to create
cutting sites for two enzymes of Kpnl and EcoRl (shown underlined), respectively, at two
ends of rbENT2 open reading frame. The sequence coding for rbENT2A was amplified
with PCR using the same forward primer while the reverse primer was replaced with 5’-
CGGAATTCTGCGCCAGCACCTGCCTG-3 ’ (EcoRl, shown underlined). The
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
amplified products were digested with Kpnl and EcoRl, and then inserted into the Kpnl-
EcoRl sites in pMH.
C-terminal truncation of rbENT2 was made with a forward primer 5’-
GGGAATTCGCGCGAGGAGACGCCCCG-3’, which was designed to anneal to 5’-end
coding sequence of rbENT2 cDNA with a EcoRl restriction site (shown underlined), and
a reverse primer 5 ’ -CGCTCGAGTCACTTCCGGAAGACCACGAAG-3 ’, which
contained codons that anneal to the sequence corresponding to amino acid residues Val
to Lys288 of rbENT2, with a Xhol restriction site. This EcoRl-Xhol fragment was
subcloned into the EcoRl-Xhol sites in pcDNA3 (Invitrogen™) or pSG5-HA vector
digested with EcoRl and Xhol. All plasmid constructs were verified for correct
orientation by restriction enzyme analysis and sequence analysis. Unless indicated
otherwise, all the reagents in this study were obtained from Invitrogen Co. (Carlsbad,
CA).
5.1.2. Transient transfection
The individual DNA expression plasmid constructed as above (see Methods 5.1.1.) was
transiently transfected into HEK or MDCK cells using LipofectAMIME™ reagent (Life
Technologies, Inc.) according to manufacturer’s instructions. For each transfection, 1 pg
DNA in 0.1 ml Opti-MEM media (Life Technologies, Inc.) was mixed with 0.1 ml Opti-
MEM media containing 6 pg and allowed equilibrating for 15 min at room temperature.
Following equilibration period, 0.8 ml Opti-MEM media was added to dilute the mixture
and then gently applied to the cells of each well. The bathing medium was replaced with
the growth medium after 6 hrs in order to stop transient transfection. In the co-
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
transfection experiments, total amount of DNA in each transfection of cells grown in a 6 -
well plate was kept constant at 2 fig by supplementing with parent vector, pcDNA3. To
normalize for transfection efficiency, cells were harvested in luciferase lysis buffer
(Promega, Madison, WI) 16 hrs thereafter. Firefly luciferase activities were measured
using Luciferase™ assay system (Promega, Madison, WI) according to the
manufacturer's instruction. Light units from firefly luciferase activity were determined
using a luminometer and were normalized to 3H-uridine uptake and to protein
concentration. Each transient transfection experiment was repeated at least three times
independently. The variation of lucifearse activity between wells observed within a
transfection was <6 %, and between transfections was <7%.
5.2. Detection of recombinant protein expression
5.2.1. Western blot analysis
Transfected HEK293 or MDCK cells grown on 60 mm petri dishes were washed once
with lx phosphate-buffered saline (PBS) and harvested by gentle scraping into 0.5 ml of
0.5% SDS in PBS containing 1% protease inhibitor cocktail (Sigma Chemical Co., St.
Louis, MO). Total cellular proteins (30 fig) were mixed with an equal volume of 2x SDS-
PAGE sample buffer (20% (v/v) glycerol, 4% (w/v) SDS, 20 mM Tris-HCl, 0.01% (w/v)
bromophenol blue, 2 % P-mercaptoethanol) and fractionated on a 1 0% (w/v) gel, followed
by transfer to a nitrocellulose membrane (Bio-Rad Laboratories, Hercules, CA) by
electroblotting. Membrane blots were immunostained with mouse anti-HA monoclonal
antibody (Covance, Princeton, NJ) diluted by 1:1000 (v/v), followed by peroxidase-
conjugated AffiniPure donkey anti-mouse IgG (Jackson ImmunoResearch
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Laboratories, Inc., West Grove, PA) at a dilution of 1:50,000 (v/v). The resulting signals
were visualized using enhanced chemiluminescence (ECL) reagent (Amersham Life
Science, Arlington Heights, EL) and were detected on BioMax MS X-ray films (Kodak).
5.2.2. Membrane orientation
MDCK cells were plated onto Lab-Tek® II Chamber Slide™ (4 well, Nalge Nunc
International, Naperville, E L ) at a density of 4.0x104 cells/cm2 one day prior to
transfection. Forty-eight hours following transient transfection with N- or C-terminal
HA-tagged transporter DNA constructs (rbENT2/pSG5.HA, rbENT2A/pSG5.HA,
rbENT2A/pMH, and its parent vector alone), cells were rinsed with lx PBS and fixed
with 3.7% formaldehyde (Sigma) solution for 30 min at room temperature. Fixed cells
were incubated in blocking solution (1% bovine serum albumin (BSA) in PBS) in the
presence (+) or absence (-) of 0.1% Triton X-100 (TX-100) for 1 hr at room temperature.
Permeabilized (P, +TX-100) or non-permeabilized (NP, -TX-100) cells were then
incubated with mouse anti-HA monoclonal antibody at 1:1000 (v/v) in blocking solution
at room temperature. After 1 hr incubation with primary antibody, cells were quickly
rinsed three times, followed by three consecutive washes (5 min each) with blocking
solution. Fluorescein isothiocyanate (FTTC)-conjugated AffiniPure donkey anti-mouse
IgG (Jackson ImmunoResearch Laboratories, Inc.) was added at a dilution of 1:100 (v/v)
for 1 hr at room temperature. Cells were washed again as described above following
incubation with FITC-labeled secondary antibody. ProLong™ antifade reagent
(Molecular Probes) was applied to preparations and sealed with clear coverslips. In order
to ascertain that TX-100 led to permeabilization of cell plasma membranes,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
peraieabilized or non-permeabilized cells were then incubated with mouse anti-y-adaptin
antibody (Transduction Laboratories, Lexington, KY) at 1:50 (v/v) in blocking solution at
room temperature, followed by 1 hr incubation of FITC-labeled secondary antibody. The
slides were examined under a microscope (Nikon Eclipse TE 300) equipped with
MetaMorph® Imaging software (Universal Imaging Corp., West Chester, PA).
5.2.3. Membrane localization
Confocal microscopy of N-terminal HA-tagged rbENT2/rbENT2A and C-terminal
truncated rbENT2 were performed on a confluent monolayer of transfected MDCK cells
cultured on Clearwells®. At 48-hr post-transfection, cell monolayers were rinsed with lx
PBS and fixed with 3.7% formaldehyde solution for 30 min at room temperature. Fixed
cell monolayers were incubated in blocking solution (1% BSA in PBS) in the presence of
0.1% TX-100 for 1 hr at room temperature. Permeabilized cell monolayers were then
incubated with mouse anti-HA monoclonal antibody at 1:1000 (v/v) and 2 /rg/ml rabbit
anti ZO-1 antibody (Zymed Laboratories Inc., South San Francisco, CA) in blocking
solution at room temperature. After 1 hr incubation with primary antibody, cell
monolayers were quickly rinsed three times followed by three 5 min washes with
blocking solution. FITC-conjugated AffiniPure donkey anti-mouse IgG and rhodamine-
conjugated goat anti-rabbit IgG (Jackson ImmunoResearch Laboratories, Inc.) were
added at a dilution of 1:100 (v/v) and 1:500 (v/v), respectively, for 1 hr at room
temperature. Following secondary antibody incubation and washing as above, the cell
monolayers grown on the Clearwell® filters were cut out and placed cell-side down on
T TU I
glass slides. ProLong antifade reagent (Molecular Probes, Eugene, OR) was then
89
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
applied to preparations and sealed with clear coverslips (18 mm2, Coming, Coming, NY).
Using a Zeiss LSM410 confocal fluorescence microscope, serial optical sections were
obtained perpendicular to the monolayer surface (i.e., xz plane) at 0.4-/tm steps.
5.2.4. Cellular colocalization
Visualization of the endoplasmic reticulum (ER) in transiently transfected HEK and
MDCK cells with either rbENT2/pSG5.HA or rbENT2A/pSG5.HA constructs were
performed by labeling with an ER marker carbocyanine dye DiOCe(3) (Molecular
Probes, Eugene, OR) for 30 min with a concentration of 10 fig/mL at 37°C in the dark
prior to fixation. To visualize lysosomes, cells were incubated with LysoTracker Red
DND-99 (Molecular Probes, Inc. Eugene, OR) at a concentration of 60 nM for 2 h at
37°C prior to fixation. After staining with ER or lysosomal markers, cells grown on glass
coverslips were washed with PBS and fixed for 30 min at room temperature with 3.7%
paraformaldehyde in PBS. Cells were then washed with lx PBS, permeabilized for 1 min
at room temperature with 0.5% TX-100 in lx PBS, and washed again with PBS.
Blocking, incubation with primary and secondary antibodies, and mounting were carried
out as described above (see Methods 5.2.2). Propidium iodide (Sigma) was used as a
nuclear marker for reference and contrast. After fixation cells were permeabilized and
treated with PBS containing 100 ptg/ml RNase A (Roche Molecular Biochemicals) for
2 0 min at room temperature prior to staining with propidium iodide at a concentration of
50 /ig/ml for 20 min at room temperature in the dark. Sample preparations were covered
with ProLong™ antifade reagent (Molecular Probes) before sealed by glass coverslips
(Coming Costar, Cambridge, MA). Cells were viewed in xy plane using a Zeiss LSM410
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
confocal fluorescence microscope.
5.3. Functional characterization of transiently transfected cells
5.3.1. Uptake studies in the transfected cells
Transiently transfected HEK293 or MDCK cells (at 24 hr post-transfection) were
detached with 0.05% trypsin-EDTA, replated at 1x10s cm2 onto Multiwell™ 12 well
plates (Falcon, Becton Dickinson Labware, Franklin Lakes, NJ), and allowed to grow for
24 hr prior to uptake studies. 3H-uridine was selected as a substrate for the following
uptake studies. HEK293 or MDCK cells at 48 hr post-transfection were washed once
with sodium-free Ringer’s solution (SFR, containing 116.4 mM choline chloride, 5.4 mM
KC1, 5.6 mM glucose, 0.8 mM KH2PO4, 0.8 mM MgSCL, 1.8 mM GaCL'FLO, 25 mM
choline bicarbonate) and allowed to equilibrate for 20 min. A dosing solution containing
5 fiM 3H-uridine (2 pCi/ml) in SFR replaced the SFR to initiate the uptake of labeled
uridine. Uptake at 5 min was terminated by washing cells in a cluster plate three times
with fresh ice-cold SFR. Washed cells were lysed using 0.5 ml of 0.5% TX-100. Twenty
microliters of cell lysates were taken for protein assay (DC protein assay, Bio-Rad
Laboratories, Hercules, CA). Five milliliters of Econosafe® scintillation cocktail were
added to the rest of each cell lysate sample and assayed for its radioactivity using LS
1801 System (Beckman Instruments, Inc., Irvine, CA).
a) Concentration dependency
Transiently transfected HEK293 or MDCK cells were spiked with 5 fiM 3H-uridine (2
juCi/ml) in the presence of 5,10, 20, 50, 100, 200,400, 600 and 800 [iM of unlabeled
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
uridine in SFR.
b) Effect of NBMPR
Transiently transfected HEK293 or MDCK cells were exposed to different concentrations
of NBMPR ranging from 1 nM to 100 fiM premixed with the dosing solution containing
5 fiM 3H-uridine (2 /iCi/ml) in SFR, to determine the NBMPR sensitivity of uridine
uptake process. The dose-response curve was produced by non-linear regression analysis
using the GraphPad Prism® version 3 for Windows (GraphPad Software, San Diego, CA)
and used to estimate the IC50 value for NBMPR.
c) Substrate selectivity
Unlabeled physiological purines (e.g., adenosine, guanosine, and inosine), pyrimidines
(e.g., cytidine, thymidine, and uridine) and a nucleobase hypoxanthine (500 fiM each)
were present in the dosing solution of 5 fiM 3H-uridine (2 /iCi/ml) in SFR.
5.3.2. Efflux studies of transiently transfected cells
Transiently transfected HEK293 or MECK cells were replated onto 6-well plates at 24 hr
post-transfection and assayed at 72 hr post-transfection as illustrated above. Cells were
washed once with SFR solution and allowed to equilibrate for 20 min, prior to preloading
3 3
with 5 /xCi/ml H-uridine for 15 min. One milliliter fresh sodium-free buffer without H-
uridine was used to briefly wash H-loaded cells three times. Radioactivity remaining in
each 100 fil of the bathing fluid sampled at a predetermined time (0,10, 20, 30,45, 60,
90, and 120 min) was measured. One hundred microliters of the fresh SFR solution was
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
replenished to keep the bathing volume constant following each sampling. Five milliliters
of Econosafe® (Research Products International, Mount Prospect, EL) were added to each
sample and its radioactivity was assayed with LS 1801 System (Beckman Instruments,
Inc., Irvine, CA).
5.4. Biochemical characterization
5.4.1. Lys-C digestion
Transfected MDCK cells grown on Multiwell™ 6 well plates (Falcon, Becton Dickinson
Labware, Franklin Lakes, NJ) were harvested by using Cellstripper™ (Cellgro®,
Mediatech, Inc., Hemdon, VA), a non-enzymatic cell dissociation solution. The
harvested cells were washed three times with lx PBS and then suspended in 100 mM
Tris-HCl pH 8.5 in the presence or absence of 0.1% TX-100. For digestion with
Lysobacter enzymogenes endoproteinase Lys-C (Sigma), it was used at 0.02 /xg/assay (20
/xl), and digestion conditions were established as described in the legends for Figures 4-
32 and 4-33. Each digestion reaction was inactivated by addition of Protease Inhibitor
Cocktail for General Use (Sigma), at a recommended 1 ml/mg proteases, followed by
boiling for 2-3 min. Digests were fractionated on 1.5-mm SDS-PAGE gel (15%).
Fractionated proteins were transferred to nitrocellulose membranes as described above in
Western blot analysis. Some SDS-PAGE gels were stained with GelCode® Blue Stain
Reagent (Pierce).
5.4.2. PNGase F treatment
The transfected MDCK cells were detached with Cellstripper™, and denatured at 1Q0°C
93
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
for 10 min. Peptide:lV-glycosidase F (PNGase F, 1000 units, England Biolabs, Beverly,
MA) treatment (described in the legend for Figure 4-34) was performed at 37 °C for 2 hr.
PNGase F-treated samples were fractionated on 1.5-mm SDS-PAGE gel (8%).
Fractionated proteins were transferred to nitrocellulose membranes as described above in
Western blot analysis. Some SDS-PAGE gels were stained with GelCode® Blue Stain
Reagent (Pierce).
6. M o d u l a t io n
6.1. Pharmacological modulation
To elucidate the possible mechanisms of modulation of basolateral uridine uptake,
pretreatments with A23187 and forskolin for 15-60 min incubation of cell monolayers
were performed in SFR solution containing either 10 /xM tamoxifen (PKC inhibitor), 10
/xM H89 (PKA inhibitor), or 10 /xM colchicine (cell cytoskeletal assembly disruptor).
Stock solutions of lOOx concentration were prepared in DMSO, containing all
pharmacological agents used for the various treatments. Negative controls were treated
with solvent alone at the final dilution concentration for all
samples. Basolateral uridine uptake (1 min) was then studied as described above (see
Methods 2.2.1).
6.2. Transcriptional and post-translational modulation
6.2.1. Procedures for nucleoside deprivation
Primary cultured RTEC cells grown in PC-1 medium for 4 days were exposed to a
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
special medium containing methotrexate (MTX), mycophenolic acid (MPA), and 5-
fluorouracil (5-FU) (all at either 100 pM or 200 fiM) in the custom-made minimal
essential medium (MEM) lacking amino acids, Asn, Asp, Glu, Gin, and Gly. The
treatment continued for three days, and the cells were fed daily with the special medium.
The control RTEC was cultured in PC-1 medium for 7 days with daily feeding.
6.2.2. Determination of Nucleoside Levels by High Performance Liquid
Chromatography (HPLC)
RTEC monolayers on culture day 7 were washed twice with ice-cold PBS and then lysed
with 0.1 % Triton X-100. Trifluoroacetic acid at a final concentration of 2% (v/v) was
added to the cell lysates, followed by centrifugation at lOOOxg for 10 min. The
supernatant was mixed with two volumes of saturated NaHCOs to neutralize the acid. An
equal volume of diethylether (99.9%) was added to the neutralized solution for
extraction. This extraction procedure was repeated two more times. The ether phase was
discarded and the aqueous phase was filtered through a 0 .2 -p.m filter before injection into
the HPLC column to determine the intracellular levels of the nucleoside species. The
HPLC system was comprised of a Hamilton PRP X-100 250 x 4.1 mm anion exchange
column (Hamilton, Reno, NV) and a Beckman System Gold 7 (Beckman, Fullerton, CA)
with both a 126 Solvent Delivery Module and a 110B Scintillation Cocktail Delivery
Module, a 166 UV Programmable Detector module, a 171 Radioisotope Detector
Module, an Applied Biosystems 783 UV-VIS Programmable Detector Module, and a
Waters 717 Plus Autosampler. The mobile phase had a flow rate of 2.5 ml/min and
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
consisted of solutions A (9:1 volumes of 1 M ammonium acetate pH 9 vs. acetonitrile)
and B (9:1 volumes of deionized H2O vs. acetonitrile] with the following schedule: 0-5
min 100% B, 5-18 min 75% B + 25% A, 18-18.5 min 35% B + 65% A, 18.5-25 min
100% B.
6.2.3. Semi-quantitative RT-PCR
Total cellular RNA of cultured RTEC cells on day 7 after treatment from day 4 through 7
with a scheme to reduce the intracellular nucleoside levels (see Methods 6.2.1.) was
extracted with TRIzol following the manufacturer’s instructions. Oligonucleotides for
simultaneously detecting two rabbit ENT transcripts were designed using the Mac
Vector® software (Oxford Molecular group, Oxford, United Kingdom). The nucleotide
sequences of the primers for ENT were: sense 5’-CGT GGGCATCGTCCTGTCC-3’ and
antisense 5 ’ -GCAGCAGATGGGGTTGAAG-3 ’ .As control, the expression level of
GAPDH mRNA was analyzed using primers in sense direction, 5 ’ -GCCAAAAGGG
TCATCATCTC-3 ’ and antisense direction, 5 ’ -CTCAGTGTAGCCCAGGATGC-3 ’. The
first cDNA strand was obtained by RT reaction, then the subsequent amplification of
rbENT2/2A mRNA isoforms present in RTEC was done by PCR with these designed
primers for ENT using a Superscript™ One-Step RT-PCR Systems. The following
schedules were used: RT, 45 min at 48 C; denaturation, 2 min at 94°C; and PCR: 94°C
for 4 min, one cycle; denaturation at 94°C for 1 min, annealing at 55°C for 1 min, and
elongation at 72°C for 1 min, 27 cycles; and 72°C for 7 min, one cycle. Using the
schedule, PCR fragments of rbENT2 cDNA (corresponding to nucleotides from 614 to
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1016 in rbENT2 (GenBank™ Accession No. AF323951)) and rbENT2A cDNA
TTWT
(corresponding nucleotides from 614 to 976 in rbENT2A (GenBank Accession No.
AF323952)) were amplified semi-quantitatively to allow comparisons levels between the
expression levels of rbENT2/2A and GAPDH transcripts. Under these conditions, the
efficiency of the RT-PCR reaction for each gene was below the plateau (i.e., within the
exponential phase of the amplification), allowing us to compare mRNA expression levels
meaningfully. Total tracheal RNA (1.0 pg) was used as template. For GAPDH, the same
amount of total RNA as described before was used, but with 23 PCR cycles only. The
general semi-quantitative RT-PCR protocol was as follows: RT-PCR reactions were
performed by first preparing a master RT-PCR mix containing 2 pg of total RNA from
RTEC and aliquoted in equal parts into two separate tubes containing 100 pmol of ENT
primers and 50 pmol of GAPDH primers, respectively. Negative RT-PCR controls were
done in the absence of RNA or reverse transcriptase. After RT-PCR procedure is over,
the content of each independent reaction tube was mixed with a DNA gel sample buffer
and separated on a 2% agarose gel, followed by staining with 0.5 pg/ml ethidium
bromide and visualization by UV light. Ethidium bromide-stained gels were scanned
and analyzed with Scion Image Beta 4.02 (Scion Corporation, Frederick, MD). All
images from the ethidium bromide-stained gels were inversed for clearer depiction of
data.
6.2.4. Evaluation of basolateral uridine uptake in nucleoside-deprived cells
Tracheal epithelial cell monolayers grown on 12 mm diameter Clearwells® on culture day
4 in culture (exhibiting PD of -70 mV and Rt o f-1.2 k£2® cm 2 ), were treated with the
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
custom-made growth medium as described above in the presence of MTX, MPA, and 5-
FU at 200 (xM. After three days of treatment, cell monolayers were washed with SFR
solution pre-equilibrated at 37°C in a 5% CO2 / 95% air environment for 20 min.
Basolateral fluid was removed and replaced with 0.8 ml dosing solution containing 2
piCi/ml 3H-uridine. Cellular uptake of 3 H-uridine was measured at 1 min, by washing
excess extracellular radioactivity. Clearwells® were washed by immersing into 100 ml of
ice-cold BRS (bicarbonated Ringer’s solution, where choline chloride was replaced with
an equimolar concentration of NaCl, K H 2PO 4 with NaFfjPCU, and choline bicarbonate
with NaHCC> 3). This procedure was repeated two more times. Cell monolayers were then
cut along the periphery and put into scintillation vials containing 0.5 ml of 0.5% Triton
X-100 for 30 min to lyse the cells. Twenty microliters of each lysate was taken for
protein assay using DC protein kit (Bio-Rad Laboratories, Hercules, CA) after thoroughly
vortexing each vial. Five milliliters of Econosafe scintillation cocktail (Research Products
International, Mount Prospect, IL) was added to the rest of each lysate for liquid
scintillation counting to assess the radioactivity in the Beckman LS 1801 System
(Fullerton, CA).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IV. RESULTS
D is c o v e r ie s a n d c o n f ir m a t io n s .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1. F u n c tio n a l E v id e n c e f o r B a so la ter a l N u c le o sid e T r a n spo rt
P r o c esses
1.1. Time course and Na+ dependency
The initial linear rate of 3H-uridine uptake from the apical fluid in primary cultured
RTEC using Na+ -containing BRS was 33.0 ± 4.0 fmol/mg protein/sec, significantly
higher than that in SFR solution (1.0 ± 0.3 fmol/mg protein/sec, Table 4-1). The time
course of 2 min uridine uptake (10 pCi/ml) from basolateral fluid revealed a linear
increase for up to 60 sec (5.8 ± 0.8 fmol/mg protein/sec estimated from the initial linear
portion). Basolateral uptake was 5 times lower compared to uridine uptake measurements
carried out from the apical compartment. No significant difference was observed between
basolateral uridine uptake from BRS or SFR solutions (p > 0.05; Table 4-1). By contrast,
> 90% of the apical uptake was dependent on Na+ (Table 4-1). Additionally, uridine
uptake from either apical or basolateral fluid observed at 4°C (re presenting nonspecific
binding of 3H-uridine) was about 10% of the values measured at 37°C.
1.2. Concentration dependency
Unlabeled uridine at concentrations ranging from 0 to 800 pM was used in an attempt to
measure saturation kinetics of 1 min uptake. Linear, diffusional component and saturable
uridine uptake were observed at 37 °C. The nonspecific diffusional uptake rate at 37 °C
was 0.07 pmol/mg protein/min/pM for uridine, and a rate of 0.06 pmol/mg
protein/min//iM was observed at 4°C. An estimate of carrier- mediated uptake is obtained
by subtracting the linear 4°C values from those observed at 37°C. The maximum
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4-1. Na+ -dependency of initial uptake rates, kinetic parameters (Km and J ^ ) , and
apparent permeability coefficients (P a p p ) of 3H-uridine (10 juCi/ml) measured from apical
and basolateral fluids of RTEC monolayers. Data are expressed as mean ± s.e.m. of 6
measurements and corrected for nonspecific adsorption.
Initial Uptake Rates Kinetic Parameters P
aPP
(x 1 0 " 6 cm/sec) BRS" SFR* Km (pk1) Jm (L :t (pmol/mg protein/min)
(fmol/mg protein/sec)
Apical 33.0 ±4.0 1.0 ± 0.3C 3.4 ± 1.8 48.5 ± 10.4 9.5 ±0.15 (A— »B)
Basolateral 5.8 ±0.8 5.6 ±0.8 210.2 ± 34.3r f 356.8 ± 22.8d 0.9 ± O.OT (B-±A)
aBRS: bicarbonated Ringer’s solution
b SFR: sodium-free Ringer’s solution
c significantly different between SFR and BRS ip < 0.05).
significantly different between basolateral and apical uridine uptake (p < 0.05).
e significantly different between B— »A and A— > B uridine transport (p < 0.05).
A->B = apical to basolateral direction
B— >A = basolateral to apical direction
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
radioactivity associated with cells at 4°C was 10-15% of the amounts detected at 37°C
(Table 4-1). Km of 210.2 ± 34.3 pM and Jm a x of 356.8 ± 22.8 pmol/mg protein/min for the
saturable component were obtained in uptake studies from basolateral fluid (Table 4-1).
The affinity of apical uptake for uridine was higher than that of basolateral uptake by two
orders of magnitude, where a Km of 3.4 ± 1.8 (iM and a Jm a x of 48.5 ± 10.4 pmol/mg
protein/min for the saturable component were obtained (Table 4-1).
1.3. NBMPR sensitivity
Uridine uptake from basolateral fluid in primary cultured rabbit tracheal epithelial cells
was inhibited in a biphasic manner in the range of nanomolar to micromolar NBMPR
concentrations (Fig. 4-1). NBMPR inhibits about 25% of total uridine uptake activity at
0.2 (iM NBMPR. The remainder of the activity was further inhibited by another 50% at
270 fiM NBMPR which probably was mediated by the less sensitive NBMPR
component. These inflection points represent IC50 values of 0.2 fiM and 270 juM,
respectively.
1.4. Substrate selectivity
Inhibition of basolateral uridine uptake was also investigated in the presence of the
nucleosides, nucleobase, and various nucleoside analogs at 500 fiM donor concentration
(Fig. 4-2). Basolateral uptake of 3H-uridine was inhibited by co-administered IDU, 2’-
dU, 3’-dU, ribavirin, and cladribine. In contrast, AZT, Ara-A, and Ara-C had no effect on
3H-uridine uptake. Interestingly, 5-FU, hypoxanthine, and ACV all exhibited weak
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
50 -
o
h m
■ # - »
C
O
O
o ^-
2 -3
501
-2 1 0 1
100'
25 -
3 4
Log [NBMPR faM )]
Fig. 4-1. Inhibition of 3H-uridine (10 /rCi/ml) uptake for 1 min from basolateral fluid in
primary cultured rabbit tracheal epithelial cell monolayers as a function of NBMPR
concentration, under Na+ -free condition. IC25 = 0.2 ± 0.09 fiM (Panel A) and IC75 = 270
± 0.7 /rM (Panel B), were estimated by non-linear curve fitting algorithms using
GraphPad Prism 3. Insets represent the continuous dose-response curve with two
inflection points and the enlarged segments of Panels A and B are drawn with broken
lines (---- ). Each data point represents mean ± s.e.m., n = 6 .
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
inhibition of 3 H-uridine uptake, which was not the case in apical uptake studies.
Basolateral 3 H-uridine uptake was significantly inhibited by all physiological purines and
pyrimidines (Fig. 4-2). Inhibition ranged from the highest of -65% by adenosine, to the
lowest of -30% by thymidine.
By direct comparison to results from preliminary studies (Mathias et al, 1996),
the interaction of these molecules at 100 jxM concentrations in BRS with the 3 H-uridine
uptake process from apical fluid occurred in a Na+ -dependent manner. Fig. 4-2
summarizes the substrate selectivity from the apical aspect of primary cultured RTEC.
The strongest interaction with up to 75% inhibition was observed for those analogs that
have substitutions on the nucleobase moiety and/or modification on the 2’-OH position
such as D D U , 2’-dU, and cladribine. Remarkably, analogs with modification on the 3’-OH
position (e.g., 3’-dU, ddl, and AZT) and those with an arabinose sugar-ring (e.g., Ara-A
and Ara-C) exhibited weak or no inhibition of apical 3 H-uridine uptake along with
nucleobases and analogues (e.g., hypoxanthine, and 5-FU). Acyclic nucleosides like
ACV did not affect the apical uptake of 3 H-uridine.
1.5. Efflux studies
The rate of 3 H-uridine effluxing into the basolateral fluid increased nonlinearly as a
function of the duration of apical preloading time for up to 60 min tested (Fig. 4-3). The
rate of efflux was about 6 times higher for the cells preloaded for 60 min, compared to
those preloaded for only 10 min. Basolateral efflux was not dependent on the presence or
absence of Na+ in the incubation buffer. Amount of uridine effluxed from cells increased
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Uridine
2'-dU
3‘-dU
IDU
Thymidine
AZT
Cytidine
Ara-C
Adenosine
Cladribine
Ara-A
Guanosine
Ribavirin
ACV
I nosine
d d l
5-FU
Hypoxanthine
13**,
H *
------------------------------------- h _ i* *
■ B - J * **
50
% Control
100
Fig. 4-2. Effect of nucleosides (adenosine, guanosine, inosine, cytidine, thymidine, and
uridine), various nucleoside analogs (ACY, Ara-A, Ara-C, AZT, 2’-dU, 3’-dU, IDU,
ribavirin, cladribine and ddl) and nucleobase analogs (hypoxanthine, 5’-FU), on 1 min
3 H-uridine (10 juCi/ml) uptake from basolateral (inhibitors at 500 ^M ,C ]) and apical
(inhibitors at 100 nM,WM) fluid in primary cultured RTEC monolayers. Asterisks(s)
represent significant decrease in uridine uptake compared to control (*p < 0.05, **p <
0.01 and ***p < 0.001) by ANOVA. Each data point represents mean ± s.e.m., n = 6.
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.41
T J
©
g 0.3-
£
0 )
0 )
.£ 0.2 -
TJ
3
I 0.1 ■
a
20 40 60 0
Time (min)
Fig. 4-3. Efflux of 3 H-uridine (10 p,Ci/ml) across the basolateral membrane of primary
cultured RTEC monolayers after apical preloading for 10 (• ), 30 (■), and 60 (A) min in
BRS. Each data point represents mean ± s.e.m., n = 6.
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
linearly for up to 10 min. Differential linear uridine efflux rates were observed in a
loading time-dependent manner (0.023 ± 0.001 pmol/min for 10 min, 0.045 ± 0.002
pmol/min for 30 min, and 0.12 ± 0.005 for 60 min preloading, respectively) into the
basolateral reservoir, that reached different plateaus. The percentage of intact uridine
detected in the basolateral bathing fluid after 1 hr of efflux compared to total cellular
radioactivity, was somewhat dependent on the duration of apical preloading: 1.85% for
10 min, 2.17% for 30 min, and 3.09% for 60 min preloading.
2. M o l e c u l a r a n d F u n c t io n a l C h a r a c t e r is t ic s o f I d e n t if ie d ENT2 a n d I t s
S p l ic e V a r ia n t f r o m R a b b it T r a c h e a
2.1. Molecular characteristics
Two rabbit equilibrative-insensitive (ei) nucleoside transporter cDNA clones, termed
rbENT2 and rbENT2A, predicted the existence of two transporter isoforms differing at
their carboxyl termini. DNA sequence analysis revealed that a 38-bp deletion (rbENT2
nucleotide sequence positions from 886 to 923, Fig. 4-4, solid line) and a 9-bp insertion
(rbENT2 nucleotide sequence positions between 977 and 978, Fig. 4-4, dashed line) were
evident in the alternatively spliced region of genomic introns/exons in the rbENT2
mRNA generation process. Their derived amino acid sequences revealed that the rbENT2
and rbENT2A isoforms were identical through amino acid 289 but diverged beyond this
point with rbENT2 having an additional 167 amino acids and rbENT2A an additional 126
amino acids with only 16% identity and 35% similarity (Fig. 4-5). Hence, alternative
splicing was speculated to be the mechanism responsible for generating the two
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 - ACGCGGGGGCGGCCGCCATGGCGCGAGGAGACGCCCCGCGGGACAGCTACCACCTGGTCG - 60
M A R G D A P R D S Y H L V G
6 1 - GGATCAGCTTCTTTATCCTGGGGCTGGGCACCCTCCTTCCCTGGAACTTCTTCATCACCG - 1 2 0
I S F F I L G I i G T L L P W N F F I T A
1 2 1 - CCATCCCGTACTTCCAGGGGCGGCTGGCGGGGGCCAACGGCACAGCCGGGGCCCTGAGCA - 1 8 0
IP Y F Q G R L A G A N G T A G A L S T
1 8 1 - CCAACCACACGGGCCCCACAGACGCCTTCAACTTCAACAACTGGGTGACACTGCTGTCAC - 2 4 0
N H T G P T D A F N F N N W V T L L S Q
2 4 1 - AGCTGCCTCTGCTGCTCTTCACCCTCCTCAACTCCTTCCTGTACCAGTGCATCCCGGAGG - 3 0 0
L P L L L F T L L N S F L Y Q C I P E A
3 0 1 - CCGTTAGGATCCTGGGCAGCCTGCTGGCCATGCTGCTGCTCTTCGCTCTCACGGCCGCGC - 3 6 0
VRIL GS LLAMLLLFALTAAL
3 6 1 - TGGTCAAGGTGGACGTGAGCCCTGGGCCCTTCTTCTCCATCACCATGGCCTCCGCCTGGT - 4 2 0
VKV DVS PGPFFSITMASAWF
4 2 1 - TCATCAACTCTTTCTGCGCAGTTCTGCAGGGCAGCCTCTTCGGGCAGCTGGGCACCATGC - 4 8 0
I N S FC A VL Q GS L F GQ L G TM P
4 8 1 - CGTCCACGTACAACACCCTCTTCCTCAGCGGCCAGGGCCTGGCCGGGATCTTTGCTGCCC - 5 4 0
STYNTLFLSG QGL AG IFA AL
5 4 1 - TTGCCATGCTCACGTCCATGGCCAGTGGTGTGGATGCCCAGACCTCCGCCCTGGGGTACT - 6 0 0
A M L T S M A S G V D A Q T S A L G Y F
6 0 1 - TCCTCACGCCCTGCGTGGGCATCGTCCTGTCCATCGTGTGCTACCTGAGCCTGCCGCACC - 6 6 0
L T P C V G I V L S I V C Y L S L P H L
6 6 1 - TGGAGTTTGCCCGCTACTACCTGGCCAAGAAACCATCGCAGGCCCCGACCCAAGAGCTGG - 7 2 0
E F A R Y Y L A K K P S Q A P T Q E L E
7 2 1 - AGACCAAAGCCGAGCTCCTCCAGTGTGATGAGAAGAACGGGGTTCCCAGCAGCCCCCAGA - 7 80
T K A E L L QC DE K NG VP SS PQ K
7 8 1 - AGGCGGCCCTGACCTCGGATGTTGACCCTGAGAAGGAGCCGGAGCTGGAGCCTGCGGAGC - 8 4 0
A A LT SD V DP EKE PEL EP AEP
8 4 1 - CCAGGGATCCGGGAAAACCCTCGGTCTTCGTGGTCTTCCGGAAGATCTGGCTGACGGCAC - 9 0 0
R D P G K P SV FV V FR K I W L T A L
S
9 0 1 - TGTGCCTTGTGTTGGTCTTCACAGTCACCCTGTCGGTCTTCCCCGCCATCACTGCCATGG - 9 6 0
C LV LV FT VT LS VF PA IT AM V
H P V G L P R H H C H G
m m a m m m m a m m m
H Q P
TCCACCAGC
' y '
9 6 1 - TGACCAGCTCCACCAGCCCCGGGAAGTGGAGTGAGTTCTTCAACCCCATCTGCTGCTTCC - 1 0 2 0
T S S T S P G K W S E F F N P I C C F L
D Q L H Q L R E V E L V F Q P H L V L P
1 0 2 1 - TCCTCTTCAACGTCATGGACTGGCTGGGGCGGAGCCTGACCTCCTACTTCCTATGGCCCG - 1 0 8 0
L F N V M D W L G R S L T S Y F L W P V
P F Q R H G L A G A E P D L L L P M A R
1 0 8 1 - TGGAGAACAGCCGGCTGCTGCCCCTGCTGGTCCGCCTGCGCTTCCTCTTCGTGCCTCTCT - 1 1 4 0
E N S R L L P L L V R L R F L F V P L F
G E Q P A A A P A G L P A L P L R A S h
1 1 4 1 - TCATGCTGTGCCATGTGCCCCAGCGTGCCCGGCTGCCCATCCTCTTCCCCCAGGCCGCTA - 1 2 0 0
HLC HVP Q RA R L PX L FP Q AA N
H A V P C A P A C P A A H P L P P G C L
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 0 1 - ACTTCATCATTTTCATGCTCCTCTTCGCCGTCTCCAACGGCTACCTGGTGTCCCTCACCA - 1 2 6 0
FI IF ML LFA VSN GY LVS LT M
L H H F H A P L R R L Q R L P G V P H H
12 6 1 - TGTGCCTGGCACCCAGGCAGGTGCTGGCGCAllHGAGGGAGGTGGCCGGCGCCCTCATGA - 1 3 2 0
C L A P R Q V L A H E R E V A G A L M T
V P G T Q A G A G A
1 3 2 1 - CCTTCTTCCTGGCCCTCGGACTCTCCTGCGGGGCCTCCCTCTCCTTCCTCTTCAAGGCCC - 1 3 8 0
F FL A LG LS CG AS LSF LFK AL
1 3 8 1 - TGCTQTGAAGCCCCCTGTGGAGCTGAGACCACCCAGCCCCTGGGGAATGACAAGCCGGGC - 1 4 4 0
In
1 4 4 1 - TCAGGCTGTGCCGGGAGGGCCGGGAGCTGCTCACCGGGGGTGCTCCGAGGTCAAGTTCAC - 1 5 0 0
1 5 0 1 - CGAGGGTCACTCGCACCCAGGAACGGATCGGGCCTTGTAGAAGACCCACTCAACGGAGCA - 1 5 6 0
1 5 6 1 - GGGACAAGAGAGTGCAACCGACCGTGGCGCTGCCCACCGTGAGACTGCATGTCAGTGCCA - 1 6 2 0
1 6 2 1 - AGGGGGCCTTGCCCACATGGGCCCTGCCCTGGGCTCAGAGTCGGAGGTGGGGCCGGGCAG - 1 6 8 0
1 6 8 1 - GTGCTCCTCCCTCCAGGCCTCAGCCGGCCCCAAGATCAAGTCTCTCCAAGCTGGGGGCAA - 1 7 4 0
1 7 4 1 - AGGATGCTGGGAGAGTGGAGGGGGGGCAGGTGCTGAGGCTCCGAGGGTGCTCTGTGGACC - 1 8 0 0
1 8 0 1 - GGGGCCTGCGCCTCCAGCTGCCCGTTGCCAACGCTCACTCTCCCTGGGTCTGGCTCTGGG - 1 8 6 0
1 8 6 1 - CACCGTGTCAGCCTGCACGTGGGCACCTCACTTTACAGCTGGCCAGTGTCTGCACACCCC - 1 9 2 0
1 9 2 1 - ACTCTCCGTGCAGCCTCCCCGAGGCCGGGGAGGGAGCAGAGCAAGACCGTCGTGCAGGCG - 1 9 8 0
1 9 8 1 - GGAAATCAGGTCCAGAGGGGAGGAGTGCACCCCAGCCCCCACCTCCTGCAGGTGCTGTTT - 2 0 4 0
2 0 4 1 - TGCCTGCCCCCTGCCCGCTCCTCTTCCTCGGGGGCTCAGGGCTGGGGCCCCACGCACTGC - 2 1 0 0
2 1 0 1 - CCCACCTTTCTGCTCCTGGATCTGTTAACAGCCACGTTTGTTTGC - 2 1 4 5
Fig. 4-4. Nucleotide and deduced amino acid sequences of rbENT2 (GenBank™
Accession No. AF323951) andrbENT2A (GenBank™ Accession No. AF323952) cDNA
clones. The entire nucleotide and deduced amino acid sequences of rbENT2 (2145 bp)
and rbENT2A (2116 bp) cDNA inserts are shown. Nucleotide numbers are shown on the
left and right. The solid line indicates a 3 8-bp deletion (rbENT2 nucleotide sequence
positions from 886 to 923) absent in the rbENT2A cDNA and the dashed line indicates a
9-bp insertion present in the rbENT2A cDNA (rbENT2 nucleotide sequence positions
between 977 and 978). The deduced amino acid sequence of rbENT2 is marked with
boldface letters. The C-terminal 126-amino acid sequence for rbENT2A is indicated by
italicized letters. The shaded region indicates a premature stop codon TGA in the
rbENT2 cDNA resulting from alternative splicing. The boxed region indicates a stop
codon TGA in the rbENT2 cDNA.
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TCTGGCTGACGGCACTGTGCCTTGTGTTGGTCTTCACA
ACCAGCTCC
ENT2 primary Transcript
V / -
#1 #2
Alternative splicing
f 4
rbENT2A mRNA generated by splicing event #1 rbENT2 mRNA generated by splicing event #2
2116 bp s 2145 bp
y / -
i
Translation
rbENT2 (49 kDa) NH3 T
289 456
jco o
rbENT2A (44 kDa) NH3 *£
T™
289
COO-
415
Fig. 4-5. Diagram of rbENT2 primary transcript and the gene products produced by
alternative splicing. Splicing event #1 (alternative splicing pathway) in which the hatched
box ( m ) is deleted to yield a 2116-bp rbENT2A message. Splicing event #2 (default
splicing pathway) in which the black box ( H I) is deleted to result in a 2145-bp rbENT2
message. Translation of the messages produces the rbENT2 and rbENT2A isoforms,
which are identical through the amino acid residue 289 but then diverge beyond this point
with rbENT2 having an additional 167 amino acids and rbENT2A an additional 126
amino acids.
110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
transporter isoforms. Alternative splicing resulted in a premature stop codon TGA
(rbENT2 nucleotide sequence positions between 1290 and 1292, Fig. 4-4, shaded region)
due to a frame shift on the open reading frame, and because a combination of the two
changes resulted in a number of base pairs that was not a multiple of three. The rbENT2
cDNA (GenBank™ Accession No. AF323951) is 2145 bp long with an open reading
frame comprised of 1371 bp (including the stop codon, Fig. 4-4, boxed region), encoding
a protein of 456 amino acids with a predicted molecular mass of 50 kDa. This open
reading frame is flanked by a 17-bp 5’-untranslated region and a 757-bp 3’-untranslated
region. In contrast, the rbENT2A cDNA (GenBank™ Accession No. AF323952) has an
open reading frame encoding a protein of 415 amino acids which is shorter by 41 amino
acid residues compared to rbENT2. The molecular mass of rbENT2A protein was
predicted to be 44 kDa. The deduced amino acid sequence of rbENT2 exhibits significant
similarity to several sequences in GenBank™, including Na+ -independent nucleoside
transporters (Jarvis and Young, 1986; Yao et al., 1997; Boleti et al., 1997; Griffiths et al.,
1997b), adenosine-pyrimidine nucleoside transporter (LdNTl) from the protozoan
parasite Leishmania donovani (Vasudevan et al., 1998), and adenosine transporter from
Toxoplasma gondii (Chiang et al., 1999). rbENT2 showed 89% identity and 92%
similarity in primary amino acid sequence to hENT2 (GenBank™ Accession No.
AAC39526) and displayed 87% identity and 91% similarity to rENT2 (GenBank™
Accession No. 054699). This suggests that rbENT2, rENT2, and hENT2 may share
similarities in their structures. Moreover, rbENT2A is 67% identical and 76% similar to
rbENT2 at the amino acid level. Membrane topology of rbENT2 and rbENT2A based on
Kyte-Doolittle approach was predictable as follows: TMPred™ (available in the ExPASY
111
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Molecular Biology Server, www.expasy.ch). The rbENT2 was predicted to have 11
transmembrane domains (TMDs) connected by short hydrophilic regions, except for an
extracellular loop linking putative TMDs 1 and 2, and a large cytoplasmic loop linking
putative TMDs 6 and 7 (Fig. 4-6A). By contrast, the rbENT2A protein was predicted to
have either 6 or 7 TMDs (Fig. 4-6B and 4-6C). The first six putative TMDs of rbENT2A,
comprising the amino acid segment spanning from N- terminus to the point of alternative
splicing (located between TMD 6 and TMD 7), have a predicted topology profile
identical to that of the WT rbENT2 protein (Fig. 4-6A). Both proteins contain the same
potential IV-linked glycosylation sites (Asn4 7 , Asn5 6 , Fig. 4-7, asterisks) on an
extracellular loop between putative TMDs 1 and 2 based on PPsearch™, available in the
ExPASY Molecular Biology Server, www.expasy.ch). Asn4 7 7 4 8 is a conserved IV-linked
glycosylation site in both ENT1 and ENT2 isoforms from multiple species (Griffiths et
al, 1997a; Griffiths et al., 1997b; Yao et al., 1997). PPsearch™ also highlighted Ser2 2 7
(Fig. 4-7, arrow) in both rbENT2 and rbENT2A, as a possible protein kinase A (PKA)
phosphorylation site located on an intracellular loop between putative TMDs 6 and 7.
2.2. Tissue distribution
The tissue distribution pattern of these genes was assessed next. In a variety of rabbit
tissues examined, this novel splicing event was ubiquitously present. Using Northern blot
analysis, a band of ~1.8-2.0 kilobases corresponding to the tracheal rbENT2/rbENT2A
mRNA was observed (Fig. 4-8A). The rbETN2A mRNA was not easily resolved from
that of rbETN2, since these two mRNAs only differ in 28 nucleotides. Therefore, RT-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
COOH
Fig. 4-6. Predicted membrane topological model of rbENT2 and rbENT2A. Using
TMPred Server (Hofmann and Stoffel, 1993), rbENT2 topology {Panel A) was predicted
to have 11 TMDs with an intracellular N-terminus, an extracellular C-terminus, and a
large intracellular loop between TMDs 6 and 7. rbENT2A was predicted to have either 6
{Panel B) or 7 TMDs {Panel C) with an intracellular N-terminus.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TM D 1
hENTl 1TTSHI
hENT2 HaRGD.
rENT2 lAHGN.
rbENT2A HaRGD.
rbENT2 KARGD.
)MSQNVSLVTAEL SKDA 60
GAG----------------- NSTAR 52
.GT--------------------NSSAE 51
GA------------------- NGTAG 51
GA-------------------jjjfGTAG 51
TMD1
TMD2
hENTl
hENT2
rENT2
Qi
\PAA
XLSTNHTG gEDAFl
TFWTNHTS
rbENT2A ALgTNHTG
rbENT2 ALgTNHTG
STDTF
aPDAF
TMP3
TMD2
TMD4
TMD3
hENTl
hENT2
rENT2
rbENT2A
rbENT2
SVWF
SVWE
.SHTAPIM
jSTgSTLFL
STSSTLFL
STTOTLFL
ISTMNTLFL
TMD5
TMD4
TMD6
hENTl
hENT2
rENT2
rbENT2A
rbENT2
VL
CVG VL
TMD5 TMD6
hENTl EGPGEj
hENT2 SQAQAj
rENT2 -QAPV8
rbENT2A SQAPTj
rbENT2 SQAPTJ
t
— LD ■ISKGffl------------ E Br a g k e e s g v s v s n ------------ SQW rN -E SH |jlK A I 2 8 5
LE AE S lQ SDH-NGIPSS S q k v a l t l d l d l e k e p e s e p d e ® 2KPGKPB8VFTV 2 8 6
LE AE Bl g a d Hk n g i p v s Bq q a g p t l d l d p e k e l e l g l e e w 2KPGKPKjVFW 2 8 5
LE I a e JLQCDHKNGVPSS s q k a a l t s d v n p e k e p e l e p a e S i RDPGKPK8VFW 2 8 6
LE AE BLQCDHKNGVPSS Sq k a a l t s d v d p e k e p e l e p a e S i RDPGKPRSVFW 2 8 6
TMD7 TMDS
hENTl LKNISVLAFSVCFIFTITIGMFPAVTVEVKSSIAGSSTWERYSl
hENT2 FQKIWLTALCLVLVFTVTLSVFPAITAMVTSST-SPGKWSQF|
rENT2 FRKIWLTALCLVLVFTVTLSVFPAITAMVTTSSNSPGKWSQF|
rbENT2A FRKSHPVGLPRHHCHGDQLHQLHQ------------------------PREVEL'
rbENT2 FRKIWLTALCLVLVFTVTLSVFPAITAMVTSST-SPGKWSEFi
r SCFLT
jlCCFLL
1ICCFLLI
1LVLP
lICCFLLl
IFDWLgR 3 4 5
1MDWLHR 3 4 5
iWLgR 3 4 5
RHGLAg- 3 3 4
IWLgR 3 4 5
TMD7
TMD9
TMD8
hENTl SLTAVFMWPGKDSRft SSLVLARLVF'
hENT2 SLTSYFLWPDEDSRL»LLVCLRFLFVKLFM*CHVPQgSR
rENT2 SLTSYFLWPDEDSQLWLLVCLRFLFvI l FMICHVP
rbENT2A -------------------AEPDLLKMARGEQPAAAH-AGgPALPLHAS
rbENT2 SLTSYFLWPVENSRLBLLVRLRFLFviLFMlCHVPQi
.LLHCNIKPBrY ITWFEHDiWFIFFMAAFAF 4 0 5
PILFPQE
J p iI F W Q E
iuVPCAPl
]PILFPQA
iTFXTFMLLFAV 4 0 5
iTFITFMLLFAI 4 0 5
3PAAHPLPPGC 3 8 4
3IIFMLLFAV 4 0 5
TMD9 TMD10
TMD10 TMD11
hENTl SNGYLASLCMCFGPKKVKPAEAETAGAIMAFFLCLBLALMVFSFLFRAIV 4 5 6
hENT2 SNGYLVSLTMCLAPRQVLPHEREVAGALMTFFLAlB l S cH s LSFLFKALL 4 5 6
rENT2 SNGYFVSLTMCLAPRQVLPHEREVAGALMTFFLALgLSC|
rbENT2A LLHHFHAPLRRLQRLPGVPHH--------------------------VPw!
rbENT2 SNGYLVSLTMCLAPRQVLAHEREVAGALMTFFLAL1LSC1
ISLSFLFKALL 4 5 6
IA------------------- 4 1 5
ISLSFLFKALL 4 5 6
TMD11
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 4-7. Deduced amino acid sequence of rbENT2 and rbENT2A compared with both
human equilibrative nucleoside transporters hENTl (Griffiths et al, 1997a) and hENT2
(Griffiths et al, 1997b) and the rat equilibrative nucleoside transporter rENT2 (Yao et
al, 1997). Alignment was performed using ClustalW (Thompson et al, 1994) service at
the European Bioinformatics institute (EBI). Identical amino acids among five sequences
are shown in white on a black background. Spaces introduced to optimize the alignment
are indicated by dashed lines. Labeled solid lines over hENTl (Griffiths et al., 1997b)
and under rbENT2 indicate the predicted TMDs of these two transporters using TMPred
Server (Hofmann and Stoffel, 1993). The numbers at the right indicate the amino acid
positions in each sequence. The putative A-glycosylation sites and the potential PKA
phoshorylation site are indicated by asterisks and by arrow, respectively.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
397 nt->- mm *&* rite
U p V P P W p -4-368 nt
Fig. 4-8. Detection of rbENT2 and rbENT2A transcripts by Northern blot analysis (Panel
A), RT-PCR (Panel B) and RPA (Panel C). Panel A: Detection of ENT2 by Northern blot
analysis in rabbit trachea and rat jejunum (Yao et al., 1997). The RNA blot was
hybridized with a biotinylated rbENT2 gene-specific probe. The blot was then stripped
and rehybridized with a GAPDH probe to ascertain uniform RNA loading. RNA size
markers are indicated on the left, and the approximate sizes of the hybridized signals are
indicated on the right. Panel B\ Total RNA isolated from rabbit tracheal epithelial cells,
kidney medulla, kidney cortex, and intestine were reverse transcribed using oligo (dT)i8
primer. The resulting cDNAs were amplified by Methods 3.2. The RT-PCR products
were separated by electrophoresis through 2% agarose gel and visualized by UV with
ethidium bromide staining. DNA size (in bp) markers are 506, 396, 344, and 298. Panel
C: Total RNA isolated from rabbit tracheal epithelial cells (50 ftg), kidney medulla (30
Hg), kidney cortex (30 fig), and intestine (30 fig) were protected with 100 pg of the
anti sense cRNA probe and digested with RNaseTl/A before precipitation, and then
separated by polyacrylamide gel as described under Methods 3.3. The expected sizes of
the protected fragments are indicated, nt, nucleotides.
116
GAPDH
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PCR was performed to characterize the expression intensities of rbETN2A with respect to
rbENT2. RT-PCR analysis (Fig. 4-8B) showed that transcripts of rbENT2 and rbENT2A
coexist in various tissues, including small intestine, kidney cortex, kidney medulla and
trachea. DNA sequence analysis confirmed that nucleotide sequences of these RT-PCR
products are identical to the corresponding segments of tracheal rbENT2 and rbENT2A.
To rule out the possibility of any RT-PCR artifacts, RPA was performed using an anti
sense cRNA that hybridizes to the message, thus circumventing both reverse transcription
and polymerase chain reaction steps. Biotin- labeled, 513-nucleotide long, antisense RNA
probe was used to selectively detect mRNA fragments, corresponding to rbETN2 and
rbENT2A, differing by 28-nucleotides in length (Fig. 4-8C). Consistent with RT-PCR
results, two ribonuclease-protected fragments with the expected sizes (397 and 368
nucleotides, respectively, for rbENT2 and rbETN2A) were detected in the same tissue
samples of trachea through kidney. The overall rbENT2 message level remained constant
across the tissues examined (Fig. 4-8B and 4-8C). In contrast, rbENT2A transcript was at
a lower intensity than rbENT2 in all tissues examined. The relative expression levels of
rbENT2 to rbETN2A messages had a fixed ratio of 5 to 1 in all samples.
2.3. Expression and detection
2.3.1. Apparent molecular mass
To examine whether or not the engineered recombinant rbENT2 and rbENT2A constructs
were expressed as proteins with the expected sizes in HEK293 and MDCK cells, the
cDNAs encoding rbENT2 and rbENT2A were individually subcloned into HA-tagged
117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pSG5 vectors, since antibodies against native rbENT2 were not available. The constructs
(rbENT2/pSG5.HA and rbENT2A/pSG5.HA) encoding rbENT2 and rbENT2A were
transfected into HEK293 (Fig. 4-9A) or MDCK (Fig. 4-9B) cells for in vitro heterologous
protein expression. The HA-tagged recombinant rbENT2 and rbENT2A proteins were
detected in immunoblots with apparent molecular mass of 49 and 44 kDa, respectively
(Fig. 4-9A and 4-9B). As expected, no band was detected in the cell lysate from cells
transfected with parent vector (pSG5.HA), indicating that the HA-tagged rbENT2 and
rbENT2A recombinant transporters were indeed expressed in HEK293 or MDCK cells.
2.3.2. Membrane localization
To investigate the membrane localization of rbENT2 and rbETN2A, MDCK cells grown
on Clearwells® were transiently transfected with pSG5.HA/rbENT2 or
pSG5.HA/rbENT2A, stained with specific anti-HA primary antibodies for identification
of the fusion proteins, and visualized with FTTC-conjugated secondary antibodies by laser
scanning confocal microscopy. In separate control experiments, anti-HA antibodies
demonstrated no staining in mock-transfected MDCK cells. Thus, only cells expressing
HA-tagged transporters were labeled by anti-HA antibodies. Cells were double labeled
with antibodies recognizing tight junctional marker ZO-1. ZO-1 was localized below
tight junctions in the cytoplasmic domain. Figure 4 shows confocal microscopy of
MDCK cells expressing rbENT2-HA and rbENT2A-HA, where an x-z cross-section was
shown. Some fractions of pSG5.HA/rbENT2 or pSG5.HA/rbENT2A were sorted to
basolateral membrane regions. Confocal fluorescence micrographs (xy plane) of MDCK
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
p
z
<
C M
<
< N
z
U J
X 2
kDa
50
< # —49 kDa
« * W .
<— 44 kDa
35
Fig. 4-9. Detection of recombinant, HA-tagged, rbENT2 and rbENT2A proteins
expressed in HEK 293 (Panel A) or MDCK (Panel B) cells by Western blot analysis. Cell
membrane proteins were prepared from HEK293 or MDCK cells transiently transfected
with HA-tagged rbENT2 and rbENT2A DNA constructs. Samples (20 jug/lane) were
subjected to 10 % SDS-polyacrylamide gel electrophoresis and transferred to a
nitrocellulose membrane. Membrane blot was incubated with mouse anti-HA monoclonal
antibody [1:1000 (v/v) dilution], followed by wash and incubation with a peroxidase-
conjugated AffiniPure donkey anti-mouse IgG at a dilution of 1:50,000 (v/v). The
resulting signals were visualized on x-ray films by ECL. The apparent molecular masses
of the protein bands are indicated by arrows on the right, and the positions of molecular-
mass markers (in kDa) are indicated on the left.
119
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
cells, transiently transfected with HA-rbENT2 and HA-rbENT2A {green, Fig. 4-10), were
acquired in 1.0-^m increments. Tight junctions were stained for ZO-1, shown in red (Fig.
4-10). HA-rbENT2 and HA-rbENT2A are predominantly localized to the basolateral
plasma membrane region below tight junctions (Fig. 4-10). These data were in agreement
with our previous functional studies performed in primary cultured rabbit tracheal
epithelial cells that Na+ -independent nucleoside transporter proteins are primarily sorted
to basolateral membrane.
To further confirm whether MDCK cells transiently transfected with
rbENT2/pcDNA3 or rbENT2A/pcDNA3 were expressing transporters in a polarized
manner, [3 H]-uridine was applied either apically or basolaterally to cells grown on
Clearwells®. Na+ -independent uridine accumulation after 5 min was significantly greater
in cells expression rbENT2/2A than in mock-transfected controls. Importantly, uridine
uptake from the basolateral fluid was three-fold greater than that from compared to the
apical fluid (Fig. 4-11).
2.3.3. Cellular colocalization
In a previous study mitochondria isolated from rat liver exhibited 2’-
deoxyguanosine transport which was not blocked by NBMPR (Watkins and Lewis,
1987), whereas mitochondria isolated from rat testis possessed high-affinity binding of
NBMPR (Camins et al., 1996). These results suggested the existence of possible
nucleoside-selective transport systems in the mitochondria responsible for the movement
of nucleosides between the cytosol and the inner mitochondrial matrix, which contain
120
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 4-10. Confocal fluorescence micrographs (xy and xz sections) ofMDCK cells grown
on permeable supports, demonstrating polarization of HA-rbENT2 and HA-rbENT2A to
the basolateral membrane region. A and C: Confocal fluorescence micrographs (xy plane)
ofMDCK cells, transiently transfected with HA-rbENT2 (A) HA-rbENT2A (C),
respectively, were acquired in 1.0-pm increments. HA-rbENT2 or HA-rbENT2A
fluorescence is shown in green, while tight junctions are stained for ZO-1, shown in red.
B and D. Confocal fluorescence micrographs (xz plane) showing distribution of HA-
rbENT2 (B) HA-rbENT2A ( .D), respectively, along the apical-basal axis, respectively.
HA-rbENT2 and HA-rbENT2A are predominantly localized to the basolateral plasma
membrane region below tight junctions. Scale bar is 10 pm. AP, apical membrane; B M ,
basal membrane.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8 S« 30
« o £
9-S
3 ^
© 5.
'■ § 0 5
■1^
3 «
£ 20
rbENT2 rbENT2A Mock
rbENT2 rbENT2A Mock
Fig. 4-11. Functional localization of MDCK cells transiently transfected with
pSG5.HA/rbENT2 (closed bar), pSG5.HA/rbENT2A (hatched bar) constructs and
pSG5.HA parent vector alone (open bar). 3 H-uridine uptake was performed after MDCK
epithelial cell monolayers were polarized by growth on 0.4 pm Clearwells® for 7 days.
3 H-uridine uptake (5 pM) was measured at 5 min from either apical (AP, Panel A) or
basolateral (BL, Panel B) side in the absence of sodium. The rbENT2- or rbENT2A-
mediated apical or baoslateral uridine uptake was corrected by the uptake data observed
in the presence of 1 mM unlabeled uridine. Points represent mean ± s.e.m., n = 6.
122
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
enzymes for nucleoside and nucleotide metabolism. To study the subcellular distribution
of rbENT2 and rbENT2A proteins expressed in transiently transfected MDCK and HEK
cells, confocal fluorescence microscopy was employed. DiOCe(3) was used to stain the
mitochondria at low concentrations (-0.5 pM or -0.1 pg/mL) and also stain the
endoplasmic reticulum when used at higher concentrations (-5-50 pM or -1-10 pg/mL).
As indicated by the combination of red and green filters showing in yellow (Fig. 4-12), a
reticular staining pattern colocalizes with the staining profile of DiOCe(3), a hexyl-chain
carbocyanine dye known to stain specifically the ER of mammalian cells.
The lysosomal nucleoside transport process in human fibroblasts exhibited broad
permeant selectivity and NBMPR sensitivity (Pisoni and Thoene, 1989). Lysosomal
nucleoside transporters predominantly function in export of degraded nucleic acids from
lysosomes to cytoplasm because lysosomes contain enzymatic activities capable of
degrading nucleic acids to their constituent nucleosides. Next, whether any rbENT2 and
rbENT2A were also present in lysosomes was investigated and late endosomes using
LysoTracker Red as a marker, a weak base, fluorescent amine that accumulates in acidic
compartments (Bucci et al., 2000). Cells were incubated with LysoTracker Red for 2hr
prior to fixation, permeabilization, and subsequent immunostaining with anti-HA primary
antibody. Punctate structures throughout the cytoplasm, that correspond to lysosomes or
large endosomes, were clearly observed (Fig. 4-13). Furthermore, none of these
structures colocalized with either rbENT2 or rbENT2A (Fig. 4-13). Additional evidence
for the occurrence of nucleoside transport processes in intracellular membranes of
mammalian cells was demonstrated in cultured human choriocarcinoma (BeWo) cells
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 4-12. rbENT2 and rbENT2A colocalize with an ER marker. Transiently transfected
HEK293 (A-F) or MDCK (G-L) cells, grown on Lab-Tek® II Chamber Slide™,
expressing HA-tagged rbENT2 (A -C and G -I ) or HA-tagged rbENT2A (D-F and J-L)
were incubated with an ER marker carbocyanine dye DiOCg for 30 min with a
concentration of 10 pg/mL at 37°C in the dark and then fixed, permeabilized, and stained
with mouse anti-HA monoclonal antibody [1:1000 (v/v) dilution], followed by washing
and incubation with a rhodamine-conjugated AffiniPure™ donkey anti-mouse IgG at a
dilution of 1:500 (v/v). Cells were viewed with red excitation/emission settings to detect
rbENT2 (A and G) or rbENT2A (.D and J) and with green excitation/emission settings to
detect the carbocyanine dye, DiOCe (B, E, H, and K). Merged yellow color containing
green and red images is shown in C , F, I, and L. There is staining throughout the
particular region for rbENT2 and rbENT2A. The scale bar, applicable to all panels, is 25
pm.
124
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 4-13. rbENT2 or rbENT2A does not colocalize with a lysosomal marker. Transiently
transfected HEK293 (A-F) or MDCK (G-L) cells, grown on Lab-Tek® II Chamber
Slide™, expressing HA-tagged rbENT2 (A-C and G-I) or HA-tagged rbENT2A (D-F and
J-L) were incubated with 60 nM LysoTracker Red DND-99 for 2 hr and then fixed,
permeabilized, and stained with mouse anti-HA monoclonal antibody [1:1000 (v/v)
dilution], followed by washing and incubation with FITC-conjugated AffiniPure™
donkey anti-mouse IgG at a dilution of 1:500 (v/v). Cells were viewed with green
excitation/emission settings to detect rbENT2 (B and H) or rbENT2A (E and K) and with
red excitation/emission settings to detect LysoTracker Red (A, D, G, and J). There is no
overlap of green staining for rbENT2 or rbENT2A with the punctate red structures
labeled by LysoTracker Red in transiently transfected HEK or MDCK cells as shown in
C, F, i, and L. The scale bar, applicable to all panels, is 25 pm.
125
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
using a proteoliposome reconstitution (Mani et al, 1998). The data suggested that the
presence of es-mediated (82-85%) and ei-mediated (15-18%) transport in nuclear
envelopes might play an important role in the translocation of nucleosides between the
cytosol and the luminal compartments of nuclear membranes for DNA and RNA
synthesis. Fig. 4-14 shows transiently transfected HEK or MDCK cells with rbENT2 and
rbENT2A stained in green, and the nucleus stained in red with propidium iodide. In
virtually all cells expressing rbENT2 or rbENT2A, staining was tightly restricted to the
region immediately surrounding the nucleus (Fig. 4-14), as well as the perinuclear areas
(Fig. 4-14).
2.4. Functional characteristics
2.4.1. In HEK heterologous expression system
a) Validation of HEK heterologous expression system
Using 3 H-uridine as a substrate, the endogenous uridine uptake in confluent HEK293 cell
monolayers grown on 12-well plates was examined. To determine if there was any
detectable Na+ -dependent endogenous uridine uptake, time course of 5 jtiM 3 H-uridine
uptake was measured in the presence and absence of Na+ (Fig. 4-15 A). Na+ -dependent
uridine uptake was significantly greater than Na+ -independent uptake in HEK293 cells,
suggesting that the endogenous uridine uptake activity in HEK293 cells was sodium-
dependent. Thus, all uridine uptake studies in HEK293 cells transfected with ei-
transporter gene in this study were carried out in the absence of Na+ to rule out the
involvement of Na+ -dependent nucleoside transport processes. To determine next
whether uridine uptake by HEK293 cells was mediated by saturable Na+ -independent
126
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 4-14. rbENT2 or rbENT2A does not colocalize with a nucleus marker. Transiently
transfected HEK (A-F) or MDCK (G-L) cells, grown on Lab-Tek® II Chamber Slide™,
expressing HA-tagged rbENT2 (A-C and G-I) or HA-tagged rbENT2A (D-F and J-L)
were fixed, permeabilized, and stained with mouse anti-HA monoclonal antibody [1:1000
(v/v) dilution], followed by washing and incubation with a FITC-conjugated
AffiniPure™ donkey anti-mouse IgG at a dilution of 1:500 (v/v). Cells were then treated
with PBS containing 100 pg/ml RNase A for 20 min at room temperature prior to
staining with propidium iodide at a concentration of 50 pg/ml in the dark for 20 min at
room temperature. Cells were viewed with green excitation/emission settings to detect
rbENT2 (B and H) or rbENT2A (E and K) and with red excitation/emission settings to
detect propidium iodide (A, D, G, and J). Merged green and red images are shown in C,
F, I, and L. The scale bar, applicable to all panels, is 25 pm.
127
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
transport processes, time course of 5 juM 3 H-uridine uptake under Na+ -free conditions
was measured in the presence and absence of excess unlabeled uridine (Fig. 4-15B).
Endogenous uridine uptake in HEK293 cells under these latter conditions was
significantly abolished in the presence of 1 mM unlabeled uridine, indicating that
HEK293 cells do exhibit Na+ -independent, saturable uridine uptake process(es). To
further determine what types of nucleoside transport processes are present in HEK293
cells, NBMPR was used to differentiate between functionally distinct subtypes of
equilibrative nucleoside transporters. Equilibrativre NBMPR-sensitive (es) nucleoside
transporters are usually potently inhibited by NBMPR at nanomolar concentrations,
whereas NBMPR-insensitive (ei) nucleoside transporters are less sensitive and can only
be inhibited by NBMPR at or above the micromolar concentrations (Griffith and Jarvis,
1996). Different concentrations of NBMPR were used to characterize the NBMPR
sensitivity of nucleoside transport processes of HEK 293 cells. The dose response curve
is shown in Fig. 4-16. The IC50 value for NBMPR was 0.9 nM, suggesting that HEK293
cells possess a single class of nucleoside transport process, which is NBMPR-sensitive
and sodium-independent. Endogenous uridine uptake by HEK293 cells was almost
completely blocked by 100 nM NBMPR, therefore, 100 nM NBMPR was included in the
subsequent transfection studies to abolish equilibrative NBMPR-sensitive nucleoside
transport processes.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1800
A
10 15 0 5
Time (min)
600
£ 400
O )
I
0
1 200
15 0 10 5
Time (min)
Fig. 4-15. Functional characterization of endogenous nucleoside transport in
untransfected HEK293 cells. Panel A, Time course of 5 juM 3 H-uridine uptake was
measured at different time points in the presence (■) and absence (□) of Na+ in HEK293
cells. Points represent mean ± s.e.m., n = 6. Panel B, Time course of Na+ -independent
3 H-uridine uptake was measured at different time points in the absence (■ ) and presence
(□) of 100 nM NBMPR, and in the absence (■) and presence ( • ) of 1 mM non
radiolabeled uridine in HEK293 cells. Points represent mean ± s.e.m., n = 6.
129
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
-3 - 2 - 1 0 1 2 3 4
Log [NBMPR (nM)]
Fig. 4-16. Effect of NBMPR on 5 pM 3H-uridine uptake by HEK293 cells. 3H~uridine
uptake was measured at 5 min as a function of NBMPR concentrations under Na+ -free
condition. IC50 values of NBMPR for endogenous uridine uptake in HEK293 cells was
0.9 nM, estimated by non-linear curve fitting algorithms using GraphPad Prism 3. Points
represent mean ± s.e.m., n = 6 .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
b) Time course
To determine the initial linear uptake range used for subsequent studies, time course of 5
jxM 3 H-uridine uptake was measured at a 15-min interval using HEK293 cells transiently
transfected with either rbENT2/pcDNA3 or rbENT2A/pcDNA3 constructs (Fig. 4-17A).
Significant increase (~5 fold) in the 3 H-uridine uptake was observed at 5 min in rbENT2-
transfected HEK293 cells over that in the mock-transfected HEK293 cells while there
was no significant difference between rbENT2A-mediated and mock-mediated uridine
uptake (Fig. 4-17B). Thus, five minutes of uptake time was determined for subsequent
kinetic and inhibition studies. With the addition of 1 mM unlabeled uridine to the donor
solution containing 5 fiM 3 H-uridine, uridine uptake mediated by rbENT2 was
completely abolished (Fig. 4- 17A). Significant inhibition of 3 H-uridine uptake by
rbENT2 was observed in the presence of 1 mM NBMPR (Fig. 4-17A).
c) NBMPR, dilazep, or dipyridamole sensitivity
To further characterize the NBMPR sensitivity of rbENT2-mediated uridine uptake in
transfected HEK293 cells, 3 H-uridine uptake was examined in the absence or presence of
graded concentrations of NBMPR, dilazep, or dipyridamole ranging from 1 nM to 100
/xM (Table 4-2). rbENT2-mediated uridine uptake was inhibited by less than 30% in the
presence of 1 fiM NBMPR and by only 65% with 100 fiM NBMPR. These data suggest
that the recombinant transporter was NBMPR-insensitive since uridine uptake was not
significantly inhibited by nanomolar NT-inhibitors, strongly confirming that the cDNA
insert encodes the NBMPR-insensitive equilibrative nucleoside transporter ei (rbENT2).
131
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
B
0
c
•I
200-
si
£ L m
150-
•S
.£
100-
5 O )
II
o
50 ■
E
a
0 ■
10 15
Time (min)
Fig. 4-17.3 H-uridine uptake by HEK293 cells transiently transfected with either
rbENT2/pcDNA3 or rbENT2A/pcDNA3 constructs. Panel A, Na+ -independent 3 H-
uridine (5 /tM) uptake was measured at different time points in the absence (■) and
presence (□) of 0.1 /iM NBMPR, and in the absence (■) and presence ( • ) of 1 mM
non-radiolabeled uridine in HEK293 cells transiently transfected with rbENT2/pcDNA3.
Points represent mean ± s.e.m., n = 6. Panel B, 3 H-uridine uptake was measured at 5 min
under Na+ -free condition. ■ and 0 represent rbENT2- and rbENT2A-mediated uridine
uptake, respectively. Asterisk(s) represent significant difference in uridine uptake
compared to mock (□ ) (*p < 0.05) by ANOVA.
132
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4-2. Inhibition of rbENT2-mediated uridine uptake by NBMPR, dipyridamole, and
dilazep. Na+ -independent 3 H-uridine (5 ptM) uptake was measured at different
concentrations of NBMPR, dipyridamole, and dilazep in HEK293 cells transiently
transfected with rbENT2/pcDNA3. Points represent mean ± s.e.m., n = 6.
Inhibitor
Uridine Uptake (% of control)
Concentration
rbETN2- Mock
Non - 100 ±5.5 100 ± 5.9
NBMPR 10 nM 61.3 ± 1.3“ 22.8 ± 3.0 *
1 fiM 74.8 ± 5.5“ 7.6 ±0.9 b
100 pM 36.4 ±8.4° 1.50 ±0.02*
Dipyridamole 10 nM 40.2 ± 7.0“ 19.6 ±1.2*
1 fiM 26.4 ± 1.1“ 2.0 ±0.1*
100 juM 3.9 ± 0.4“ 0.66 ±0.15*
Dilazep 10 nM 69.2 ± 2.8“ 72.4 ±9.4*
1 pM 52.3 ± 0.7“ 5.9 ±0.6*
100 pM 13.1 ± 2.0“ 0.98 ±0.09*
“ significantly different from rbENT2-mediated uridine uptake (p < 0.05).
* significantly different from mock-mediated uridine uptake (p < 0.05).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
d) E fflu x studies
To examine the efflux of radiolabeled uridine from rbENT2-transfectd HEK293 cells, 5
pCi/ml 3 H-uridine was preloaded into cells for 15 min, followed by washing cells with
fresh sodium-free buffer without 3 H-uridine. Radioactivity appearing in each 100 p 1
sample from the bathing fluid sampled was measured at predetermined time intervals.
One hundred microliters of the fresh sodium-free buffer was replenished to keep the
reservoir volume constant. As shown in Fig. 4-18, time-dependent efflux of radioactivity
was found in HEK293 cells expressing rbENT2 in the absence of uridine in the
extracellular medium, compared with that from mock-transfected HEK293cells
suggesting that rbENT2 has similar characteristics of bidirectional translocation of
nucleosides to those of both hENT2 and rENT2.
e) Concentration dependency
To determine the kinetic characteristics of uridine uptake by rbENT2-transfected
HEK293 cells, the concentration dependency of uridine uptake was investigated by
increasing the concentration of unlabeled uridine from 5 pM to 800 pM. Uridine uptake
by rbENT2-transfected HEK293cells saturated at higher concentrations of uridine. Fig. 4-
19 shows the total uptake data fit by nonlinear regression to the Michaelis-Menten
equation. The apparent Km and Vmu for uridine uptake by rbENT2-transfected HEK293
cells are 156.5 pM and 419.3 pmol/mg protein/5 min, respectively. The value was
similar to that characterized in rabbit tracheal epithelial cell monolayers grown on
permeable Clearwells® in primary culture (Table 4-1).
134
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 20 40 60 80 100 120
Time (min)
Fig. 4-18. Efflux of 3 H-uridine by rbENT2-transfected HEK293 cells after 3 H-uridine
preloading. rbENT2-transfected HEK293 cells (■) grown on the 6-well plates were
preloaded with 1 ml of the dosing solution containing 5 /xCi/ml 3 H-uridine for 15 min.
NBMPR (0.1 /xM) was present in all assays to inhibit endogenous es transport activity in
HEK293 cells. After incubation, the transfected cells were washed three times with
sodium-free buffer, and were replaced with 1 ml fresh sodium-free buffer without 3 H-
uridine. Radioactivity remaining in each 100 /xl of the receiver fluid sampled at a
predetermined time was measured. Control cells (□) were transfected with the pcDNA3
plasmid without an insert. Points represent mean ± s.e.m., n = 6.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 200 400 600 800
Concentration O t/M )
Fig. 4-19. Concentration dependency of 5 fiM 3 H-uridine uptake by HEK293 cells
transiently transfected with rbENT2/pcDNA3 construct. 3 H-uridine uptake was measured
at 5 min as a function of unlabeled uridine concentrations ranging from 0 to 800 fiM.
NBMPR (0.1 fiM) was present in all assays to inhibit endogenous es transport activity in
HEK293 cells. The rbENT2-mediated uridine uptake (—) was calculated as the
difference between the uptake data observed in HEK293 cells transfected with
rbENT2/pcDNA3 ( • ) and those in HEK293 cells transfected with pcDNA3 alone
(O).The apparent Km and V m a x for uridine uptake by rbENT2-transfected HEK293 cells
are 156.5 pM and 419.3 pmol/mg/5 min. Points represent mean ± s.e.m., n = 6.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
f) Substrate specificity
To detennine the substrate specificity of rbETN2, the effects of endogenous nucleoside
on 3 H-uridine uptake was examined by rbENT2-transfected HEK293 cells. At the time of
spiking transfected HEK293 cells with 3 H-uridine, competing agents all at 1 mM
concentration were added simultaneously and 5 min uptake studies were performed.
Theses agents include: guanosine, adenosine, inosine (purine nucleosides); uridine,
thymidine, cytidine (pyrimidine nucleosides). As shown in Fig. 4-20, a significant
inhibition of 3 H-uridine uptake (> 40%) was observed in the presence of endogenous
nucleosides, suggesting that rbETN2 were broadly selective for physiological purine and
pyrimidine nucleosides.
g) Coexpression
The spliced variant rbENT2A reveals no apparent transporter activity in transiently
transfected HEK293 cells although expressed when detected by an antibody directed
against a HA-tagged version of the protein. To further assess the functional role of
rbENT2A, attempts involved HEK293 cells transiently cotransfected with rbENT2 and
rbENT2A, where the uptake of model substrate uridine were measured. When co
transfected with rbENT2, the spliced variant rbENT2A appeared to negatively regulate
uridine uptake (Fig. 4-21 A) and efflux (Fig. 4-21B) in a dose-dependent manner.
However, being aware of limitations in heterologous expression systems (i.e.,
functionality observed in such a system may not necessarily exactly parallel that of in
vivo), the messages were expressed in another cell line duplicating the previous set of
studies.
137
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
■= 100
50
Control
Fig. 4-20. Concentration dependency of 3 H-uridine uptake by HEK293 cells transiently transfected with
rbENT2/pcDNA3 construct. Na+ -independent 3 H-uridine (5 pM) uptake was measured at 5 min in the
absence (control, open bars) and in the presence of 1 mM unlabeled nucleosides (solid bar). NBMPR (0.1
pM) was present in all assays to inhibit endogenous es transport activity in HEK293 cells. Results are
expressed as percentages of control values. G, guanosine; A, adenosine; C, cytidine; T, thymidine; U,
uridine; I inosine. Bars represent mean ± s.e.m., n = 6. *Significantly different from the control (p < 0.05).
138
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
300
® j= 200
Q . f f i
rbENT2 1 1 1 1
rbENT2A 0.5 1 2
pcDNA3 2 1.5 1 - 3
5
O 4
E
CL
X 3
3
3 =
U J 2
o
c
' ■ 5 1
1
3
0
40 80 100 120 0 20 60
Time (min)
Fig. 4-21. Functional characterization of 3 H-uridine uptake by HEK293 cells transiently
co-transfected with rbENT2/pcDNA3 and rbENT2A/pcDNA3. Panel A, Na+ -independent
3 H-uridine (5 fiM) uptake was determined at 5 min in HEK293 cells transiently co
transfected with different amounts (fig) of rbENT2/pcDNA3 and rbENT2A/pcDNA3 as
indicated. The total amount of DNA was kept constant by adding the pcDNA3 vector.
Panel B, 3 H-uridine efflux studies were performed in HEK293 cells transiently
transfected with rbENT2 (• ) , rbENT2A (O), or rbENT2/2A (■, rbENT2:rbENT2A =
1:1 (jxg)) grown on the 6-well plates as described in Fig. 4-18. Mock-transfected cells
(□) were transfected with the pcDNA3 plasmid without an insert. *Significantly different
from the control (p < 0.05). Points represent mean ± s.e.m., n = 6.
139
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.4.2. In MDCK heterologous expression system
a) Validation of MDCK heterologous expression system
Using 3 H-uridine as a substrate, background endogenous uridine uptake in confluent
MDCK cell monolayers grown in 12-well cluster plates was examined next. To
determine whether uridine uptake by MDCK cells was mediated by saturable Na+ -
independent transport processes, time course of 5 fiM 3 H-uridine uptake under Na+ -free
conditions was measured in the presence and absence of 1 mM unlabeled uridine (Fig. 4-
22A). Endogenous uridine uptake in MDCK cells under Na+ -free conditions was
significantly abolished in the presence of 1 mM unlabeled uridine, indicating that MDCK
cells exhibit Na+ -independent, saturable uridine uptake processes). To further determine
the specific type of Na+ -independent nucleoside transport processes present in MDCK
cells, NBMPR was used to differentiate functional distinction between es and ei subtypes
of equilibrative nucleoside transporters (Griffith and Jarvis, 1996). Fig. 4-22B showed a
dose-response curve of NBMPR, characteristic of the NBMPR sensitivity of Na+ -
independent, equilibrative nucleoside transport processes in untransfected MDCK cells.
The IC50 value for NBMPR was 24.3 nM, suggesting the presence of an NBMPR-
sensitive nucleoside transport process in native MDCK cells. Thus, all the subsequent
functional studies in transfected MDCK cells were carried out in the presence of 100 nM
NBMPR in SFR. These conditions exclude Na+ -dependent and Na+ -independent es types
of nucleoside transport processes.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 5 10 15
Time (min)
100-
50-
3 4 2 ■ 1 0 2 5
Log [NBMPR] fiM)
Fig. 4-22. Functional characterization of endogenous equilibrative nucleoside transport in
untransfected MDCK cells. Panel A: Time course of 5 fiM 3 H-uridine uptake by MDCK
cells under Na+ -free condition. 3 H-uridine uptake was measured at 15-min intervals in the
absence (□) and presence (■) of 1 mM unlabeled uridine. Points represent mean ±
s.e.m., n = 6. Panel B: Effect of NBMPR on 5 fiM 3 H-uridine uptake by MDCK cells
under Na+ -free condition. 3 H-uridine uptake was measured at 5 min as a function of
NBMPR concentrations. IC50 = 24.3 ± 1.1 nM was estimated by non-linear curve fitting
algorithms using GraphPad Prism 3. Points represent mean ± s.e.m., n = 6.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
b) E ffect o f ep ito p e tagging o n m em b ra n e ex p ressio n an d fu n ctio n a lity
In order to determine whether the addition of HA tag at the N-terminus of rbENT2 and
rbENT2A affected their functionality, we examined the uridine transport afforded by
pSG5.HA/rbENT2 and pSG5.HA/rbENT2A fusion constructs (see Methods 5.1.1.)
transiently transfected in MDCK cells. Two days after transfection with
pSG5.HA/rbENT2 or pSG5.HA/rbENT2A, MDCK cells displayed Na+ -independent
uridine uptake ~3-fold higher than the mock-transfected MDCK cells (Fig. 4-23).
Furthermore, pSG5.HA/rbENT2 or pSG5.HA/rbENT2A-mediated uridine uptake was
significantly abolished in the presence of 1 mM unlabeled uridine (data not shown),
indicating that HA-tagged recombinant proteins were functionally similar to their wild
type counterparts.
c) C on cen tration d ep en d en cy
Expression constructs harboring rbENT2 or rbETN2A were individually transfected into
MDCK cells and the kinetic parameters and pharmacological profiles of uridine uptake
were examined, respectively. Time course of 5 jiM 3 H-uridine uptake by rbENT2 or
rbENT2A-transfected MDCK cells was linear for 15 min (data not shown). Hence, 5 min
uptake studies were performed to determine the functional and pharmacological
classification of the heterologously expressed nucleoside transporters. rbENT2 had a Km
for uridine of 272.8 ± 31.9 fiM and V m a * of 0.6 ±0.1 nmol/mg protein/5min (Fig. 4-24A).
By contrast, rbENT2A displayed a higher affinity (Km = 41.6 ± 9.8 fiM and lower
capacity (V m a X = 1.3 ±0.1 nmol/mg protein/5 min) compared to rbENT2 (Fig. 4-24B).
142
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pSG5.HA/rbENT2 pSG5.HA/rbENT2A Mock
Fig. 4-23.3 H-uridine uptake by MDCK cells transiently transfected with
pSG5.HA/rbENT2, pSG5.HA/rbENT2A constructs (closed bars), and pSG5.HA parent
vector alone (open bar). 3 H-uridine uptake (5 pM) was measured at 5 min in the presence
and absence of 1 mM unlabeled uridine. The rbENT2- or rbENT2A-mediated uridine
uptake was corrected by the uptake data observed in the presence of 1 mM unlabeled
uridine. Data represent mean ± s.e.m., n = 6.
143
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
3
2
800 200 400 600
0.5'
0 200 400 600 800
Uridine Concentration fcsM )
Fig. 4-24. Concentration dependency of 5 jxM 3 H-uridine uptake by MDCK cells
transiently transfected with rbENT2/pcDNA3 {Panel A) and rbENT2A/pcDNA3 {Panel
B) constructs. 3 H-uridine uptake was measured at 5 min as a function of unlabeled uridine
concentrations ranging from 0 to 800 fzM . NBMPR (0.1 jxM) was present in all assays to
inhibit endogenous es-type transport activity in these MDCK cells. The rbENT2- or
rbENT2A-mediated uridine uptake (—) was calculated as the difference between the
uptake data observed in MDCK cells transfected with rbENT2/pcDNA3 ( • ) or
rbENT2A/pcDNA3 (O) and those in MDCK cells transfected with pcDNA3 alone (□),
as shown in Panels A and B, respectively. Points represent mean ± s.e.m., n = 6.
144
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
d) NBMPR sensitivity
NBMPR inhibits 3 H-uridine uptake mediated by rbENT2A with an IC50 of 0.1 ptM, but
not rbENT2 (IC50 = 200 fiM). At NBMPR concentrations where rbENT2 still retained
100% of its activity at 1 fiM, rbENT2A activity was completely inhibited (Fig. 4-25).
These data suggested that the recombinant rbENT2 was NBMPR-insensitive. However,
the pharmacological profile of rbENT2A towards NBMPR was “es-type-like”. The
rbENT2 data herein are comparable to the pharmacological profile of NBMPR
characterized in our recent studies of basolateral uridine uptake in primary cultured
RTEC (IC2 5 = 0.2 pM, IC75 = 270 fiM).
e) Substrate selectivity
To determine the substrate specificity of rbETN2 and rbENT2A, the effects of
endogenous nucleosides on 3 H-uridine uptake by rbENT2- and rbENT2A-transfected
MDCK cells were examined. As shown in Fig. 4-26,3 H-uridine uptake by these two
variants was significantly inhibited by 40-90% in the presence of unlabeled nucleosides
including guanosine, adenosine, inosine (purine nucleosides); uridine, thymidine, cytidine
(pyrimidine nucleosides); and hypoxanthine (purine nucleobase, a known ENT2
substrate). Thus, rbETN2 and rbENT2A were broadly selective for physiological purine
and pyrimidine nucleosides.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
75" O
+ - »
c
o
o
«t—
o
''S
50-
i k
25"
4 0 1 2 3 -4 3 2 1
Log [NBMPR ( jjM )\
Fig. 4-25. Effect of NBMPR on 5 fiM 3 H-uridine uptake by MDCK cells transiently
transfected with rbENT2/pcDNA3 and rbENT2A/pcDNA3 constructs. 3 H-uridine uptake
was measured at 5 min as a function of NBMPR concentrations under Na+ -free condition.
NBMPR (0.1 /xM) was present in all assays to inhibit endogenous es-type transport
activity in these MDCK cells. IC50 values of NBMPR for rbENT2- ( • ) and rbENT2A-
mediated uridine uptake (▲) were 200.3 ± 7.8 jiM and 0.1 ± 0.004 piM, respectively,
estimated by non-linear curve fitting algorithms using GraphPad Prism 3. Points
represent mean ± s.e.m., n = 6.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Adenosine
G uanosine
Inosine
Cytidine
Thymidine
Uridine
Hypoxanthine
0 20 50 75 100
% of Control
Fig. 4-26. Substrate selectivity of various nucleosides on 5 pM 3 H-uridine uptake by
MDCK cells transiently transfected with rbENT2/pcDNA3 and rbENT2A/pcDNA3
constructs. 3 H-uridine uptake was measured at 5 min in the absence (control) and in the
presence of 1 mM unlabeled nucleosides under Na+ -free condition. Closed bars and open
bars represent rbENT2- and rbENT2A-mediated uridine uptake, respectively. NBMPR
(0.1 pM) was present in all assays to inhibit endogenous es-type transport activity in
these MDCK cells. Asterisk(s) represent significant decrease in uridine uptake compared
to control (*p < 0.01) by one way-ANOVA of variances, followed by Tukey’s procedure
for contrasting multiple group means. Each data point represents mean ± s.e.m., n = 6.
2.5. Functionality, expression, and subcellular distribution of the C-terminal
147
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
truncated rabbit ENT2
Based on predictions derived from hydropathy analysis and results from studies of
orientation of these transporters in cell membrane, it was hypothesized that the C-termini
(from amino acid residues 290 to 456/415, Fig. 4-5) of rbENT2/2A play a determining
role in functional differences. To examine the role of this fragment of rbETN2/2A in
functionality, expression, and subcellular distribution, the C-terminal amino acid residues
(defined as starting from the acid 290 to 456 corresponding to the rbENT2 protein) after
the point of alternative splicing where rbENT2 and rbENT2A begin to structurally
diverge.
2.5.1. Functionality and expression
Truncation of the C-terminus resulted in a sharp decrease in transport activity in
transiently transfected MDCK cells compared to wild type rbENT2. The abolished
transport activity of the C-terminal truncated mutant was not dependent on methods of
transfection (LipofectAMINE™ and electroporation) or types of cell lines (MDCK and
HEK293). To examine whether the decrease in transport activity of the C-terminal
truncated mutant resulted from impaired protein synthesis, Western blot analysis was
carried out. Similar levels of the HA-tagged recombinant rbENT2 and its truncated
counterpart were detected with apparent molecular mass of 49 and 31 kDa (Fig. 4-27A),
respectively, indicating that lack of transport activity was not due to insufficient protein
biosynthesis.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 4-27. Detection ofHA-tagged, truncated rbENT2 proteins expressed in MDCK cells
by Western blot (Panel A) and confocal fluorescence microscopy (Panel B and C). A:
Cell membrane proteins were prepared from MDCK cells transiently transfected with
HA-tagged, truncated rbENT2 DNA construct. Samples (20 pg/lane) were subjected to
10% SDS-PAGE gel and transferred to a nitrocellulose membrane. Membrane blot was
incubated with mouse anti-HA monoclonal antibody [1:1000 (v/v) dilution], followed by
wash and incubation with a peroxidase-conjugated AffiniPure donkey anti-mouse IgG at
a dilution of 1:50,000 (v/v). The resulting signals were visualized on x-ray films by
enhanced chemiluminescence (ECL). The apparent molecular weights of the protein
bands were indicated with arrows on the right, and the positions of molecular-mass
markers (in kDa) are indicated on the left. B. Confocal fluorescence micrographs (xy
plane) of MDCK cells, transiently transfected with HA-tagged, truncated rbENT2 DNA
construct, were acquired in 1.0-gm increments. HA-tagged, truncated rbENT2
fluorescence is shown in green, and tight junctions, stained for ZO-1, are shown in red.
C. Confocal fluorescence micrographs (xz plane) showing distribution of truncated
rbENT2 protein along the apical-basal axis. Scale bar is 10 pm. AP, apical membrane;
BM, basal membrane.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.5.2. Subcellular distribution
To determine whether the “non-functional” C-terminal truncated mutant failed to target
to the cell membrane surface, the N-terminal HA epitope was immunostained and
visualized along with ZO-1 using confocal microscopy. Although a significant amount of
cytoplasmic staining for the HA-tag was observed, there were distinct and diffuse sties of
membrane localization as well (Fig. 4-27B and 4-27C). These data taken together
indicate that the C-terminal truncated form of rbENT2/2A was properly synthesized and
sorted into the plasma membrane.
2.6. Structural validation
2.6.1. Verification of membrane orientation
Computer-derived hydropathy profiles and comparative protein analysis serve as
a useful starting point in developing working models for the topology of novel membrane
proteins. Such analyses, however, can lead to the formulation of different topological
models as in the case of the equilibrative nucleoside transporters. Experimental data is
needed to distinguish between the different models and identify important domains that
are present on each side of the membrane for structure-function analysis. Initial
hydropathy analysis of the rbENT2 and its splice variant rbENT2 A proteins cloned from
rabbit trachea suggested that the rbENT2 protein possessed either 10 or 11 TMDs,
whereas the rbENT2A protein was predicted to have either 6 or 7 TMDs (Fig. 4-28).
Each of the proposed models has two possible membrane orientations with N-terminus
residing intra- or extra-cellularly. The number of putative TMDs considerably varied
depending upon the secondary structure prediction algorithms (TMPred
150
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
rbENT2 (10 TMDs)
rbENT2A (7 TMDs)
' x S
- -COOH
COOH
rbENT2 (11 TMDs)
rbENT2A (6 TMDs)
HO OO - _ „ - .
-5,-cX
COOH
Fig. 4-28. The proposed topological models of rbENT2 and rbENT2A. Potential
membrane-spanning a-helices are numbered. The putative iV-glycosylation sites and the
potential PKA phosphorylation site are indicated by arrow (— ► ) and asterisks (★),
respectively. A novel C-terminal tail of rbENT2A that is different from rbENT2 is
indicated by dashed line (— ).
151
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(Hofmann and Stoffel, 1993) and TMHMM (Sonnhammer et al., 1998)) employed to
predict TMDs. In these studies, suspensions of dissociated MDCK cells transiently
transfected with rbENT2/2A under two conditions: cell intact cells (i.e.,
nonpermeabilized (NP) condition) and permeabilized (P) cells. A cell permeabilizing
agent, 0.1% TX-100, were used for P conditions. To establish the validity of these two
conditions, a controlled study was performed where the P and NP cells were tested by
using an antibody directed against y-adaptin, a cytoplasmic marker. Since y-adaptin is
located inside the cells, the antibody will find its specific epitope only if in the plasma
membranes are permeabilized enough to allow free access of the antibody to the cytosol
(Fig. 4-29). This method has been used conjunction with immunofluorescence to
determine the membrane topology of numerous transporter proteins including hPepTl
(Covitz et al., 1998), cytidine 5’-monophosphate (CMP)-sialic acid transporter (Eckhardt
et al., 1999), concentrative Na+ -nucleoside transporter rCNTl (Hamilton et al., 2001).
MDCK cells grown on chamber slides were transiently transfected with
pSG5.HA/rbENT2 or pSG5.HA/rbENT2A (HA epitope tagged on the N-terminus of
proteins) for the determination of N-terminal location. In the case of an intracellularly
located N-terminus, the NP condition should not have resulted in a fluorescence signal.
However, if the N-terminus was extracellularly located, both P and NP conditions should
allow for the anti-HA antibody to reach the N-terminus. Florescent cells were only
observed under the permeabilized condition, confirming the computer-predicted
intracellularly located N-terminus for both rbENT2 and rbENT2A (Fig. 4-30). Following
the elucidation of N-terminal orientation, corresponding C-terminal locations for both
equilibrative nucleoside transporters were deduced. For the 11 TMD model of rbENT2,
152
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 4-29. Validation of cell plasma membrane intact nonpermeabilized (NP) condition, and permeabilized
(P) condition. Transiently transfected MDCK cells were incubated in the presence and absence of a cell
permeabilizing agent, 0.1% Triton-X 100 (TX-100). The P (PanelA) and NP (Panel B) conditions were
then tested by using an antibody directed against y-adaptin, a cytoplasmic marker. Since y-adaptin is
located inside the cells, the antibody would find its specific epitope only in the permeabilized environment.
153
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
N-HA
rb£NT2
N-HA
rbENT2A
Fig. 4-30. Orientation of the N-terminus of rbENT2 and rbENT2A. MDCK cells were
grown on Lab-Tek® II Chamber Slide™ and transiently transfected with the N-terminal
HA-tagged rbENT2 (Panels A and B) or rbENT2A (Panels C and D) DNA constructs
using LipofectAMINE™. Two days after transfection, cells were fixed in 3.7%
formaldehyde and incubated in the presence or absence of 0.1% TX-100. Permeabilized
(Panels A and C) or non-permeabilized (Panels B and D) cells were then incubated with
mouse anti-HA monoclonal antibody [1:1000 (v/v) dilution], followed by wash and
incubation with a FITC-conjugated AffiniPure donkey anti-mouse IgG at a dilution of
1:50,000 (v/v). Cells were then subjected to indirect immunofluorescence. If the N-
terminus was extracellularly located, both P and NP conditions should have allowed for
the anti-HA antibody to reach the epitope tag. However, if the N-terminus is
intracellularly located, under the NP condition a green signal (from FITC) would not be
observed.
154
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the C-terminus has to be located extracellularly. However, the predicted 6 TMD model
for the splice variant rbENT2A would have an intracellularly located C-terminus, in
contrast to the 7 TMD model with an extracellularly located C-terminus.
To further validate the C-terminal location for rbENT2 and rbENT2A, MDCK
cells grown on chamber slides were transiently transfected with rbENT2/pMH or
rbENT2A/pMH harboring HA epitope tagged on the C-terminus of proteins. When these
cells permeabilized with TX-100, they showed similar to staining patterns for HA epitope
to those for the nonpermeabilized cells, suggesting that the C-termini of both rbENT2 and
rbENT2A were localized extracellularly (Fig. 4-31). In control MDCK cells transfectd
with parent vector only (pSG5.HA or pMH), no immunoreactive signals were observed.
These results taken together support a structural model for rbENT2A with 7 TMDs, an
intracellular N-terminus, and an extracellular C-terminus. In addition, rbENT2 has 4 extra
TMDs giving it a total of 11, sharing the same topology for the 1-7 TMDs as that for
rbETN2A.
2.6.2. Lys-C digestion
PeptideMass™ was employed to predict the molecular mass of fragments of rbENT2/2A
cleaved with Lys-C. PeptideMass™ identified a complete set of lysine residues located
along the entire length of the predicted polypeptide chains for rbETN2/2A (Fig. 4-32A
and 4-33A). Accessibility of the identified lysine sites to the Lys-C enzyme under
permeabilized or cell-intact conditions, differentiating the cleavage sites located on
putative intracellular inter-TMD loops from those on extracellular ones, was studied.
Furthermore, differences in primary amino acid sequence after the alternative splicing
155
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
point marked by key lysine resides, that are absent in rbENT2A, were
P NP
C-HA
rbENT2
C-HA
rbENT2A
Fig. 4-31. Orientation of the C-terminus of rbENT2 and rbENT2A. MDCK cells were
grown on Lab-Tek® II Chamber Slide™ and transiently transfected with the C-terminal
HA-tagged rbENT2 {Panels A and B) or rbENT2A {Panels C and D) DNA constructs
using LipofectAMINE™. Two days after transfection, cells were fixed in 3.7%
formaldehyde and incubated in the presence or absence ofO.l % TX-100. Permeabilized
{Panels A and C) or non-permeabilized {Panels B and D) cells were then incubated with
mouse anti-HA monoclonal antibody [1:1000 (v/v) dilution], followed by wash and
incubation with a FITC-conjugated AffiniPure donkey anti-mouse IgG at a dilution of
1:50,000 (v/v). Thereafter cells were subjected to indirect immunofluorescence. If the C-
terminus was intracellular, under the NP condition a green signal should not have been
observed. However, if the C-terminus is extracellular, both P and NP conditions should
allow for the anti-HA antibody to reach the epitope tag.
156
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
B 1 2
4 8 .4 — > « _ _ | ^ Molecular Weight (kDa)
«
Cleavage site
Expected NP P
4-------37.1
4-------33.7
Lys^-Val11 8 1 2 .7 1 4 .3 1 4 .3
4---------30.6
4---------27.1
Lys?2 4 -Lys?25 2 4 .0 - -
Lys?3 7 -Ala?38 2 5 .5 - 2 4 .1
L y s^ A sr? 47 2 6 .5 - 2 7 .1
lllliiilli
L y s^ A la 2 5 6 2 7 .4 - -
iliiiiiiiiiii
Lysf^-Glu267 2 8 .6
- 3 0 .6
IlllllllllllllllllllllB L ys^-lle290
3 1 .2
- 3 3 .7
l l l l l l l l l l
Lys?23-Trp?24
3 4 .8
3 7 .1 3 7 .1
. 4---------14.3
LygMS-Ala454 4 9 .7 4 8 .4
Lys-C + +
Fig. 4-32. Digestion with Lys-C protease of the HA-tagged rbENT2 transiently
transfected in MDCK cells grown on 6-well plates. A: Potential Lys-C protease digestion
maps of the 50-kDa HA-tagged rbENT2 transporter. The molecular weight of the
theoretical proteolytic fragments has been calculated by PeptideMass™ based on the
rbENT2 primary sequence deduced from its cDNA sequence (C, Expected). Extracellular
cleavage sites are marked by ( - ^ - ) that hypothetically will be exposed to lys-C in the
nonpermeabilized (NP) condition, whereas the intracellular ones (marked by » -^ ) will
not. However, all marked sites are supposed to be cleaved when the cells are treated with
TX-100 (i.e., permeabilized (P) conditions). B: Cells were harvested by using
157
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Cellstripper™ (Cellgro®, Mediatech, Inc.). The harvested cells were washed three times
with lx PBS by centrifugation and resuspended with 100 mM Tris-HCl pH 8.5 in the
presence (Lanes 2-3, P condition) or absence (Lane 1, NP condition) of 0.1% TX-100.
Lys-C was used at 0.02 jug/assay (20 ju .1 ) for 4 hr treatment (Lanes 1-2). Each digestion
reaction was inactivated by addition of Protease Inhibitor Cocktail for General Use
(Sigma), at a recommended 1 ml/mg proteases, followed by boiling for 2-3 min. Digests
were fractionated in 1.5-mm thick SDS-PAGE gels (15%). Gels were then either
transferred to nitrocellulose membranes or stained by GelCode® Blue Stain Reagent
(Pierce). Molecular weight markers are indicated on the right. Each assay is
representative of at least three separate experiments. C: Three digested fragments under
the NP condition, and six fragments under the P condition were observed. Compared to
Panel A for the predicted TMDs, all the expected bands under NP condition were
observed, but under the P condition the parent band was missing which is probably due to
complete cleavage.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
B
45.9-
35.4-
32.1 •
24.0-
Lys-C
TX-100
1 2 3
C
Cleavage site
Molecular Mass (kDa)
Expected NP P
m Lys11 7 -Val118
12.7
S M i
L ys^ -L ys225 24.0 24.0 -
: I : ; : : : : : : : : : : : : : : : : : : : : : : : : : : : :
i_ys237-Ala238 25.5 - -
............
Lys246-Asn247 26.5 - -
Lys25S -AIa256 27.4 - -
+ +
Lys266-Glu267 28.6 32.1 -
+ +
Lys289-Ser290 31.1 35.4 -
Fig. 4-33. Digestion with Lys-C protease of the HA-tagged rbENT2A transiently
transfected in MDCK cells grown on 6-well plates. A: Potential Lys-C protease digestion
maps of the 44-kDa HA-tagged rbENT2 transporter. The molecular weight of the
theoretical proteolytic fragments has been calculated by PeptideMass™ based on the
rbENT2A primary sequence deduced from its cDNA sequence (C, Expected).
Extracellular cleavage sites are marked by (— ^ ) that hypothetically will be exposed to
lys-C in the nonpermeabilized (NP) condition, whereas the intracellular ones marked by
(“■ ^ ) will not. However, all marked sites are supposed to be cleaved when the cells are
treated with TX-100 (i.e., permeabilized (P) conditions). B: Cells were harvested by using
Cellstripper™ (Cellgro®, Mediatech, Inc.). The harvested cells were washed three times
159
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
with lx PBS by centrifugation and resuspended with 100 mM Tris-HCl pH 8.5 in the
presence (Lanes 1 and 3, P condition) or absence (Lane 2, NP condition) of 0.1% TX-
100. Lys-C was used at 0.02 /rg/assay (20 |il) for 4 hr treatment (Lanes 1-2). Each
digestion reaction was inactivated by addition of Protease Inhibitor Cocktail for General
Use (Sigma), at a recommended 1 ml/mg proteases, followed by boiling for 2-3 min.
Digests were fractionated on 1.5-mm thick SDS-PAGE gels (15%). Gels were then either
transferred to nitrocellulose membranes or stained by GelCode® Blue Stain Reagent
(Pierce). Molecular weight markers are indicated on the right. Each assay is
representative of at least three distinct experiments. C: Compared to Panel A for the
predicted TMDs, no fragments under the NP condition (Lane 2), but only three fragments
under the P variation (Lane 1) were observed.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
evaluated. Extracellular cleavage points are assumed to be accessible to Lys-C in the non-
permeabilized condition, whereas the intracellular ones are inaccessible. In other words,
all the potential cleavage sites would be accessible to Lys-C, when cells are treated with a
permeabilizing agent.
For rbENT2 three fragments (Fig. 32B) under NP condition and 6 fragments
under P condition were observed. Lane 1 (Fig. 32B) on the SDS-PAGE showed all the
expected bands predicted by PeptideMass™. In Lane 2 (Fig. 32B), the parent band was
missing, which is probably due to over-exposure to Lys-C that results in complete
cleavage. Additionally, two bands that would arise from the cleavage at specific lysine
residues (Lys2 2 4 , Lys2 5 5 ) located on the large intracellular loop connecting TMDs 6 and 7
were absent (Fig. 32B, Lane 2).
According to PeptideMass™ prediction, there is only one possible extracellular
lysine cleavage site for rbENT2A. To our surprises, no fragments under NP condition
were observed for rbENT2A (Fig. 33B). Three fragments under the Lane 1 (Fig. 33B)
variation were visualized under SDS-PAGE. However, an expected band at 12.7 kDa
from the predicted lysine residue located on the 2n d extracellular loop was absent in Lane
2 (Fig. 33C). Since this cleavage site (Lys1 1 7 ) is identical between rbENT2 and
rbENT2A, the 12.7 kDa signal was observed in the case of rbENT2. By contrast, in Lane
1 (Fig. 33B) three bands that would arise from specific lysine residues (Lys2 4 6 , Lys2 5 5 ,
and Lys2 6 6 ) located on the large intracellular loop connecting TMDs 6 and 7 were
missing.
161
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.6.3. PNGase F treatment
Both rbENT2 and rbENT2A isoforms contain the same potential A-linked glycosylation
sites (Asn4 7 , Asn5 6 ) located on the first extracellular loop between TMDs 1 and 2,
according to the prediction based on a post-translational modification algorithm
( .PPsearchm , http://www2.ebi.ac.uk/ppsearch). Blots of membranes obtained from SDS-
PAGE of total lysates from MDCK cell preparations expressing rbENT2/2A exhibited
two consistent bands. The larger bands, at 49-50 kDa for rbENT2 and 45-46 kDa for
rbENT2A, correspond to the predicted molecular size of the HA-tagged transporters (Fig.
4-9). The slightly smaller bands had molecular sizes of 44-45 kDa and 40-41 kDa in blots
of rbENT2 and rbENT2A, respectively (Fig. 4-34). Since cell lysates may also containe
intracellular organellar membranes, in addition to the cell plasma membranes, we
hypothesized that the smaller bands reflected a population of rbENT2/2A which was
synthesized by MDCK cells but not subjected yet to post-translational modifications, i.e.,
glycosylation. In any event, PNGase-F treatment of cell lysates from MDCK cells
expressing rbENT2/2A completely eliminated the larger bands (at -50 kDa and -46 kDa
for rbENT2 and rbENT2A, respectively), indicating -45 kDa and -41 kDa of
deglycosylated products in rbENT2 and rbENT2A, respectively.
3. F u n c t io n a l a n d M o l e c u l a r M o d u la t io n o f B a s o l a t e r a l ENTs
3.1. Pharmacological modulation of basolateral nucleoside transport activity
In an attempt to reveal strategies in enhancing the therapeutic index of nucleoside drugs,
possible mechanisms of modulation of uridine uptake from the basolateral
fluid in primary cultured RTEC were studied using various pharmacological agents
162
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A rbENT2
B rbENT2A
49 kDa
45 kDa
£ 4 ■ - 4 4 kDa
4
PN G ase-F +
PN G ase-F +
Fig. 4-34. Deglycosylation of the HA-tagged rbENT2 (Panel A) and rbENT2A (Panel B)
with PNGase F. The transfected MDCK cells were denatured at 100°C for 10 min and
treated with PNGase F (1000 units) at 37°C for 2 hr. These cell lysates were fractionated
on 1.5-mm thick SDS-PAGE gels (8%). Gels were then either transferred to
nitrocellulose membranes as described above on Western blot analysis, or stained by
GelCode® Blue Stain Reagent (Pierce). The potential N-glycosylation sites are identical
in both rbENT2 isoforms. After treatment with PNGase F, slight decreases of ~5 kDa in
both rbENT2 (Panel A) and rbENT2A (Panel B) were observed, indicating some
deglycosylation was conferred through asparagines residues.
163
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(Table 4-3). When uptake was assessed with cells incubated for 15 min with an adenylyl
cyclase activator, forskolin (10 fiM), and a calcium ionophore, A23187 (50 fiM), and
A23187 decreased basolateral 3 H-uridine uptake by 75%, whereas forskolin significantly
increased it by 50%. When A23187 and forskolin were co-administered, no additive
effect was observed. The fairly specific PKC inhibitor, tamoxifen (Rowlands et al,
1995), at 10 fiM after a 30 min pretreatment did not reverse the effect of A23187.
Following a 30 min pretreatment with a PKA inhibitor, H89 (Penn et al., 1999), at 10
fiM, the stimulation of uridine uptake by forskolin was completely blocked (Table 4-3).
Pretreatment for 15-60 min with a microtubule depolymerizing agent, colchicine, at 10
fiM completely reversed the decrease in uridine uptake elicited by A23187, while no
effect on forskolin stimulation of uptake was observed (Table 4-3). Treatment of
colchicine alone had insignificant effect on basolateral 3 H-uridine uptake.
In Figure 4-35A, kinetic analysis of uridine uptake studied in the presence of
A23187 is shown. A23187 significantly decreased Jm ax (356.9 ± 23.4 vs.183.0 ± 10.0
pmol/mg protein/min for control vs. A23187, respectively, n = 6, p < 0.05), but no
significant change in Km was observed (210.2 ± 34.3 vs. 220.3 ± 8.0 fiM for control
vs. A23187, respectively, n = 6 ,p> 0.05). Forskolin-mediated alteration in nucleoside
uptake significantly increased the affinity (Km of 210.2 ± 34.3 vs. 110.7 ± 7.0 fiM for
control vs. forskolin, respectively, n = 6 ,p < 0.05), while decreasing the maximal rate
(J m a x of 356.9 ± 23.4 vs.181.9 ± 3.9 pmol/mg protein/min for control vs. forskolin,
respectively, n = 6, p < 0.05) for basolateral uptake of uridine (Figure 4-35B and Table 4-
4).
164
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4-3. Effect of A23187 and forskolin on kinetic properties of basolateral 3 H-uridine
uptake in RTEC monolayers. Data are expressed as mean ± s.e.m., n = 6.
Km (/xM) Jm ax (nmol/mg protein/min)
Control 210.2 ± 34.3 0.36 ± 0.02
A23187 220.3 ± 8.0 0.18 ±0.01*
Forskolin 110.7 ±7.0“ 0.18 ±0.00*
“ significantly different from control (p < 0.05).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
300i
2 0 0 -
1 0 0 -
a ,
a>
|
o
E 200 0 400 600 800
c l
& 3001
( Q
a
15
v 200-
c
2
^ 100 ■
0 200 400 600 800
Uridine Cone. (jjM )
Fig. 4-35. Concentration dependency of 1 min 3 H-uridine (10 jaCi/ml) uptake from the
basolateral fluid in primary cultured RTEC monolayers after treatment with 50 fiM
A23187 for 15 min (Panel A), and 10 fiM forskolin for 15 min (Panel B) as a function of
unlabeled uridine concentrations ranging from 0 to 800 pM at 37°C (■) and 4°C (□) in
the SFR solution. Carrier-mediated uptake (—) observed with forskolin and A23187
pretreatment was compared with non-specific absorption and with that of control (—)
observed at 37°C. The values of uptake for untreated controls have been corrected for
non-specific binding. Each data point represents mean ± s.e.m., n = 6.
166
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4-4. Basolateral 3 H-uridine (10 piCi/ml) uptake for 1 min by tracheocyte
monolayers after a 15 min pretreatment with 10 /xM forskolin; 10 /xM H89 for 30 min
pretreatment, followed by 15 min with 10 pM forskolin; 10 pM colchicine for 2 hrs
followed by 15 min incubation with 10 pM forskolin; 15 min pretreatment with 50 pM
A23187; 10 pM tamoxifen for 30 min followed by 15 min with 50 pM A23187; 10 /xM
colchicine for 2 hrs followed by 15 min incubation with 50 pM A23187; a combination
of 10 /xM forskolin and 50 juM A23187 for 15 min in SFR solution. Each data point
represents mean ± s.e.m., n = 6.
Treatment Control (%)
Forskolin
Alone 51 ± l l a t
Pretreatment with H89 10 ±12 t
Pretreatment with colchicine 49 ± 10*T
A23187
Alone 70 ±15feI
Pretreatment with tamoxifen 74 ± 9b 4 -
Pretreatment with colchicine 12 ± 10 I
Forskolin +A23187 71 ± l l H
a is significantly different from control ip < 0.05).
fe is significantly different from control ip < 0.01).
Tincrease
■ I decrease
Control uptake rate was 37.5 ± 1.1 fmol/mg protein/min, n = 6.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.2. Transcriptional and post-translational modulation of ENTs
3.2.1. Extent of depletion on intracellular nucleoside pool following nucleoside
deprivation
HPLC was used to evaluate the extent of intracellular nucleoside pool after combined
treatment of de novo biosynthesis inhibition (MTX, MPA, and 5-FU) and deprivation of
precursor amino acids (Asn, Asp, Glu, Gin, and Gly). The area under curve containing
peaks is used as the content of total nucleosides. Total nucleosides after 72 hr treatment
was decreased (Fig. 4-36). HPLC analysis revealed a 70-75% depletion of total
nucleosides compared to control. Comparisons are being made between groups
comprised of equal numbers of cells (i.e., equal protein content) (p < 0.05).
3.2.2. Time course of rbENT2/2A mRNA levels following nucleoside deprivation
Nucleoside deprivation caused a 2-3 fold induction of both rbENT2/2A mRNAs (Fig. 4-
37). However, a lesser induction of rbENT2 mRNA than that of rbENT2A was observed
(Fig. 4-37). Maximal induction was observed at 72 hr for both rbENT2 and rbENT2A
mRNA levels (Fig. 4-37). Re-introducing metabolically inhibited/nucleoside depleted
cells to PC-1 medium restored the levels of rbENT2/2A mRNA levels to the original
levels (Fig. 4-37). Levels of GAPDH mRNA remained unchanged in response to either
nucleoside deprivation or re-instatement to PC-1 medium.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A
0.04
Control
B
100 fiM
tn
0 .0 3
inhibitors
8100
200//M
inhibitors
o
o
C 0 .0 2
C O
42
0.00
o >
>
o
0
Control 10° / /M 200//M
*
inhibitors inhibitors
Fig. 4-36. Extent of depletion on intracellular nucleoside pool following nucleoside
deprivation. Primary cultured RTEC cells were grown in PC-1 medium. On day-4, cells
were exposed to MTX, MPA, and 5-FU at 100 pM {hatched bar) or 200 fiM {closed bar)
in the custom-made minimal essential medium lacking amino acids, Asn, Asp, Glu, Gin,
and Gly. The treatment was continued for three days, with daily changes of medium. The
control group {open bar) was cultured in PC-1 medium for 7 days with daily changes of
PC-1 medium from day 4 to day 7. The peaks representing total nucleosides of interest
from typical HPLC analyses are shown in Panel A. The area under curve representing
total nucleosides was decreased after a 72 hr treatment as shown in Panel B.
Comparisons are being made between groups of equal number of cells {p < 0.05).
Asterisks represent significant difference compared to control {p < 0.05).
169
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Cont 12 hr 24 hr 72 hr R.l.
rbENT2A"
rbENT2
B
300
*
A
o
* H e
c 200
o
O
o
100
0
Contr 12 hr 24 hr 72 hr R.l.
Fig. 4-37. Time course of rbENT2/2A mRNA levels following nucleoside deprivation by
semiquantitative RT-PCR. Semiquantitative RT-PCR analysis was performed using 1.0
fig of total RNA from RTEC in the custom-made MEM lacking several amino acids
(Asn, Asp, Glu, Gin, and Gly) in the presence of 200 fiM MTX, MPA, and 5-FU for 12,
24, and 72 hr, or PC-1 media (Contr) for 72 hr. Seventy-two hour treated cells were re
introduced (R.L) to PC-1 medium for 24 hr prior to harvesting RNA for semiquantitative
RT-PCR analysis (see Methods 6.2.3.). Panel A, representative RT-PCR products
separated with 2% agarose gel electrophoresis and stained with ethidium bromide
showing the bands of expected sizes at 397 (rbENT2), 358 (rbENT2A), and 489 bp
(GAPDH). All the ethidium bromide-stained gels were depicted as inverse images for
clarity in data presentation. Panel B, close (rbENT2) and open (rbENT2A) bars represent
the mean ± s.e.m., observed in optical density in arbitrary units from three independent
experiments, each done in triplicate. All the results were normalized with respect to that
of a housekeeping gene, GAPDH. Asterisks represent significant difference compared to
control (p < 0.05).
170
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.2.3. Evaluation of basolateral uridine uptake in nucleoside-deprived cells
Basolateral uridine uptake was stimulated by 2-fold after 72 hr of treatment with both de
novo biosynthesis inhibitors (MTX, MPA, and 5-FU) and deprivation of precursor amino
acids, Asn, Asp, Glu, Gin, and Gly (Fig. 4-38). The levels of rbENT2 and rbENT2A
mRNA had increased 3-fold compared to those observed for PC-1 control cells (Fig. 4-
37). Similarly, when metabolically inhibited/nucleoside depleted cells were re-introduced
to PC-1 medium for 24 hr prior to uptake studies, uridine uptake from basolateral fluid
decreased 2-fold compared to that observed for cells subjected to nucleoside deprivation
(Fig. 4-38).
3.2.4. Effect of actinomycin D or cycloheximide on rbENT2/2A mRNA level and
basolateral transporter activity
The induction of rbENT2/2A mRNA by deprivation of precursors required for de novo
biosynthesis of nucleosides, could be the result of transcriptional or post-transcriptional
effects. Thus, experiments were performed after the addition of a general inhibitor of
transcription, actinomycin D or unrelated general translation inhibitor, cycloheximide.
Actinomycin D (5 /tg/ml, a concentration at which general
RNA synthesis in inhibited by greater than 90%, based on the decrease in intensity of
GAPDH signal (data not shown)) or cycloheximide (2.5 /ig/ml) was added to the
metabolically inhibited/nucleoside-depleted or control RTEC cells, and rbENT2/2A
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
— 200
o
3 - >
■ # - *
e
o
2 1 0 0 ■
o
Control Treated
Re-introduced
to PC-1
Fig. 4-38. Evaluation of basolateral uridine uptake in nucleoside-deprived cells. Primary
®
cultured RTEC grown on 12 mm diameter Clearwells on culture day 4 were maintained
in custom-made MEM lacking selective amino acids in the presence of 200 pM MTX,
MPA, and 5-FU {hatched bar) or PC-1 media {open bar) for 72 hr. Some of the 72 hr-
treated cells were re-introduced to PC-1 medium for 24 hr prior to uptake studies {closed
bar, R.I.). Na+ -independent 3 H-uridine (5 pM) uptake from basolateral fluid (5 pM) was
evaluated at 1 min. Asterisks represent significant difference in basolateral uridine uptake
compared to control {p < 0.05). Each data point represents mean ± s.e.m., n = 6.
172
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mRNA levels were determined (Fig. 4-39). Both mRNA levels decreased by more than
80% witih actinomycin D treatment (Fig. 4-39), as expected from transcription blockade.
Cycloheximide did not alter the rbENT2/2A mRNA signal induction in cells deprived of
nucleosides (Fig. 4-39). GAPDH mRNA levels were lower by >90% in the presence of
unselective mRNA synthesis blocker, actinomyin D while unaltered with cycloheximide.
Whether the induction of rbENT2/2A mRNAs requires de novo protein synthesis
in response to both nucleoside deprivation and metabolic inhibitors MTX, MPA, and 5-
FU was determined. Next the effect of a general translational (i.e., protein synthesis)
inhibitor, cycloheximide was also investigated. 3 H-uridine uptake from the basolateral
fluid of RTEC cells was evaluated in the presence or absence of 2.5 /xg/ml cycloheximide
as a protein synthesis inhibitor which non-selectively blocks the peptidyl synthetase
activity of eukaryotic ribosomes. Primary cultured RTEC incubated in the presence of
cycloheximide for 18 hr, at 54 hr post-treatment with the custom MEM to induce
rbENT2/2A mRNA synthesis, resulted in a decrease by 70 ± 2% compared to the cells
that were continued to grow in the custom MEM in the absence of cycloheximide (Fig. 4-
40). Greater than 85% inhibition of induced nucleoside transporter activity was observed
when RTEC pre-exposed to the custom MEM were treated with 5 pg/ml actinomycin D
(Fig. 4-40).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A B
rbENT2A— T
Contr CHX ActD
rbENT2
GAPDH
Contr CHX ActD
Fig. 4-39. Effect of actinomycin D or cycloheximide on induced rbENT2/2A mRNA
level. Day-4 primary cultured RTEC were maintained in custom-made MEM lacking
selective amino acids in the presence of 200 fiM MTX, MPA, and 5-FU for 72 hr.
Actinomycin D {ActD, 5 pg/ml) or cycloheximide {CHX, 2.5 jug/ml) was added to the
day 6 RTEC for 18 hr. Control cells were continued to grow in the custom MEM in the
absence of actinomycin D and cycloheximide. In Panel A, after 18 hr of treatment, total
RNA was extracted, and levels of 397 (rhENT2), 358 (rbENT2A), and 489 bp {GAPDH)
mRNA expression were analyzed by semi-quantitative RT-PCR (see Methods 6.2.3.).
RT-PCR products were electrophoresed in 2% agarose gel and stained with ethidium
bromide to identify the bands of expected sizes. All the ethidium bromide-stained gels
were depicted as inverse images for clarity in data presentation. Panel B, close (rbENT2)
and open (rbENT2A) bars represent the mean ± s.e.m. observed in optical density in
arbitrary units from three independent experiments, each done in triplicate. All the results
were normalized with respect to the GAPDH signal. Asterisks represent significant
difference compared to control (p < 0.05).
174
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
c
o
o
120
80
40
*
.
Control CHX ActD
Fig. 4-40. Effect of actinomycin D or cycloheximide on induced basolateral transporter
activity. Primary cultured RTEC grown on 12 mm diameter Clearwells on culture day 4
were maintained in custom-made MEM lacking specific amino acids in the presence of
200 fiM MTX, MPA, and 5-FU for 72 hr. Actinomycin D {ActD, 5 /tg/ml) or
cycloheximide {CHX, 2.5 /xg/ml) was added to the day 6 RTEC for 18 hr prior to uptake
studies. Control cells were allowed to grow in the custom MEM in the absence of
actinomycin D and cycloheximide. Na+ -independent 3 H-uridine (5 fiM) uptake from
basolateral fluid was evaluated at 1 min. Asterisks represent significant difference in
basolateral uridine uptake compared to control (p < 0.05). Each data point represents
mean ± s.e.m., n = 6.
175
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V. DISCUSSION
R e l a t i v e r o l e a n d c o n t r ib u t io n o f ENT2/2A in m e d ia tin g
NUCLEOSIDE TRANSPORT ACROSS RABBIT TRACHEAL EPITHELIAL CELLS.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1. F u n c t io n a l E v id e n c e f o r B a s o l a t e r a l C a r r ie r -m e d ia te d
N u c le o s id e T r a n s p o r t P r o c e s s e s in RTEC M o n o la y e r s
Our study provided evidence for the presence and substrate selectivity of nucleoside
transport processes in the basolateral aspect of RTEC. Transepithelial transport of various
molecules using air-interface RTEC mononlayers was previously characterized (Mathias
et al., 1995; Mathias et al., 1996a; Mathias et al., 1996b). Here it has been demonstrated
that transport of uridine in the apical-to-basolateral direction was 80x higher than that of
a paracellular marker mannitol. Since uridine itself is relatively hydrophilic and
impermeable across lipid bilayers, the presence of nucleoside carriers functionally
expressed in both apical and basolateral aspects of epithelial cell membranes is required
to complete transepithelial transport of nucleosides.
1.1. Basic characteristics of transporters
A Na+ gradient is not required in uridine transport across the basolateral membrane
(Table 4-1), similar to that found in the T84 intestinal epithelial cells (Mun et al., 1998).
Concentration gradient may be the main driving force for nucleoside exit across the
basolateral membrane in primary cultured RTEC. The >97% decrease in basolateral
uridine uptake at 4°C compared to that at 37°C suggested that this process was non-
diffusional (e.g., carrier-mediated). Compared to apical uptake, the basolateral uptake
rate was five times lower (Table 4-1). This suggests the presence of a distinct basolateral
nucleoside transport process from the apical one that plays a role in net transcellular
absorption of nucleosides across the tracheal epithelium. The functional ability to efflux
apically preloaded uridine into basolateral fluid, chemically intact, confirmed the bi-
177
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
directional nature of nucleoside transport across the basolateral membrane of cultured
RTEC (Fig. 4-3). Moreover, the amount of intact uridine effluxed in a period of 1 hr
depended strictly on the duration of apical preloading that determines the level of
accumulated uridine inside epithelial cells. Concentration gradient, therefore, could be
the chief driving force of bi-directional nucleoside transport (Fig. 4-3).
Analogous to observations by Leung et al. (2001) and by Ward and Tse (1999), at
least two major subtypes of equilibrative nucleoside transporters are implicated from the
biphasic inhibition profile afforded by NBMPR (Fig. 4-1). The two inflection points
separated by 3 orders of magnitude in NBMPR concentration are characteristic of ei and
es type transporter pharmacology (Fig. 4-1). The ei subtype of nucleoside transporters is
relatively insensitive to nanomolar ranges of NBMPR, but can be inhibited in the
micromolar range. The es system, by contrast, displays a characteristic inhibition at
nanomolar ranges of NBMPR (Hyde et al., 2001). The Na+ -independent basolateral
uridine uptake in primary cultured RTEC was mediated by both ei-type (-30%) and es-
type (-70%) transport processes, which was supported by a biphasic dose response to
NBMPR inhibition. A schematic diagram of the two identified processes is illustrated in
Figure 5-1.
1.2. Relative distribution and population
Since Na+ -independent equilibrative nucleoside transporters primarily govern the
shuttling of substrates across the basolateral membrane of RTEC in culture, knowledge of
the relative populations of cell types in this model is helpful to determine the relative
distribution of these transporters. Several types of secretory cells are present in various
178
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
animal species, that tend to vary in their relative distribution or abundance in the upper
airway. Goblet and serous cells have been detected, while Clara cells populate the
bronchiolar sections. In the human airway, ciliated cells comprise half of the entire
epithelium, while the goblet cells are comparatively rare (Rennard et al., 1991). In the
rabbit, Clara cells are found abundantly everywhere in the upper airway epithelium. The
density of secretory cells in our primary cultured RTEC model is extremely similar to the
rabbit trachea in vivo (Plopper et al., 1983; Harkema et al., 1991). About 18% Clara cells,
1% mucus cells, 38% basal cells, and 43% ciliated cells have been previously
characterized by Mathias et al. (1995). Statistically, the most probable location of the
equilibrative nucleoside transporters in this model is among the basal and ciliated
epithelial cells, since these clearly make up a combined 80% of the primary cultured
RTEC.
Nevertheless, quantification of the relative population of concentrative versus
equilibrative nucleoside transporters in native trachea is somewhat a bottleneck due to the
very nature of these proteins. And also there are no specific radioligands or labeled
antibodies available for each type of transporter. Radiolabeled NBMPR is perhaps the
best option available right now, but again it will not come close to delineating any
specific isoform of equilibrative nucleoside transporters, since it can bind to both es and
ei subtypes.
Kinetic analysis demonstrated that basolateral uridine uptake consists of two
components: carrier-mediated and passive diffusion. Overall, more than 80% of uptake
was predominated by the carrier-mediated component (Table 4-1). The Km of 210 fiM is
consistent with 120 fiM to 260 jxM affinities reported for sodium-independent uridine
179
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
transport processes in mammalian cells (Cass et al., 1999). Kinetic parameters of
saturable processes for both apical and basolateral uridine uptake indicated that apical
uptake was likely to be a high-affinity/low-capacity transport process, whereas a low-
affmity, high-capacity transport process may exist on the basolateral side (Table 4-1).
Other basolateral transporters (e.g., facilitated glucose transporters (Takata, 1996) and
peptide transporters (Terada et al., 1999)) also exhibit a lower affinity towards substrates.
Similarly, differential transport mechanisms for apical vs. basolateral transport of
nucleosides have been demonstrated in rabbit kidney (Williams and Jarvis, 1991), rat and
rabbit intestine (Chandrasena et al., 1997). These latter reports show a Na+ -dependent
secondary active nucleoside transport process in the apical cell membrane with the
permeation across basolateral membranes mediated by a Na+ -independent facilitated
transporter. Shuttling of uridine across the basolateral membrane of RTEC may be
mediated via a similar mechanism. In polarized epithelial systems, molecules of
nutritional value (such as glucose, peptides as well as nucleosides) must cross basolateral
membranes of absorptive epithelia in vectorial transport from the lumen to systemic
circulation (Takata, 1996; Terada et al., 1999).
1.3. Drug delivery applications
Our studies of uridine uptake involving nucleoside analogs of therapeutic significance
indicate that the apical and basolateral nucleoside transport systems play an important
role in governing their accessibility to systemic circulation (Fig. 4-2). Presence of the 3’
hydroxyl group in nucleoside analogs seems to be critical to retaining high inhibitory
capacities to both apical and basolateral uridine transport systems. Additionally,
180
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Lumen V ________ J Blood
Pyrimidine
Na+
es
Purine
r \
Fig. 5-1. Carrier mediated basolateral nucleoside transport in primary cultured
RTEC.
halogenation of the purine or pyrimidine bases of nucleoside analogs is also an effective
structural determinant for high-affinity interaction with both apical and basolateral
transport systems in primary cultured RTEC. These two requirements are consistent with
structure-inhibitory profiles reported for Na+ -dependent nucleoside transport in human
intestine (Patil et al., 2000) and kidney (Brett et al., 1993). Moreover, nucleobases and
their analogs (hypoxanthine and 5-FU) seem to have stronger interactions with the
basolateral but not apical nucleoside transport system. Overall, the basolateral Na+ -
independent nucleoside transport process(es) appear to be more tolerant to substitutions
or structural modifications of nucleosides. Selectivity of nucleoside transport is evidently
isoform-specific (Patil et al, 2000), as seen in our observations on substrate selectivity,
which may vary depending on cell or tissue types. Our results are consistent with reports
of differential inhibitory patterns of nucleoside transport in various systems (Dresser
181
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
et al, 2000; Lum et al., 2000; Patil et al, 2000). Thus, it is highly likely that the transport
of uridine along with other physiological nucleosides may be mediated via the same
process in our in vitro model.
1.4. Molecular rationale
Ultimately, the cloning of the rabbit analog of an equilibrative nucleoside transporter
from the tracheal epithelium partially fulfilled the necessary knowledge to molecularly
characterize this process. Relative roles of rbENT2 or rbENT2A in mediating nucleoside
transport in the basolateral membrane of tracheal epithelial cells is essential to delineate.
Their distribution within the specific cell types of the trachea is unknown. Through
detection of the mRNA message of both transporters, it can be assumed that in this
specific setting they both coexist at approximately the same ratio. However, if present in
other tissue/cell types, the relative levels of either rbENT2 or rbENT2A can vary greatly.
The mechanisms of alternative splicing, its regulation, and the factors influencing the
turnover rates of both transporters will play important roles in answering these questions.
Although it is clear that in epithelial tracheocytes most of the basolateral function in
nucleoside uptake is of Na+ -independent nature, the quantitative contributions of rbENT2
and rbENT2A are still unknown. Antisense approach can be used to transiently reduce
the levels of rbENT2 and rbENT2A transcripts in primary cultured RTEC in order to get
an idea of the relative involvements of the activity of each transporter. Although this
approach has been successfully used in other systems, its feasibility here remains
questionable until the steady state kinetics of rbENT2 and rbENT2A protein turnover in
primary cultured RTEC are characterized.
182
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2. M o l e c u l a r, C e ll u l a r, an d St r u c t u r a l Sim il a r it ie s and
D iffe r e n c e s bet w e e n Id e n tified R a b b it E N T 2 a n d its Splice
V a r ian t
2.1. Genetic information
Our results clearly demonstrated that alternative splicing of the rbENT2 primary
transcript representing rbENT2 produces two homologous isoforms, differing in their C-
termini, termed rbENT2 and rbETN2A (Fig. 4-5). Northern analysis with a rbENT2-
specifc cDNA probe detected the presence of a message with approximate 2000
nucleotides in size (Fig. 4-8A), consistent with the reported ENT2 cDNA (ENT2_Rat
jejunum, accession AFO15305, 1678 bp; ENT2_Human placenta, accession AF029358,
1576 bp; ENT2_Mouse Brain accession AF183397, 2372 bp). Higher specificity of RPA
and RT-PCR analysis (Fig. 4-8B and 4-8C) was required to confirm the presence of the
alternatively spliced variant, rbENT2A, whose expression was detected in a number of
rabbit tissues. An equal ratio with its wild type throughout the tissues studied suggested
that the event is of constitutive nature. This type of expression contrasts with other
phenomena where the splicing event appears to be regulated by distinct key physiological
cues in response to certain external stimuli or developmental prompts in a growing
organism. Moreover, whether the difference in mRNA steady state levels translates to
functional diversity is currently unknown.
2.2. Tissue distribution
The existence of several different ENT isoforms is expected to have a physiological
183
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
significance that is related either to their unique properties, or to their regulation. In either
case, their tissue distribution should provide some clues as to what their role is on a large
scale. The expression of mRNA for the ENT2/2A isoforms in a range of rabbit tissues has
been determined using a RPA assay. Similar studies for ENT1 (Griffiths et al, 1997a;
Mackey et al, 1998; Mackey et al, 1999), ENT2 (Griffiths et al, 1997b; Crawford et al.,
1998) and ENT3 (Hyde et al, 2001) have been reported in animal tissues. A search of the
EST database (BLAST search, http://www.ncbi.nlm.nih.gov/BLAST) revealed that
hENTl is expressed in many neoplastic tissues, and so is likely to play an important role
in drug uptake by tumors (Griffiths et al, 1997a; Mackey et al, 1998; Mackey et al,
1999). Identification of ENT2 mRNA in rat and human tissues indicated that, like the
ENT1 isoform, this transporter is also expressed in a broad range of cell types including
brain, heart, placenta, thymus, pancreas, prostate, kidney, and is particularly abundant in
skeletal muscle (Griffiths et al, 1997b; Crawford et al, 1998). Examination of the EST
database (http://www.ncbi.nlm.nih.gov/BLAST) revealed that ENT3 is expressed in a
range of mouse and human tissues in addition to kidney and placenta, including adult
breast, colon, testis, fetal liver and spleen, and in a number of neoplastic tissues (Hyde et
al, 2001). The substrate selectivity of mouse or human ENT3 has not yet been
established. Due to its unique pattern, ENT3 may have a distinct physiological impact.
2.3. Functional characteristics
2.3.1. In HEK heterologous expression system
When transiently transfected with rbENT2 cDNA, the HEK293 cell system is
functionally active with properties of an equilibrative nucleoside transporter, exhibiting
184
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
similar general characteristics as previously described in other systems (Osses et al,
1996; Crawford et al., 1998). Rabbit ENT2 transports nucleosides in a Na+ -independent
manner (Fig. 4-17A). It is concentration-dependent (Fig. 4-19), and NBMPR-insensitive
(Table 4-2, IC 5< > > 10 fiM). Interestingly, rbETN2A reveals no apparent transport activity
when transiently expressed in HEK293 cells (Fig. 4-17B), but shows functional activity
when transiently tansfected in MDCK cells (Fig. 4-24). Moreover, when transiently
cotransfected with its wild type (rbENT2) in HEK293 cells, this splice variant seems to
negatively regulate the functional activity of rbENT2 in a dose-dependent manner (Fig.
4-21). It was hypothesized that the spliced variant rbENT2A appears nonfunctional in
HEK293 cells because it doesn't reach the plasma membrane in nonpolarized expression
systems. The non-functionality could be due to the fact that HEK293 cells are not
polarized epithelial cells, or due to difference in system-specific protein sorting and
trafficking, where rbENT2A is localized in some intracellular compartments (i.e.,
lysosome) due to the different C-terminal sorting motifs it contains as a result of
alternative splicing. Its functional nature can be described along the lines of providing the
epithelial cells with a wider range of substrate selectivity, not a higher capacity of uptake.
Maximal velocity of uridine uptake may not change in a system that is cotransfected with
both rbENT2 and rbENT2A probably because of limited expression sites for equilibrative
nucleoside transport elements. Hence in the functional studies an additive or synergistic
effect was not observed when MDCK cells were cotransfected with increasing amounts
of plasmids carrying the gene for both transporters (data not shown). However, when
each cloned ENT was overexpressed alone in MDCK cells, it showed functional
characteristics with distinct V m ax and Km values (Fig. 4-25). In a compare/contrast
185
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
example, STAT3j3, a splice variant of transcription factor STAT3, lacks 50 amino acids
near the C-terminus. STAT3 and STAT3(3 are co-expressed in various cell types.
STAT3p is transcriptionally inactive on the intercellular adhesion molecule 1 (ICAM-1)
promoter. Low amounts of STAT3(3 significantly decreased trans-activation by STAT3,
while high amounts of STAT3J3 completely inhibited STAT3 -mediated transcriptional
activation. Therefore, STAT3P is a dominant negative regulator of transcription.
(Caldenhoven et al., 1996). Table 5-1 summarizes the kinetic parameters of uridine
uptake in HEK293 cells by rbENT2 and rbENT2A.
2.3.2. In MDCK heterologous expression system
Alternative splicing of ENT1 variants has been reported for the mouse homologues (Kiss
et al., 2000; Handa et al., 2001). One of the variants encodes for two additional amino
acid residues in a large intracellular loop, resulting in a lower functional activity. The
other has a distinct sequence in its 5'-untranslated region (Kiss et al., 2000). However,
their kinetic and pharmacological properties (as compared to those of the wild-type,
mENTl) were not determined yet. Many reports of alternative splice variants resulting in
diverse consequences exist. These include variants with similar activities (e.g., 5-HT7
receptor (Heidmann et al., 1998), and glycine transporter 1-c (Kim et al., 1994)) or those
that are non-functional with respect to the wild type (e.g., CFTR lacking exon 9 (Pagani
et al, 2000) and 5-HT6 receptor with a 289-bp deletion (Olsen et al., 1999)). On the
other hand, some nonfunctional spliced variants like the HI receptor (GHR1-279) (Ross
et al., 1997) act as negative regulators of their wild type. The hPepTl regulating factor
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(Saito et al, 1997), generated from the same pre-mRNA as hPepTl, acts as a regulator of
hPepTl function as its name implies.
A novel splice variant (rbENT2A) differing in its C-terminus from the Na+ -
independent, NBMPR-insensitive nucleoside transporter (rbENT2) was reported for the
first time. Alternative splicing of rbENT2 led to lowered expression and es-\ike
functional properties, but had no effect on ion independence. The rbENT2 transcript has a
higher RNA expression level than rbENT2A, revealed by our RT-PCR and RPA results
(Fig. 4-8B and 4-8C). Expression of rbENT2A variant containing the novel C-terminus
produced nucleoside transporters with different kinetic characteristics (Fig. 4-25) and
divergent pharmacological properties (Fig. 4-26). The splice variant exhibited a higher
affinity (Km = 0.04 mM) towards substrates compared to the Km values of rbENT2 (Km =
0.27 mM) and other cloned ENT2s (hENT2, 0.20 mM (Griffiths et al, 1997a; Griffiths et
al., 1997b); rENT2, 0.30 mM (Yao et al., 1997)). The splice variant exhibited a lower
capacity (V m ax - 0.6 nmol/mg protein/5 min) towards substrates compared to the V m a x
values of rbENT2 (V m ax =1.3 nmol/mg protein/5 min). Interestingly, rbENT2A behaves
as an es type-like nucleoside transporter having an IC50 value of 0.1 juM NBMPR (Fig. 4-
26). The wild types rbENT2, hENT2 (Griffiths et al., 1997a; Griffiths et al, 1997b) and
rENT2 (Yao et al., 1997) were all NBMPR-insensitive, and their IC50 values of NBMPR
were > 1 fiM (Fig. 4-26). Although rbENT2A possesses higher sensitivity to NBMPR
than other ENT2s, rbENT2A is still less sensitive to NBMPR compared to genuine es-
type nucleoside transporters (hENTl, IC50 = 3.6 nM (Griffiths et al, 1997a; Griffiths et
al., 1997b); rENTl, IC50 = 4.6 nM (Yao et al., 1997)). Kinetic measurements indicated
that both rbENT2 and its splice variant proteins were indeed capable of mediating uridine
187
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
uptake. A similar consequence of alternative splicing was observed in the mouse cationic
amino acid transporter (mCAT2) (Stevens et al, 2000). The mCAT2 gene is a single
copy that encodes two distinct protein isoforms, mCAT2 and mCAT2A. One of these has
almost identical transport properties with mCATl while the other isoform exhibits quite
distinct transport properties in its apparent Km and V m a x. Although mCAT2 and mCAT2A
proteins are nearly identical (97%), they are distinguished by a domain of 41 amino acids
that comprises the primary difference between them. Table 5-1 summarizes the kinetic
parameters of uridine uptake in MDCK cells by rbENT2 and rbENT2A.
Table 5-1. Kinetic parameters of uridine uptake.
Uptake Km (pM) V m a x (nmol/mg protein)
Primary cell culture
Apical (CNT)
3.4 ± 1.8 24.3 ± 5 .2 (/30 sec)
Basolateral (ENT)
210.2 ± 3 4 .3 0.34 ± 0 .0 2 (/min)
H EK cells
rbENT2-transfected
156.1 ± 4 0 .4 4.2 ± 0.4 (15 min)
M D CK cells
rbENT2-transfected 272.8 ±31.8 2.40 ± 0.08 (/5 min)
rbENT2A-transfected 41.6 ±9.8 1.26 ± 0.07 (/5 min)
2.3.3. Functionality, expression, and subcellular distribution of the C-terminal
truncated rbENT2
When compared with the rbENT2 C terminus, the novel variant rbENT2A C terminus
shares little amino acid identity after the alternative splicing point that is 41 amino acid
residues shorter (Fig. 4-5). However, uridine uptake by the C-terminal truncated protein
with the apparent lack of TMD 7 (predicted by TMPred™) in transiently transfected
188
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MDCK cells was non-functional. It indicates that the novel C terminus of rbENT2A
provides the information necessary for its distinct function from that of the wild type. It
also suggests that deletion of C-terminal tail appears to impair the acquirement of
functional conformation of the protein that is thought to form the intrinsic substrate
channel of the equilibrative nucleoside transporters.
The differences between the 10- and 11 -transmembmae topology for rbETN2 and
between the 6- and 7-transmembrane topology for rbENT2A are the localization of the
N- and C-termini on either face of the plasma membrane. Interestingly, the C-terminal
truncated form of rbENT2 significantly impaired Na+ -independent uridine uptake in
transiently transfected MDCK cells. Western blot analysis indicated that C-terminal
truncated rbENT2 was produced (Fig. 4-28A), however, confocal microscopy indicated
that such structural change abolished surface membrane expression of the truncated
protein (Fig. 4-28B and 4-28C). The consequences of the truncation of C-terminus and
the limitations of the cellular processing machinery probably play a role in increased
retention time of the impaired protein in the cellular compartments and the decreased
surface expression of the C-terminal truncated mutant. Moreover, since it was observed
that the C-termini of rbETN2 and rbENT2A are located on the outside face of the plasma
membrane, there is a possibility that these transporters could be regulated by extracellular
factors. The transmembrane topology of rbENT2A is unique when compared to other
ENT family.
Using a chimera approach, Sundaram et al. (1998) defined TMDs 3-6 of
hENTl/rENTl transporters as responsible for interactions of the transporters with
NBMPR and vasoactive drugs, and likely to form parts of the substrate translocation
189
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
channel. Additionally, domain swapping methods allowed to narrow down the major sites
of NBMPR interaction in chimeric nucleoside transporters to be TMDs 3-4 and TMDs 5-
6 (Sundaram et al., 2001b). As a prelude to such investigation, it is essential to clarify
orientation and membrane topology of rbENT2/2A transporter proteins. Our results
indicate a difference between rbENT2 and rbENT2A could be in the predicted TMD 7.
This can result in differential NBMPR sensitivities, which is a unique case compared to
literature reports.
Notable species differences were not observed among rbENT2 and rENT2 or
hENT2 in terms of kinetic parameters and hydropathy profiles (Yao et al., 1997; Griffiths
et al., 1997b). However, whether these transporters differ in their ability to recognize
nucleoside permeants or structurally related nucleoside analogs needs to be further
investigated for comparison. There are interspecies difference (Griffiths et al., 1997a;
Yao et al., 1997; Crawford et al., 1998; Sundaram et al., 1998) found in equilibrative
nucleoside transporter family: (1) hENT 1 -mediated transport is potently inhibited by
coronary vasodilators (dipyridamole, dilazep, draflazine) whereas rENT 1 -mediated
transport is unaffected. A surprising 3-log unit difference was observed in coronary
vasodilator IC50 values of hENT (very potent), nevertheless rENTl was ultimately
characterized as a es-type nucleoside transporter (Sundaram et al., 1998) based on its
primary structure. Importantly, using chimeric construct approach of hENT 1 and rENTl
demonstrated that transmembrane domains 3-6 of hENT 1 were essential for interaction of
vasodilators to equilibrative nucleoside transporters (Sundaram et al., 1998). Thus, this
molecular information would benefit our understanding on the design of more specific
190
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
inhibitors to the e/-type nucleoside transporters.
2.4. Polarized distribution
The transepithelial absorption of nucleosides requires transporter proteins on the apical
and basolateral membranes of RTEC. Our in vitro studies have shown that RTEC
monolayers possess the machinery required for Na+ ~dependent apical, as well as bi
directional and Na+ -independent basolateral nucleoside transport (see Results 1). The
equilibrative family of nucleoside transporters is less extensively studied (Hyde et al,
2001), compared to the functionally distinct concentrative group (Ritzel et al., 2001a;
Ritzel et al., 2001b). It has been hypothesized that the basolateral exit of nucleosides and
other nutrients is the rate-limiting step in transepithelial transport (Takata, 1996; Terada
et al., 1999), hence the studies discussed herein focus on two newly cloned, putative
basolateral nucleoside transporters from the ei family. Since antibodies for rbENT2/2A
are not available, their subcellular distribution was studied using HA-tagged constructs of
the transporters. MDCK epithelial cell line is extensively used for studies of membrane
trafficking and protein targeting (Bissonnette et al., 1999; Sweet et al., 2000; Mangravite
et al, 2001), and was utilized here as well.
Nucleosides are known to undergo active reabsorption within the kidney by CNTs
(Mangravite et al., 2001). Expression of green fluorescence protein (GFP)-tagged CNTs
in polarized MDCK cells has demonstrated that the larger GFP tag (compared to HA)
does not alter the substrate selectivity, but only modestly affects the kinetic activity of the
transporters (Mangravite et al, 2001). By using confocal microscopy and functional
studies CNTs have been clearly localized to the apical membrane of MDCK cells
191
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(Mangravite et al., 2001). Data presented here show that the addition of HA tag did not
significantly affect the functionality of the transporters compared to the untagged
versions, expressed in MDCK cells (Fig. 4-23). Confocal microscopy, with functional
studies indicated that both HA-tagged equilibrative nucleoside transporters (rbENT2 and
rbENT2A) are predominantly localized to the basolateral membrane of the MDCK cells
(Fig. 4-10 and 4-11). Consistent with our earlier observations that primary cultured
RTEC appear to possess basolateral equilibrative transport machinery (see Results 1), this
finding suggests that the two newly identified proteins may play a partial role in
transepithelial transport of nucleosides.
2.5. Subcelluiar localization
After permeabilized cells were double-stained, both rbENT2 and rbENT2A were
distributed in a pattern that overlapped with markers for the ER in the transiently
transfected HEK and MDCK cells (Fig. 4-12). This is a phenomenon seen for many other
plasma membrane proteins and reflects relatively slow transit of secretory proteins
through this compartment on their way to the cell surface (Lippincott-Schwartz et al,
2000). Although many of the signals accumulated intracellularly colocalized with ER
markers, it is possible that some of the protein was present in other, smaller cellular
locations (e.g., Golgi apparatus).
Our results provide a limited picture of the steady-state distribution of
rbENT2/2A proteins in heterologous expression systems that can be combined with
future data describing pulse-chase labeling experiments, to suggest a major route for the
subcelluiar trafficking of these molecules. It is possible that the two ENTs examined
192
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
have a significantly different metabolic half-life compared with other nucleoside
transporters, with many of the mutant proteins being degraded or immediately shuttled to
various membranes after synthesis in the ER. This interpretation and approach are
consistent with other systems, where for example the reduced surface levels of prion
proteins (PrPs) and their ER localization in stably transfected BHK-21 cells have been
studied (Ivanova et al., 2001). The subpopulation of cells in which rbENT2/2A can be
visualized may represent those cells in which the rate of transfected protein synthesis has
temporarily exceeded the rate of degradation, possibly due to fluctuations in the
metabolic state of individual cells as they grow and divide under the influence of the
transfecting vector/reagent. Hence it will be imperative to repeat similar studies in
native/primary cells once antibodies become available for rabbit ENT isoforms.
Reduced delivery of rbENT2/2A isoforms to the cell surface and their
accumulation in the ER suggests the possibility that ER quality control processes are
involved in the metabolism of these proteins. Secretory and membrane proteins are
subject to a stringent surveillance mechanism that causes retention in the ER of misfolded
or nonfunctional forms, such as those carrying mutations, and this prevents them from
reaching the cell surface (Hurtley and Helenius, 1989; Ellgaard et al., 1999). Proteins
retained in the ER are then eliminated by a pathway that is now known to involve reverse
translocation of the polypeptide chain into the cytoplasm, followed by proteasomal
degradation (Bonifacino and Weissman, 1998). Whether our studies behave in a manner
consistent with the involvement of such a mechanism needs to be studied. Proteasome
inhibitors cause a marked accumulation of PrPs involved in such mechanisms in cultured
CHO cells, and a significant prolongation of their metabolic half-lives (Wickner et al.,
193
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1999). Presumably, ER retention and degradation depend on certain molecular features
that are common between various families of proteins. It is attractive to hypothesize that
such early post-translational modifications are recognized by cells and trigger retention of
proteins along the secretory pathway, which in our case could be the ER.
The work reported here significantly extends other studies of the equilibrative
nucleoside transporter family, by systematically comparing rbENT2 and the spliced
variant rbENT2A using both light and confocal microscopic techniques in conjunction
with a variety of intracellular markers and by taking into account the possible effects of
cell background and transient transfection expression levels. The multiple lines of
evidence provide a general understanding on how ENTs heterologously expressed in
HEK293 or MDCK cells are trafficked, sorted, and transferred within cytosolic
compartments.
GFP tagging has been widely used to visualize the localization and movements of
proteins within cells (Lippincott-Schwartz et al, 2000). However, the success of this
approach requires evidence that attachment of the GFP moiety does not interfere with the
function of the target protein. Since the physiological role of rbENT2/2A remains
unknown, one must rely on indirect measurements as information to document the
functionality of ENT-GFP chimeras. If attachment of a GFP reporter adjacent to the N- or
C-terminus of either rbENT2 or rbENT2A will not affect a number of key attributes of
these molecules, including their biosynthetic maturation, subcelluiar localization, and
biochemical properties, these studies may place the foundation for the use of ENT-GFP
chimeras in a wide range of experiments on the distribution, trafficking, and molecular
194
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
transformations of nucleoside transporters in living cells or in animals.
2.6. Putative structural features
cDNAs that have been previously isolated encode proteins with the functional and
pharmacological characteristics resembling equilibrative nucleoside transporters: the
rabbit-ei-ENT2, and spliced variant es-ENT2A (Fig. 4-24 and 4-25). Topological analysis
of the predicted amino acid residues encoded within rbENT2 was consistent with a
topology and orientation reported by Kiss, et al. (2000) for the mouse isoform, Yao, et al.
(1997) for the rat isoform, and Griffiths, et al. (1997b) for the human isoform. The major
findings of this study are two fold: first, the orientation of putative secondary structures
of rbENT2 and rbENT2A proteins in the plasma membrane of MDCK cells has been
experimentally demonstrated (Fig. 4-30 and 4-31), and second, through treatment of
rbENT2 and rbENT2A with a specific hydrolytic enzyme (Fig. 4-32 and 4-33) it has been
demonstrated that the terminal 125 to 160 amino acid residues which encompass the C-
terminal tails of rbENT2 and rbENT2A are the primary variation that is a direct
consequence of alternative splicing of the precursor gene. Results from
immunofluorescent localization (Fig. 4-30 and 4-31) of the N-and C-terminally tagged
rbENT2/2A proteins clearly show that the N-termini of rbENT2 and rbENT2A are on the
cytoplasmic side of the plasma membrane, whereas the C-termini of both transporters are
extracellular. Protease cleavage assay (Fig. 4-32 and 4-33) of in vitro translated N-
terminally tagged rbENT2/2A proteins suggests that the predicted 11-TMD model for
rbENT2 and a 7-TMD model for rbENT2A were partially correct based on the
information derived from computer aided hydropathy analysis. The number and size of
195
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the resultant proteolytic fragments of rbENT2 or rbENT2A exposed to Lys-C were
essentially correct, with some unique exceptions that are discussed in detail below.
2.6.1. Membrane orientation
Studies followed the same general stipulation, that under predetermined experimental
conditions ensuring cell membrane integrity vs. membrane permeablization, selective
antibodies and site-specific enzymes (i.e. endoproteases or endoglycosidases) can react
with their targets within rbENT2 and rbENT2A, only if those targets are accessible to the
aqueous buffer containing the cross-reactive substance (i.e. anti-HA antibody, Lys-C, or
PNGase-F). In an effort to provide a more comprehensive understanding of the functional
divergence between rbENT2 and its spliced variant rbENT2A, the membrane orientation
of these newly cloned transporters was elucidated employing two consecutive
experiments: localization of their N- and C-termini and the existence of putative inter-
TMD loops. The pattern of proteolytic fragments observed on SDS-PAGE was also
considered as a rough test for predicted differences in C-termini of rbENT2 and
rbENT2A. Biologically tagged epitope detection and enzymatic cleavage were used to
determine and validate the predicted membrane topology of rbENT2/2A, as well as to
provide more direct biochemical evidence for the deduced differences in primary amino
acid sequences between rbENT2 and rbENT2A after the alternative splicing region. Such
experimental approaches have been widely utilized before. Evidence of substantial
structural conservation among mammalian NBMPR-sensitive nucleoside transporters was
collected through a series of studies where enzymatic cleavage was employed as a probe
(Kwong et al., 1993).
196
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Putative transmembrane domains integrate into the plasma membrane according
to topogenic determinants. Transmembrane domains of rbENT2/2A described by the
putative membrane topology were consistent with other equilibrative nucleoside
transporters from different species, and lacked homology to the mammalian I n
dependent concentrative nucleoside transporter family (Wang et al., 1997; Ritzel et al.,
1997; Ritzel et al., 1998) or to prokaryotic nucleoside transporters (Westh Hansen et al.,
1987; Craig et al., 1994). hnmunofluorescent determination of the accessibility of N- or
C-terminally fused HA-tag in MDCK cell preparations expressing rbENT2 or rbENT2A
reflected the predicted topology. Extracellular or intracellular localization of the N- or C-
termini of rbENT2/2A proteins were predicted based on reports on ENT2 isoforms from
the mouse or human. The membrane orientation of rbENT2 was really two fold, with the
observed result of intracellular N- and extracellular C-termini highly favored. Secondly,
the spliced variant, rbENT2A, had four different possibilities in transmembrane
orientation. This was primarily due to the extra factor of a possible 7th TMD (compared
to a less favored 6 TMD model) predicted by the computer aided hydropathy plots.
Results clearly indicated an intracellular N- and extracellular C-termini, which could only
conform to the 7 TMD model for rbENT2A. In summary, our results clearly show that
the N-termini of rbENT2 and rbENT2A are on the cytoplasmic side of the plasma
membrane (Fig. 4-30), whereas the C-termini of both transporters are extracellular (Fig.
4-31). Predicted TMD 7 in rbENT2A must anchor itself in the membrane according to
the results. Having roughly defined the orientation of consecutive transmembrane
segments, membrane integration was further characterized by site-specific cleavage of
the proteins. Due to relatively different abundance and predicted localization of amino
197
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
acid lysine residues between rbENT2 and rbENT2A, Lys-C endoprotease was chosen to
confirm the predicted difference in primary and secondary structures. Whether or not
these variations are a direct determinant of measured differences in kinetics or
pharmacology between rbENT2 and its spliced variant rbENT2A need further
investigation.
2.6.2. Lys-C digestion
Remarkable differences in functionality between rbENT2 and rbENT2A that were
observed previously in the MDCK cell heterologous expression system led to
experiments designed to provide structural basis for these differences. Analysis of
accessibility of select lysine residues to enzymatic cleavage was performed in MDCK
cell preparations expressing rbENT2/2A. Studies were aimed to determine the molecular
structure and orientation in the plasma membrane of these transporters. Briefly, all
experiments were performed based on predictions in PeptideMass™. Only a few results
in the set of studies did not turn up as expected. There is evidence for the existence of
select lysine residues on the predicted largest intracellular loop in both proteins in
environments that somehow shield them from Lys-C catalyzed hydrolysis. There is a
224 255
possibility that the polypeptide segment encompassing Lys through Lys in rbENT2
237 246 255
(Fig. 4-32), as well as a different segment that contains Lys , Lys , and Lys m
rbENT2A (Fig. 4-33), form secondary complexes or membrane associated domains that
somehow shield them from enzymatic attack. Lysine1 '7 located on the predicted 2n^
extracellular loop of both proteins is shielded from Lys-C digestion only in rbENT2 A
(Fig. 4-33). This subtle difference suggests that this residue might be located deeper
198
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
inside TMD3 for the spliced variant rbENT2A, which results in its inaccessibility to Lys-
C. Whether or not these residues (Lys237,246,2 5 5 , as well as Lys1 1 7 ) are involved in
determining the sensitivity of rbENT2A towards NBMPR still needs further
investigation. Distinct folding patterns, interaction with another unknown subunit, or
association with the plasma membrane are some other possible explanations. Lastly, the
above results offer possibilities that rbENT2 and rbENT2A can have dissimilar tertiary
structures, with apparent completely different C-terminal tails.
Our current understanding of the structure of rbENT2A has certain implications
about specific domains in rabbit equilibrative nucleoside transporters that may possibly
play key roles in the function of these proteins. Considering the studies described here,
these differences between rbENT2 and rbENT2A can be summarized by three major
points. The differences between rbENT2 and rbENT2A are three-fold: 1) the segment
starting at the first extracellular loop, spanning to the end of the second extracellular
loop; 2) the intracellular loop between TMDs 6 and 7; 3) entire length of the segment
following amino acid residue 290 until the C-terminus. Moreover, since our results for
the rabbit isoform of ENT2 strongly agree with the topology reported for the mouse and
human isoforms (Yao et al., 2001a; Yao et al., 2001b), the interpretation of rbENT2A
secondary structure becomes very important for rationalizing the inhibitor insensitivity
and divergent kinetics of ei ENT family. Apparently similar consequence of alternative
splicing was observed in mCAT proteins exhibiting distinct functional properties
(Stevens et al., 2000). Closs et al. (1993) has identified a domain of 41 amino acids that
comprises the only difference among mCATl, mCAT2, and mCAT2A proteins using a
domain swapping approach. This data indicate that perhaps substrate recognition and the
199
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mechanism of substrate translocation are established by this 41-amino acid domain
(Closs et al, 1993).
Many approaches have been taken to study the topology of integral membrane
proteins, yet each method displays certain limitations. Immunocytochemical localization
of peptide epitopes is a preferred method for studying transmembrane topology because it
is possible to study the full-length unmodified, and functional protein. Yet it is not always
possible to generate antibodies to epitopes of interest and it is difficult to distinguish
between buried extracellular epitopes and intracellular ones. Another approach involves
the detection of post-translational modification of residues as an indication of membrane
sidedness. Several integral membrane proteins such as the cystic fibrosis transmembrane
regulator (Chang et al., 1994), GluRl glutamate receptor (Hollmann et al., 1994), Glut
1 glucose transporter (Hresko et al., 1994), the SGLT1 glucose transporter (Turk et al.,
1996), and most recently GAT-1 and the GLYT1 glycine transporter (Bennett and
Kanner, 1997; Olivares et al., 1997) have been studied using glycosylation site insertion.
The insertion of such sites throughout a protein has the potential of altering the native
topology. In addition, the insertion of glycosylation sites into the loops of some proteins
can be detrimental to function as has been shown for the Glut-1 glucose transporter
(Hresko et al., 1994), GAT-1 (Bennett and Kanner, 1997), and GLYT1 (Olivares et al.,
1997).
In the absence of x-ray crystallography data of complex integral membrane
proteins such as ENT2 family, one must infer the transmembrane topology of such
proteins using methods such as those described above. While each of these approaches
has limitations, valuable information may be obtained as long as results are interpreted
200
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
with caution. Given the significant sequence homology between the members of this
transporter family, it is expected that the model of rbENT2 proposed here is
representative of equilibrative nucleoside transporters. These results provide a foundation
for future topology studies of this transporter family using alternative methodologies.
2.6.3. PNGase F treatment
These PNGase F treatment studies provided preliminary evidence for the existence of a
possible large glycosylated loop connecting TMDs 1 and 2 of both rbENT2 and rENT2A
(Fig. 4-34). Results suggest that this loop might face the extracellular side of the protein,
as indicated by the topology model in Fig. 4-6A and 4-6C. Oligosaccharide chains of
glycoproteins have been implicated in a number of biological processes, including
intracellular recognition, protein folding, protein stability, and intercellular sorting. The
role of the AMinked oligosaccharide chains on rbENT2 is not known at the present time.
Further studies are required to conclusively include or exclude Asn4 7 and Asn5 6 as sites of
iV-linked glycosylation in rbENT2/2A (Fig. 4-7). Accordingly, these asparagines should
be replaced with inert residues singly or in combination by site-directed mutagenesis in
order to confirm that they are the only sites of glycosylation in rbENT2/2A. Importantly,
the possible utilization of Asn4 7 and Asn5 6 as sites of postrtranslational glycosylation
implies that rbENT2 and rbENT2A both contain extracytosolic loops with the same basic
orientation within the plasma membrane.
3. M o d u la t io n o f R a b b it ENTs
3.1. Pharmacological modulation of basolateral nucleoside exit of RTEC monolayers
201
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The development of new approaches for improving the transportability of nucleoside
drugs into target tissues is ongoing. Various strategies have been hypothesized to enhance
the therapeutic index of nucleoside analogues. New inhibitors for nucleoside transporters,
including reagents that are targeted to the equilibrative nucleoside transporters, are now
being used for their cardioprotective properties in a site- and event-specific manner
(Seubert et al., 2000). Ultimately the manipulation of nucleoside transporter function
through an understanding of the regulation of transport activity on a protein level could
be used as a more direct tool to achieve this end.
3.1.1. Modulation by protein kinases
Protein kinases are known to regulate the short-term (~ minutes) function of various
transporters including both Na+ -dependent and Na+ -independent nucleoside transporters
(Davis et al, 1998; Melikian and Buckley, 1999). Modulation of basolateral nucleoside
transport bears therapeutic significance in the development of more efficient drug
delivery strategies of nucleoside analogs via pulmonary routes.
Some second messengers appear to be involved in the modulation of nucleoside
transport across the basolateral membrane of primary cultured RTEC. Forskolin (an
adenylyl cyclase activator) increased uridine uptake by 50%, while A23187, a calcium
ionophore, decreased it by 75%. Elevation of intracellular cAMP followed by activation
of PKA appears to be the stimulatory signaling pathway, as evidenced by a complete
reversal of forskolin-induced response with a PKA inhibitor, H89 (IC50 = 10 juM (Penn et
al., 1999)) (Table 4-5). The observed changes in kinetic parameters after exposure of
primary cultured RTEC to forskolin (Fig. 4-35) suggest the involvement of a transporter
202
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
protein with putative PKA phosphorylation site(s). Identification of such a transporter at
Blood Lumen
P > ENT
10//M Forskolin
10//M H-89
ENT
cAMP
A"!
<
Fig. 5-2. cAMP dependent modulation of ENT activity.
the molecular level requires further studies. Figure 5-2 is a schematic representation of
possible regulation of ENTs via cAMP dependent mechanisms.
3.1.2. Modulation by cytoskeletal components
Tamoxifen, a PKC inhibitor (IC50 -10 /tM (Rowlands et al., 1995)), failed to reverse the
effect of A23187 pretreatment on basolateral uridine uptake (Table 4-6), suggesting that
modulation through elevation of intracellular Ca2 + levels ([Ca2 + ]0 is independent of PKC
(Tai et al., 1996). Previous reports have shown that changes in [Ca2 + h or Ca2 + /calmodulin
activity have an effect on microtubule (MT) assembly/disassembly and membrane sheet
203
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
retraction. These MT-related processes regulate interactions of motor proteins in MT-
based intracellular transport (Benjamins and Nedelkoska, 1994; Wang et ah, 1996). Our
results provide evidence that colchicine blocked the decrease in uridine uptake elicited by
A23187, suggesting that a MT-dependent process may participate in Ca2 + -mediated
modulation of substrate uptake (Table 4-6). The increase in uridine uptake evoked by
forskolin was not affected by colchicine, however, indicating that MTs were not required
in PKA-mediated modulation of uridine transport (Table 4-6). Changes in uridine uptake
kinetics under PKA activity modulation were not observed in the co-treatment of RTEC
monolayers with A23187, as illustrated in Fig. 4-35. The role of insertion and retrieval of
existing transporter proteins to and from the basolateral membrane in nucleoside
transport need to be further investigated by more direct means (e.g., immunofluorescence
and cell membrane capacitance measurements). Figure 5-3 illustrates the suggested Ca2 +
dependent regulatory pathways involved in ENT modulation. At least two types of post-
translational modifications of equilibrative nucleoside transporters - phosphorylation and
glycosylation - may play regulatory roles in the rabbit trachea. PPsearch™
(http://www2.ebi.ac.uk/ppsearch) highlighted Ser2 2 4 in both rbENT2 and rbENT2A (Fig.
4-7), as the only locus for possible phosphorylation by PKA. No PKC phosphorylation
sites were predicted. Phosphorylation by kinases like PKA or PKC has been identified as
a cellular mechanism of functional modulation on the protein level. Changes in functional
activity after inducing cellular kinases in cultured primary cells were measured.
However, whether or not the PKA-dependent phosphorylation occurs at the
conserved Ser2 2 4 residue requires further studies. The effect of N-glycosylation on the
204
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
rbENT2 transport activity is unknown, and needs to be investigated. JV-glycosylation of
Lumen
ENT
ENT
pool
MT
Blood
Fig. 5-3. Ca2 + dependent modulation of ENT activity.
r
AO
hENTl at Asn had impact on the transport activity when expressed in Xenopus oocytes,
indicating that glycosylation of hENTl seems to be essential for transporter activity
(Sundaram et al, 2001a).
Possibilities to enhance the therapeutic index of nucleoside drugs arise from these
findings. The existence of various cell surface receptors in the upper airway that
modulates levels of intracellular second messengers has been well characterized. The co
administration of adrenergic receptor agonists, that act through elevation of intracellular
cAMP levels can potentially enhance the absorption of inhaled nucleoside analogues
(Freitag et al, 1996). On the other hand, the presence of cholinergic receptor
205
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
antagonists that prevents the elevation of [Ca2+]i can sum up to a similar outcome if
included in similar pulmonary formulations of nucleoside drugs (Reinheimer et al.,
2000).
3.2. Transcriptional modulation of rabbit ENTs
Information on the elements regulating the equilibrative nucleoside transporter gene
transcription is limited. Recently, it has been established that the genomic structures of
the hENTl and mENTl genes are located on chromosome 6 (Coe et al., 1997) and 17
(Choi et al., 2000) of human and mouse genome, respectively, but the promoter region
has not been characterized. Very little is known about the factors that regulate the
primary processing and the further fate of the equilibrative nucleoside transporter
mRNAs. The intron/exon structure for the equilibrative nucleoside transporter genes may
suggest differential excision/splicing of certain exons and, thus, the generation of variant
mRNA species and their translation products. The steady-state levels of functional
equilibrative nucleoside transporters may not necessarily correlate with the levels of
corresponding mRNAs. Moreover, the metabolic turnover rates of equilibrative
nucleoside transporters (i.e., translation versus degradation rates) and turnover rates of
the corresponding mRNAs (transcription versus degradation rates) may not necessarily
change in concert and are probably regulated differentially depending on cell phenotype,
environmental cues, and signaling pathways. Differentially polyadenylation as well as the
specific regulatory sequences that reside in the 3’-untranslated regions of the
equilibrative nucleoside transporter mRNAs may control the degradation rates of these
transcripts
206
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.2.1. System for metabolic stress
The data presented here describe a cellular response to deprivation of amino acid
precursors for de novo nucleoside biosynthesis and inhibition of rate-limiting steps in this
pathway. Namely, the induction of rbENT2 and rhENT2A mRNA expression and
stimulation of basolateral nucleoside transport was studied. Present results demonstrate
that precursor availability regulates mRNA abundance and transport activity of the
basolateral rbENT2/2A transporters in a constitutive and reversible manner. Results
obtained with actinomycin D and cycloheximide strongly suggest that these effects
resulted, at least in part, from increased gene transcription and protein synthesis.
Initially, to establish the validity of intracellular nucleoside level depletion in
primary cultured RTEC as a model for studying regulation of gene expression, it was first
essential to demonstrate that custom MEM containing MTX, MPA, and 5-FU modulates
levels of total nucleosides without significant toxicity. The schematic diagram illustrated
in Figure 5-4 outlines the reasoning behind the proposed method of nucleoside level
depletion. This is essentially the same way that intracellular nutrients regulate the
expression of glucose transporter (Dyer et al., 1997), amino acid transporter (Wasa et al.,
1996; Hyatt et al., 1997; Aulak et al., 1999), dipeptide transporters (Walker et al., 1998;
Ihara et al., 2000), and nucleoside transporters (Valdes et al., 2000). Antimetabolite
drugs, such as 5-FU and raltitrexed inhibit thymidylate synthase TS enzyme of de novo
biosynthetic pathway, resulting in depletion of intracellular nucleotide pools
(Rauchwerger et al., 2000). A concentration-dependent decrease of total cellular
nucleosides (Fig. 4-36) was achieved with our combination treatment in the absence of
207
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Lumen 5-carbon sugar
U . 5-FdU
Blood
5-FdU
10-formyl-TH
1 c aicar pathways
FAICAR 7 * ^
I MPA / \
j) MTX^I I - 5-FdUMP
dTMP
^^hM TX
FGAR
purine
pyrimidine
I O-formyl-THI
IMP - = ^ XMP XMP « = > GTP
V
f
Fig. 5-4. Model of nucleoside starvation.
GAR, glycinamide ribonucleotide; FGAR, 5-phosphoribosyl-N-
formylglycinamide; AICAR, aminoimidazole carboxamide ribonucleotide;
FAICAR, 1 -(5-phosphoribosyl)-5-formamido-4-imidazolecarboxamide; THF,
tetrahydrofolate; IMP, inosine 5’-monophosphate; XMP, xanthosine 5’-
monophosphate; dUMP, deoxyuridine 5 ’ -monophosphate; dTMP, deoxyuridine
5'-monophosphate; 5-FdU, 5-fluorodeoxyuridine; 5-FdUMP, 5-
fluorodeoxyuridine 5 -monophosphate.
key amino acid precursors for purine or pyrimidine biosynthesis, without significant
cytotoxicity over 72 hr. This is possibly due to the fact that epithelial tracheocytes do not
divide very rapidly (Puchelle, 2000).
3.2.2. Effect of starvation and antimetabolites
This study demonstrated that rbENT2 and rbENT2A gene expression is induced by
deprivation of primary cultured RTEC from key precursors and inhibitors for rate-
limiting synthetic enzyme. Similarly, Valdes et al. (2000) demonstrated that following
208
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
nucleoside starvation in vivo the expression of the CNT1 nucleoside transporter was
induced in the Wistar rat jejunum, which indicates that our observation on regulatory
properties of nucleoside transporter expression following nucleoside deprivation may not
be restricted to one cell type. The increase of rbENT2/2A mRNA levels in cells
challenged for de novo nucleoside biosynthesis could be due to an increase in either
transcription rate of genes or mRNA stability, or both. To determine if nucleoside
deprivation alters the transcription rate of rbENT2/2A genes, nuclear run-off assays have
to be performed using nuclei isolated from RTEC to determine the relative changes of
hybridization signals for rbENT2/2A over genomic DNA. If the transcription rate
remains unchanged during nucleoside deprivation, the data may suggest that transcription
of rbENT2/2A are not induced during nucleoside starvation, therefore, the activity of the
rbENT2/2A proteins may be altered by nucleoside depletion at the post-transcriptional
level. Regulation of such nature can be due to re-sorting and re-organization of the
existing intracellular pools of ENTs to the plasma membrane. Such events have been
reported in rat proximal tubule brush border membrane vesicles enriched with the 1 3 -
amino acid taurine transporters (Chesney et al., 1990).
There are many ways by which nucleoside transport activity can be modulated
(Pastor-Anglada et al., 2001). These include alterations in the expression of transporter
proteins and/or modulation of their activity. In addition to changes in nucleoside
transporter activity due to transcriptional or translational regulation (Soler et al., 1998;
Soler et al., 2000; Soler et al., 2001a; Soler et al., 2001b), the activity of existing
transporter proteins can be changed by different pharmacological agents and endogenous
signaling molecules (Fideu and Miras-Portugal, 1992; Jones et al., 1994; Mun et al.,
209
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1998; Sen et al, 1998; Sen et al., 1999). The increase of basolateral uridine uptake in
nucleoside-depleted cells paralleled the increase of rbENT2/2A mRNA levels. This
increase in transporter activity may reflect an induced rbENT2/2A protein synthesis. The
mechanisms that sense nucleoside availability and those involved in transfer of this
information to transporter gene expression in the trachea are not known. Nutrient-
dependent changes in gene expression have been documented for a number of proteins,
including amino acid transporters (Wasa et al., 1996; Hyatt et al., 1991; Aulak et al,
1999), and dipeptide transporters (Walker et al, 1998; Ihara et al, 2000). It should also
be noted that the mRNA level of alternatively spliced isoform rbENT2A in response to
nucleoside deprivation was up-regulated constitutively with rbENT2. However, further
work is needed to test whether the observed stimulation in basolateral uridine uptake was
due to an increase in transporter protein expression in one particular isoform (e.g.,
rbENT2) or due to concerted increases in multiple isoforms of nucleoside transporters.
The steady state level of mature mRNA might be affected by post-transcriptional
processing or via modulation of message stability. The role of a transcription inhibitor,
actinomycin D, on the levels of rbENT2/2A mRNA in primary cultured RTEC was
studied. Actinomycin D blocked nucleoside deprivation-mediated increase of rbENT2/2A
mRNA levels. This indicates that the rbENT2/2A mRNA levels were up-regulated in
metabolically inhibited/nucleoside-depleted cells in a manner dependent on transcription
(Fig. 4-39). As expected, protein synthesis inhibition had no effect on rbENT2/2A
mRNA levels.
To determine if general protein synthesis was required for the enhanced
basolateral uridine uptake in nucleoside-depleted cells, uridine uptake from the
210
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
basolateral fluid of RTEC cells treated for 72 hr with MTX, MPA, and 5-FU was
measured in the presence or absence of cycloheximide. Cycloheximide generally inhibits
all protein synthesis at low concentrations. Stimulation of basolateral transporter activity
following nucleoside deprivation was abolished by cycloheximide (Fig. 4-40), suggesting
that new synthesis of the rbENT2/2A transporter proteins takes place in the nucleoside-
deprived cells that leads to upregulation of nucleoside transport activity. Stimulation of
rbENT2/2A mRNA expression and activity thus may result from increases in both gene
transcription and protein synthesis. However, other intracellular events, such as reduced
mRNA decay or alteration in mRNA translation efficiency and membrane trafficking of
transporter proteins cannot be excluded from our studies. Direct quantification of
rbENT2/2A protein when intracellular nucleoside level is depleted in cases of nucleoside
shortage will shed more light on the regulatory mechanisms involved, by looking into the
efficiency of mRNA translation.
In the study of transcription modulation, a condition where intracellular
nucleoside pool is significantly compromised by the agents, MTX, MPA, and 5-FU, that
inhibit de novo biosynthetic pathway with differing mechanisms of action, was
established. Under such conditions, when cells are deprived of important basic building
blocks, hindering growth and differentiation, cells may respond to up-regulate ENT
transcription. Furthermore, modulating ENT protein expression and transporter function
that are involved in impact of nucleosides from extracellular space need to be further
investigated. It has been confirmed that, when de novo biosynthesis of nucleosides is
compromised, up-regulation of ENT transcription in response to such deprivation may be
an important alternative salvage mechanism.
211
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.2.3. Therapeutic significance
These findings may be applicable in multiple therapeutic situations. Antifolate resistance
can develop by circumvention of the inhibition of de novo nucleotide biosynthesis via the
salvage of extracellular nucleosides and nucleobases. ENT-mediated transport inhibitors
(e.g., NBMPR, dipyridamole) can enhance the cytotoxicity of non-nucleoside anticancer
antimetabolites by blocking nucleoside salvage mechanisms. For example,
coadministration of AZT, and antimetabolites (e.g., MTX and 5-FU) with dipyridamole
leads to enhancement of the chemotherapeutic effect of human neoplasias by selectively
blocking ENT-mediated transport (Zhen et al., 1992). Similar strategies of the combined
therapy are being widely explored. ENT inhibitors can also be used to enhance the
cellular uptake of various chemotherapeutic agents via secondary Na+ -dependent CNTs
through the inhibition of ENT-mediated efflux. For instance, dipyridamole enhances the
toxicity of vidarabine in L1210/C2 cells, which express es/mENTl, ezVmENT2, and
c(f/mCNT2 transport activities (Dagnino et al, 1991). Therefore, clinical efficacy of
numerous antimetabolites can be improved by overcoming acquired or inherent
mechanisms of resistance involving ENTs. Moreover, Pressacco et al. (1995a; 1995b)
reported that inhibition of DNA synthesis by hydroxyurea and 5-FU in the human bladder
cancer cell line led to an increase in the es-type ENT expression by up to 10-fold
(Pressacco et al., 1995b). Up-regulation of es expression by 5-FU and raltitrexed offers a
means to augment the effects of drugs like gemcitabine, which enters cells via carrier-
mediated mechanism. Therefore, combining a nucleoside analogue with agents that
212
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
increase nucleoside transporter expression at the cell surface may be potentially useful for
cancer treatment.
213
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
VI. CONCLUSION
H ig h l i g h t s a n d g a p s.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1. S u m m a r y o f F in d in g s
1.1. Functional characterization of basolateral nucleoside transport
Basolateral nucleoside transport in primary cultured RTEC is the final step in the
vectorial, transcellular absorption of nucleosides and nucleoside drugs from the upper
airway. A low-affinity and high-capacity transport system exists on the basolateral
aspect, exhibiting characteristics of bi-directionality, temperature/concentration
dependency, and broad specificity towards purines and pyrimidines without requiring
Na+ . Primary cultured RTEC appeared to possess basolateral equilibrative-
sensitive/insensitive (es/ei) type transport machinery, manifested as a biphasic dose
response to NBMPR inhibition. Additionally, a number of nucleoside analogs of
therapeutic relevance appeared to compete with uridine uptake from basolateral fluid.
1.2. Functional characterization and pharmacological classification of rbENT2A
cDNA clones encoding the complete protein sequence of an equilibrative nucleoside
transporter (rbENT2) and its novel C-terminal splicing variant rbENT2A were isolated by
RT-PCR approach on RNA prepared from rabbit trachea. Rabbit ENT2 corresponding
protein sequence consists of 456 amino acid residues, with rbENT2A being shorter by 41
residues. Both rbENT2 and rbENT2A transcripts are found in a variety of rabbit tissues
including intestine, kidney cortex, kidney medulla, and trachea, with different expression
levels by RPA. By utilizing a heterologous expression system - MDCK epithelial cell line
- to characterize rbENT2 and rbENT2A at the protein level, our results demonstrated that
both rbENT2 and rbENT2A were expressed with a molecular mass of 49 kDa and 44
kDa, respectively. Both rbENT2 and rbENT2A exhibited functional uridine uptake
215
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and efflux in transfected MDCK cells, without the requirement of coupling to a
secondary ion (e.g., Na+ ). Remarkably, rbENT2A displayed a higher affinity (Km = 41
p,M) and a lower capacity (V m ax = 0.6 nmol/mg protein/5 min) towards uridine, while
rbENT2-transfected cells demonstrated low affinity and high capacity towards the same
substrate (Km = 272.8 /tM, V m a x = 1.26 nmol/mg protein/5 min). Pharmacological profile
studies showed that 1 [iM NBMPR potently inhibited 3 H-uridine uptake mediated by
rbENT2A, but not rbENT2. This substantiated a novel splice variant rbENT2A (es-like)
bearing no identity and little similarity with known Na+ -independent es nucleoside
transporter group. The rbENT2A arises by alternative splicing of pre-mRNA encoding
for an ei type ENT, rbENT2.
1.3. Structural validation
Limited structural or molecular data are available to adequately explain unique functional
characteristics of mammalian equilibrative, NBMPR-insensitive ENT. We used
recombinant techniques to produce epitope tagged or truncated transporters that were
additionally exposed to selective hydrolytic cleavage with enzymes in order to
understand the structural basis for functional diversity between two newly cloned rabbit
NBMPR-insensitive ENTs. The transporter gene (rbENT2) isolated from the trachea,
kidney, and intestine of pigmented rabbit, was studied along with its alternatively spliced
variant (rbENT2A) present in these same tissues at a constant ratio. rbENT2 and
rbENT2A highly differ from each other in their size, secondary structure and membrane
spanning pattern. Molecular weight differences between rbENT2 and rbENT2A (50 and
44 kDa) mainly appear to be due to the alternative splicing event with an earlier
216
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
downstream stop codon for rbENT2A. The degree of post-translational modification with
carbohydrates was slightly different between the two proteins, as determined by ~5 kDa
difference in apparent core protein molecular weight after PNGase-F treatment. The
predicted hydropathy plots yielded 11 and 7 TMDs for rbENT2 and rbENT2A,
respectively, which could be partially confirmed, with minor differences. The
accessibility of potential enzymatic cleavage sites to endopeptidase Lys-C were evaluated
using MDCK cells transiently expressing either rbENT2 or rbENT2A. There were minor
differences in membrane spanning for the initial 6 predicted transmembrane regions for
both proteins. The major differences occurred in the C-termini for rbENT2 and rbENT2A
with only 16% identity and 35% similarity after amino acid residue 289. Results from
immunofluorescent staining of MDCK cells transiently expressing recombinant either
rbENT2 or rbENT2A proteins, under membrane permeabilized and non-permeabilized
conditions, suggest that the N-termini of both transmembrane proteins are intracellular,
whereas their C-termini are extracellular. Confocal microscopy analysis of subcellular
distribution for rbENT2 and rbENT2A revealed proteins predominantly localized in the
basolateral surface in transiently transfected MDCK cells. Western blot and confocal
microscopy analysis indicated that C-terminal truncated rbENT2 was successfully
produced and expressed by transfected MDCK cells, however, this smaller constmct
containing only the conserved TMDs between rbENT2 and rbENT2A (TMDs 1 through
6) was not functioning as a transporter. These data suggest that TMDs 7-11 of rbENT2
protein are involved in membrane targeting and perhaps substrate/inhibitor binding in
MDCK cells.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.4. Physiological regulation of basolateral nucleoside transport processes
Short-term pre-incubation of primary cultured RTEC with the calcium ionophore A23187
inhibited basolateral uridine uptake (irreversible with tamoxifen or staurosporine, PKC
antagonists) by an apparent significant reduction in the Jm ax, with Km unaffected. In
contrast, basolateral uridine uptake was increased by adenylyl cyclase activator,
forskolin, (reversible with PKA inhibitor, H89) resulting in an increased Km , but a lower
Jm ax - Uridine exit across the basolateral membrane of primary cultured RTEC occurs via a
2+
facilitative-diffusion carrier, which can be modulated by intracellular Ca levels and
PKA. The exit of intracellularly accumulated nucleosides into basolateral fluid of our in
vitro model can be modulated through at least two signal transduction pathways.
Involvement of rbENT2 or rbENT2A in cAMP dependent upregulation or Ca2 +
dependent downregulation of basolateral uridine uptake remains to be delineated from
this primary cell culture system. The changes in kinetics observed only suggest the
possible modification or re-sorting of exiting putative transporters, but do not offer
specific insight regarding the cloned proteins at hand. In order to track the behavior of
each putative protein in the primary cultured RTEC model, isoform specific antibodies
are needed. An alternative approach will allow for their individual regulation patterns to
be investigated in the future using similar conditions in heterologous over-expression
systems.
1.5. Gene regulation of ENTs in homeostasis of nucleosides
We initially established the validity of conditions where total cellular nucleoside
availability is decreased. We also evaluated the extent of depletion of intracellular
218
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
nucleoside levels in order to establish a threshold for triggering transcriptional regulatory
events. Primary cultured RTEC model deprived of nucleosides was achieved by the
application of three antimetabolites, MTX, MPA, and 5-FU. In addition, amino acids
(asparagine, aspartic acid, glutamine, glutamic acid, and glycine) that are essential for
incorporation into purine or pyrimidine rings were deprived from the culture medium.
The effects of nucleoside deprivation on gene regulation of rbENT2 and rbENT2A were
assessed by semiquantitative RT-PCR, and in separate studies by cellular 3 H-uridine
uptake measurements. Overall, we show that the regulation of the rbENT2 and rbENT2A
in rabbit tracheal epithelial cells is constitutively responding to nucleoside deprivation,
and also to re-feeding with necessary precursors. Although functional uptake
measurements in this case do not offer direct evidence for the involvement of either
rbENT2 or rbENT2A in the increased uptake of substrates under nucleoside deprivation,
they do parallel the directly assessed increase in the mRNA levels of both of these genes.
The response involves a concerted increase in gene transcription, which is followed by
protein synthesis, as could be observed by generally blocking RNA or protein synthesis
in these cells. Combination treatments to achieve both ends may have beneficial
therapeutic implications in the treatment of respiratory cancers or infections.
2. S ig n if ic a n c e o f t h e F in d in g s
We functionally identified and characterized two facilitated, Na+ -independent nucleoside
transport processes in basolateral membranes of primary cultured RTEC. The exit of
intracellularly accumulated nucleosides into basolateral fluid of our in vitro model
219
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
rbENT2 rbENT2A
Km for uridine = 272.8 //M K m for uridine = 41.6 //M
IC5 0 of NBMPR > 1 /#M
''■ 4 A ' " '
IC»of NBMPR = 100 nM
CO O H
Cytoplasm
Cytoplasm
► N-glycosylation site s ------► PKA phosphorylation site s
► PKC phosphrylation sites ?
Fig. 6-1. A schematic summary of functional findings and molecular outlook.
appears to be an important step in overall transcellular absorption of nucleosides.
Therefore, the basolateral equilibrative nucleoside transporter(s) in primary cultured
RTEC can be pivotal in governing the accessibility of luminally presented substrates to
systemic circulation, or local availability in pulmonary therapeutics.
Identification of basolateral nucleoside transporters from epithelial tracheocytes
will help to advance our understanding of functional properties and pharmacological
modulation of the primary exit mechanism of nucleosides across RTEC basolateral
membranes. These carriers also offer leads in improving the transportability of antitumor
and antiviral nucleoside analogs via pulmonary routes.
220
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.1. Functional importance
Our functional studies suggested that rbENT2 and rbENT2A might play a role in
absorption, disposition, and elimination of nucleosides and nucleoside drugs via
equilibrative nucleoside transporters. With the molecular identification of these
nucleoside transporter proteins, an understanding of the relationship between deduced
primary amino acid sequence, and putative secondary topological information provided a
structural basis for the functional differences among the two transporter isoforms.
Moreover, a better picture of the mechanisms of these two nucleoside transporters in
rabbit trachea was obtained by development of a heterologous expression system for
production of recombinant rbENT2/2A. Information on functional kinetics and substrate
selectivity of rabbit tracheal ENTs obtained in this study may be useful for the design of
in vivo targeting of nucleoside drugs and their delivery for the treatment of pulmonary
diseases. For example, as cited earlier, for nucleoside analogues of therapeutic
significance that appear to be substrates in carrier mediated absorption (primary studies
available are done for intestinal tissues), their bioavailability is significantly (usually >
50%) higher than compared to molecules that are absorbed by passive paracellular
diffusion. This can be even more advantageous when delivering the same permeants of
ENTs through the pulmonary route, because this will highly limit their first pass
elimination and metabolism.
The alternative splicing variant, rbENT2A possesses a markedly different kinetics
and distinct pharmacological profile, implicating that rbENT2A acts in conjunction with
the rbENT2 fine-tuning equilibrative nucleoside transport function in rabbit trachea. This
novel event appears to be constitutive in its origin, because both rbENT2 and rbENT2A
221
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
coexist at a constant ratio in all tissues examined. Furthermore, this even holds in
conditions of upregulation of their expression.
2.2. Molecular insights
Analysis of the predicted topology of rbENT2 was consistent with models of ENT2
isoforms reported in other species (Yao et al., 1997; Griffiths et al, 1997; Kiss et al.,
2000). Experimental evidence demonstrated the orientation of putative secondary
structures of rbENT2 and rbENT2A proteins in the plasma membrane of MDCK cells
(cytoplasmic N-termini and extracellular C-termini). Treatment of rbENT2 and rbENT2A
with a specific hydrolytic enzyme demonstrated that the divergent C-terminal 125 to 160
amino acid residues of rbENT2 and rbENT2A are the primary variations, as a direct
consequence of alternative splicing, of the precursor gene. In vitro translated rbENT2/2A
proteins appear to have predicted 11-TMD model for rbENT2 and a 7-TMD model for
rbENT2A. The structural differences between rbENT2 and rbENT2A, arising from
alternative splicing, could be partially responsible for the different functional properties
of these nucleoside transporters. Structural validation results provide substantial
structural information on functional diversity conferred by alternative splicing of the
novel rbENT2 gene. The models developed for the rbENT2 and its splice variant
rbENT2A provided a basis for investigating structure-function relationships of
equilibrative nucleoside transporter family.
Our data support the hypothesis that rbENT2/2A RNA synthesis induction in
response to nucleoside deprivation is augmented by increased protein synthesis. This
observation agrees with existing understanding of related nutrient transporter systems
222
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(i.e., for essential amino acids (Wasa et al., 1996; Hyatt et al., 1997; Aulak et al, 1999)
or di/tripeptides (Walker et al, 1998; Ihara et al, 2000))). Furthermore, the up-regulation
of rbENT2/2A mRNA levels and their specialized transport function has important
implications in antimetabolite drug resistance as a result of possible adaptive
mechanisms.
2 .3 . S u m m ary o f k ey fin d in g s
To summarize the important implications of these findings, the following four general
points of view can be considered: a) the reported experiments advance our physiological
understanding of the mechanisms involved in the transepithelial flux of nucleosides in the
rabbit trachea, defining some principles that may be applied to other epithelia; b) the
supported central hypothesis provides detailed information on molecular characteristics,
functional kinetics, and substrate selectivity of the individual transporters, allowing us to
predict in vivo organ-specific selection of nucleoside drugs for the treatment of
pulmonary diseases; c) as a result of this work, we have established a mammalian
heterologous expression system for screening the kinetics and interaction of potential
nucleoside drugs with rbENT2 isoforms, setting the stage for future studies of
intracellular trafficking, and regulatory mechanisms of the individual transporters; d) the
knowledge obtained will have an impact on the future by benefiting our molecular
understanding of structure-function relationship of equilibrative nucleoside transporters.
3. F utur e C o n sid e r a t io n s
223
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Elucidation of functional and molecular mechanisms in transepithelial transport of
nucleosides has great potential in revealing effective methods of controlling the
intracellular and extracellular levels of this important class of building blocks and
signaling molecules. Nucleoside analogues belong to a large group of therapeutics with a
wide range of clinical applications. Exploring targeted delivery of these molecules
through unconventional pathways that evade and limit first-pass metabolism is highly
important and beneficial for the ultimate goal of improving patient’s quality of life. These
studies set the stage for future investigation of structure-function relationships in
nucleoside transport enabling us to design therapeutics targeted specifically to the ei-
transporters to improve membrane permeability, and to develop ei-type transporter
inhibitors with higher potency and specificity. In a nutshell, equilibrative nucleoside
transporters continue to be a potential target for rational drug development for enhanced
systemic delivery.
With the cloning of two distinct equillibrative nucleoside transporters, a variety of
tools will be developed to characterize individual transporter subtypes, including specific
antibodies and probes to delete/knock-down expression of each subtype. These tools are
beginning to provide fundamental information that has the potential to impact our
understanding of ENT physiology or pathophysiology. It is unclear, even now, what
would happen in a species ENTx -/- case of a given host (mouse, rat, etc.). The question
of gene redundancy will be answered by using genetic approaches such as targeted gene
ablation, tissue-specific gene knockout and transgenic overexpression. From this
information and more physiological analysis, it will be possible to determine the
significance of these transport proteins.
224
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.1. What is the physiological role of rbENT2/2A? Are there any phenotypes
associated with rbENT dysfunction?
We propose to generate knock-out rabbits to answer questions regarding, but not limited
to, what are the levels of rbENT2 and rbENT2A mRNA or protein, trafficking, as well as
in their regional, cellular, and subcellular organization. The gene inactivation procedure
in vivo, or knock out, permits the creation of new strains of rabbits specifically lacking
rbENTx genes. Unlike pharmacological approaches this molecular procedure offers a
unique specificity for the target gene. This technique has been applied recently to
inactivate the expression of dopamine receptors (Kelly et al, 1998; Rietschel et al.,
2000), the monoamines vesicular transporter (Fumagalli et al, 1999) and the plasma
membrane dopamine transporter (Gainetdinov et al, 1999; Moron et al, 2002).
Transgenic animals carrying these "genetic defects" leading to a better understanding of
the relative contribution of each of the corresponding gene product regarding regulation
of the expression of transporter proteins and responses to various drugs targeting the
nucleoside transporter system. These strains of rabbits will offer unique models to test the
specificity and selectivity of nucleoside analog drugs and have often provided key
elements leading to possible clinical implications for illnesses.
This is a brief outline of the steps necessary to produce rabbits with a mutation
targeted to a specific gene, in this case rbENT2 (rbENT2A presumably comes from the
same source). The broad outline of the experiment includes: 1) the investigator constructs
a gene targeting vector containing a mutation in the target gene; 2) the targeting vector is
introduced into embryonic stem (ES) cells; 3) ES cell clones which undergo homologous
225
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
recombination with the targeting vector are identified; 4) ES cell-rabbit chimeras are
produced; 5) the chimeras breed and transmit the chromosome with the targeted gene to
their offspring; 6) homozygous animals are produced from the mating of hemizygous
chimera offspring; and 7) the phenotype resulting from the genetic mutation is
characterized.
The genetic and molecular analysis of the equilibrative nucleoside transporters is
yet to be thoroughly accomplished. Once their physiological roles are well defined, it
would be easier to speculate about their impact on the pharmacokinetics,
pharmacodynamics, and pharmacogenomics aspects of various drugs that fall within their
spectrum of substrates.
3.2. New sorting motif via alternative splicing?
The growing molecular identification of various epithelial transporter genes is revealing
that alternative splicing is common even among transporters. Physiological consequences
of alternative splicing, in certain known genes encoding transporter proteins in which
spliced isoforms have recently been identified, are still largely being studied. In some
cases, the spliced isoforms result in nonfunctional proteins, which possess a dominant
negative effect on the cotransporter function, suggesting that the presence of such
isoforms can be important in the functional regulation of the transporter. In most
transporter genes, the spliced isoforms have been shown to be functional, resulting in a
variety of physiological consequences, including, for example, changes in the polarization
of isoforms to the apical or basolateral membrane, changes in pharmacological or kinetic
properties, and changes in tissue distribution or intracellular localization. In some cases,
226
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
although the spliced isoform is functional, the consequence of splicing is still unknown.
Different regulation among isoforms enhanced by alternative splicing mechanisms is an
interesting possibility that adds to the diversity of transporter proteins.
Pre- and posttranslational regulation of ENT expression, the role of gene
amplification in increased expression, and the phenotypes of transgenic and "knock-out"
animals are only a few possible molecular leads that need further investigation. The need
for such data can be clearly seen by the lack of information required to fairly and
conclusively explain previous observations. For example, a “knock-down” scenario of
tight junctions (ZO, zonula occludens) in MDCK cells is required to somehow establish
the validity of our reasoning as to why rbENT2A was apparently observed as
nonfunctional in HEK293 cells - the idea being that these cells lacked the polarity
offered by tight junction formation that clearly separates the apical domain from the
basalateral one. Hence the spliced variant appeared to be functional in MDCK cells, but
non-functional in HEK293 cells.
3.3. New functional characteristics via alternative splicing?
Development of a cell line stably-transfected and overexpressing rbENT2 and/or
rbENT2A is needed to further characterize more subtle details about each transporter.
Inhibitory profile of nucleoside analogs performed in such a system will be of
importance, because it will provide more constant results about the individual
characteristic of each transporter, without the difficulties associated with transient
transfections (efficiency, toxicity, etc.). Hence, structure-function analysis of ENT
activity can be readily accomplished in the near future, as more data becomes available
227
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
from cellular uptake studies. In these situations, when specific ENT isoforms are
expressed in “clean” expression systems and the uptake of nucleoside analogues with
very diverse functional group modifications is evaluated, powerful databases can be
compiled which with the help of computers will be able to accurately predict the success
of future drug candidates in being nucleoside transporter substrates.
3.3.1. Implications in modulation by physiological machinery
Pharmacological modulation, and transcriptional and post-translational regulation of
individual transporters can be evaluated in such systems with greater consistency and
accuracy as well. A number of control experiments need to be done along these lines to
provide information that will fill some gaps in linking second messengers to each
transporter, rbENT2/2A. Our current conclusions included, but were not limited to the
modulation of basolateral ENT like nucleoside transport by Ca2 + and PKA.
However we need direct controls for each implicated event, such as an experiment
relatively simple to do by removing Ca2 + from the incubation buffer, followed by a repeat
in the previously described study. Additionally, the activation and effect (or the lack of
effect, as in the case of PKC) of protein kinases needs to be controlled as well, possibly
in further repeats of previously described studies, where in addition to adding
pharmacological agents that stimulate kinases, we would monitor a general
phosphorylation of histones as a control. Moreover, if all these studies are done in
overexpression conditions, valuable molecular information can be obtained about each
ENT isoform identified in these studies.
The regulatory role of second messengers must be clarified in more detail. For
228
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
example, in studies where events other than phosphorylation of rbENT2/2A were
implied, the prospect of an existing intracellular pool must be confirmed. Direct
measurements of single cell membrane capacitance after stimulation of desired cells with
drugs that were previously implied in the recruitment of active transporters from the cell
membrane (hence lowering Vm a x ) will provide additional evidence of this hypothesis.
Likewise, these types of measurements can be used as a negative control in cases where
we hypothesized that protein phosphorylation was taking place. Enough time must be
allowed for the membrane to completely turn over (20 to 30 min), with proper controls,
in membrane capacitance measurements proposed above.
3.4. What is the endogenous expression and distribution profile of rbENT2/2A?
To localize the segmental location of rbENT2 isoforms in trachea, and to examine
polarized distribution of transporter proteins at the steady state, molecular cloning has led
us to develop immunological tools such as isoform-specific antibodies, for the
investigation of transporter distribution and the assessment of transporter abundance.
There is a reported correlation between cell-surface abundance of es transporters and in
vitro sensitivity of cells to nucleoside drugs (Gati et al., 1998). To date, there are no
direct studies examining the distribution of nucleoside transporters with specific
antibodies in rabbit trachea. First, we will localize the mRNA transcripts of the rbENT2
isoforms in the segmental rabbit trachea, in situ hybridization will be performed, which
has been used to localize the distribution of a number of transporters (Ritzel et al, 1998;
Varoqui and Erickson, 2002). Second, we will raise isoform-specific antibodies to define
the polarity and intracellular distribution of membrane transporters. Finally, we will
229
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
perform cell-surface domain-selective biotinylation to support i mmunohistochemical and
confocal analyses, and to quantify cell-surface abundance of these transporters.
3.4.1. Development of antibodies for immunodetection
Antibodies that are specific for the rbENT2/2A proteins are used as immunological tools
to determine the distribution of the equilibrative nucleoside transporter proteins in the
major divisions of the rabbit respiratory tract. Enriched membrane fractions from regions
of trachea, bronchi, alveolar cells must be prepared from rabbit. The trachea membrane
preparations are assessed for the presence of specific cellular markers. The relative
abundance of the rbENT2 and rbENT2A proteins, as determined by densitometry of
immunoblots will be recorded.
Polyclonal antibodies specific for the rabbit ENT2 protein can be produced in
viral-free male New Zealand rabbits (10-20 kg) with a fusion protein corresponding to
the entire intracellular loop region of rbENT2. This region is not common between the
wild type and spliced variant, so it can be used to resolve the two orthologues. Using a
common strategy, a fusion polypeptide containing maltose binding protein (MBP) and
specific amino acid residues of rbENT2 will be produced in Escherichia coli and purified
on an amylose resin. The amino acid residues will need to span the putative alternative
splicing region (after amino acid residue 289), in order to generate two mutually
exclusive antibodies. A fusion polypeptide containing glutathione S-transferase (GST)
and the desired peptide sequences of rbENT2 and rbeENT2A will also be produced in E.
coli and purified on GST sepharose beads. The purified MBP-rbENT2 fusion polypeptide
will be used as the immunogen, while the GST-rbENT2 fusion polypeptide will be used
230
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
for preliminary screening by ELISA and immunoblotting of the wild type and spliced
variant. Hence there will be a pair for each immunogens described above. Another major
problem for the field of equilibrative nucleoside transport is a severe paucity of data on
the structure of these proteins and their arrangement in membranes. There is a need for
epitope-specific antisera to distinguish the rbENT2/2A proteins and reagents that will
specifically block transport to each of these proteins.
3.4.2. Additional applications of antibodies
Cell surface biotinylation and Western blot analysis will be used to estimate the relative
expression levels of rbENT2 isoforms on each domain of tracheal epithelial cells. Cells
will be cooled to 4°C and either the apical or basolateral domain will be derivatized with
a membrane-impermeable biotin analog sulfo-N-hydroxysuccinimidyl-biotin (sulfo-
NHS-biotin). Cells will then be lysed, incubated with streptavidin-agarose beads,
subjected to SDS-PAGE, and blotted with isoform-specific antibodies for
immunoblotting. Upon completing this goal, we will gain a better understanding of the
regional localization and the polarized distribution of different nucleoside transporter
isoforoms in rabbit trachea. Studies will provide us with important information on the
physiological roles of these transporter proteins in the translocation of nucleosides.
3.5. What is the consequence of molecular and structural markers due to alternative
splicing?
Domain swapping of predicted crucial TMD 7 between rbENT2A and rbENT2, will be a
key experiment to perform in order to validate our hypothesis regarding the regulatory
231
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
role of this domain in the splice variant (in terms of kinetics and pharmacology of
transport, as well as subcellular spatial distribution). Such experiments would be very
interesting, especially, if we can successfully replace the TMD 7 of rbENT2 with the
TMD 7 of rbENT2A (and vice versa) to see whether or not it may affect the functional
characteristics of each transporter. If this membrane spanning domain is required for
rbENT2A to be like an es-type ENT, as suggested by the data presented earlier, this
property should be reflected by the chimera. On the other hand, if we insert TMD 7 of
rbENT2 (an ei-type ENT) in the place of the splice variant sequence, one would expect
similar results as in the former scenario.
3.5.1. Biotech approach
To further examine the transmembrane topology of the rbENT2/2A transporter, a series
of constructs will be generated in which the transporter sequence will be truncated at the
COOH-terminal region of each of the putative hydrophilic loops connecting the putative
hydrophobic domains. Each of the truncated transporter fragments will be ligated into a
vector (pSG5-HA) in-frame. All plasmid constructs will be verified for correct
orientation by restriction enzyme analysis and sequence analysis. This HA fragment lacks
intrinsic translocation activity and has previously been shown to serve as a faithful
reporter for translocation when following topogenic sequences in chimeric proteins in our
study. Following transient expression of HA-tagged transporter chimeras in MDCK cells,
the transmembrane orientation of each chimera will be determined by immunofluorescent
staining under membrane permeabilized and non-permeabilized conditions. These studies
will nicely compensate our previously reported results when various biochemical
232
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
treatments were used as probes for structural validity in rbENT2/2A.
3.5.2. In S i l i c o approach
The computer-aided predictions about rbENT2/2A must be further developed into
working models. That may help to answer certain critical questions such as 1) where are
the putative structural binding domains in which substrates (e.g. uridine) or inhibitors
(e.g. NBMPR) would interact with rbENT2/2A? 2) What are the essential amino acid
residues of the cloned nucleoside transporters responsible for substrate selectivity? 3)
Can any good correlation be obtained between minimal energy of substrate-transporter
interaction predicted by computer modeling and IC5 0 values of substrates obtained by
functional characterization? All these in silico predictions about rbENT2/2A can be
incorporated into working models.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
VII. REFERENCES
M o v e r s , s h a k e r s a n d p io n e e r s.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Abbracchio, M.P. and Bumstock, G. (1998). Purinergic signalling: pathophysiological
roles. J. Pharmacol. 78, 113-145.
Abd-Elfattah, A.S., Jessen, M.E., Lekven, J., and Wechsler, A.S. (1998). Differential
cardioprotection with selective inhibitors of adenosine metabolism and transport: role of
purine release in ischemic and reperfusion injury. Mol. Cell Biochem. 180,179-191.
Abdullah, L.H., Conway, J.D., Cohn, J.A., and Davis, C.W. (1997). Protein kinase C and
Ca2 + activation of mucin secretion in airway goblet cells. Am. J. Physiol. 273, L201-
L210.
Anderson, C.M., Baldwin, S.A., Young, J.D., Cass, C.E., and Parkinson, F.E. (1999a).
Distribution of mRNA encoding a nitrobenzylthioinosine-insensitive nucleoside
transporter (ENT2) in rat brain. Brain Res. Mol. Brain Res. 70, 293-297.
Anderson, C.M., Xiong, W., Geiger, J.D., Young, J.D., Cass, C.E., Baldwin, S.A., and
Parkinson, F.E. (1999b). Distribution of equilibrative, nitrobenzylthioinosine-sensitive
nucleoside transporters (ENT1) in brain. J. Neurochem. 73, 867-873.
Aulak, K.S., Mishra, R., Zhou, L., Hyatt, S.L., de Jonge, W., Lamers, W., Snider, M., and
Hatzoglou, M. (1999). Post-transcriptional regulation of the arginine transporter Cat-1 by
amino acid availability. J. Biol. Chem. 274, 30424-30432.
Baldwin, S.A., Mackey, J.R., Cass, C.E., and Young, J.D. (1999). Nucleoside
transporters: molecular biology and implications for therapeutic development. Mol. Med.
Today 5, 216-224.
Ballarin, M., Fredholm, B.B., Ambrosio, S., and Mahy, N. (1991). Extracellular levels of
adenosine and its metabolites in the striatum of awake rats: inhibition of uptake and
metabolism. Acta Physiol. Scand. 142, 97-103.
Barros, L.F., Yudilevich, D.L., Jarvis, S.M., Beaumont, N., Young, J.D., and Baldwin,
S.A. (1995). Immunolocalisation of nucleoside transporters in human placental
trophoblast and endothelial cells: evidence for multiple transporter isoforms. Pflugers
Arch. 429,394-399.
Benjamins, J.A. and Nedelkoska, L. (1994). Maintenance of membrane sheets by
cultured oligodendrocytes requires continuous microtubule turnover and Golgi transport.
Neurochem. Res. 19, 631-639.
Bennett, E.R. and Kanner, B.I. (1997). The membrane topology of GAT-1, a (Na+ + d e
coupled gamma-aminobutyric acid transporter from rat brain. J. Biol. Chem. 272,1203-
1210.
Bissonnette, P., Noel, J., Coady, M.J., and Lapointe, J.Y. (1999). Functional expression
of tagged human Na+ -glucose cotransporter in Xenopus laevis oocytes. J. Physiol. 520,
235
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
359-371.
Boleti, H., Coe, I.R., Baldwin, S.A., Young, J.D., and Cass, C.E. (1997). Molecular
identification of the equilibrative NBMPR-sensitive (es) nucleoside transporter and
demonstration of an equilibrative NBMPR-insensitive (ei) transport activity in human
erythroleukemia (K562) cells. Neuropharmacol. 3 6,1167-1179.
Bonifacino, J.S. and Weissman, A.M. (1998). Ubiquitin and the control of protein fate in
the secretory and endocytic pathways. Annu. Rev. Cell Dev. Biol. 14, 19-57.
Bouma, M.G., Stad, R.K., van den Wildenberg, F.A., and Buurman, W.A. (1994).
Differential regulatory effects of adenosine on cytokine release by activated human
monocytes. J. Immunol. 753,4159-4168.
Boumah, C.E., Hogue, D.L., and Cass, C.E. (1992). Expression of high levels of
nitrobenzylthioinosine-sensitive nucleoside transport in cultured human choriocarcinoma
(BeWo) cells. Biochem. J. 288, 987-996.
Braiman, L., Sheffi-Friedman, L., Bak, A., Tennenbaum, T., and Sampson, S.R. (1999).
Tyrosine phosphorylation of specific protein kinase C isoenzymes participates in insulin
stimulation of glucose transport in primary cultures of rat skeletal muscle. Diabetes 48,
1922-1929.
Brett, C.M., Washington, C.B., Ott, R.J., Gutierrez, M.M., and Giacomini, K.M. (1993).
Interaction of nucleoside analogues with the sodium-nucleoside transport system in brush
border membrane vesicles from human kidney. Pharm. Res. 10,423-426.
Brown, J.R., Cornell, K., and Cook, P.W. (2000). Adenosine- and adenine-nucleotide-
mediated inhibition of normal and transformed keratinocyte proliferation is dependent
upon dipyridamole-sensitive adenosine transport. J. Invest. Dermatol. 115, 849-859.
Brundege, J.M. and Dunwiddie, T.V. (1997). Role of adenosine as a modulator of
synaptic activity in the central nervous system. Adv. Pharmacol. 39, 353-391.
Bucci, C., Thomsen, P., Nicoziani, P., McCarthy, J., and van Deurs, B. (2000). Rab7: a
key to lysosome biogenesis. Mol. Biol. Cell 11,467-480.
Bumstock, G. (1990). Overview. Purinergic mechanisms. Ann. N. Y. Acad. Sci. 603,1-
17.
Caldenhoven, E., van Dijk, T.B., Solari, R., Armstrong, J., Raaijmakers, J.A., Lammers,
J.W., Koenderman, L., and de Groot, R.P. (1996). STAT3j3, a splice variant of
transcription factor STAT3, is a dominant negative regulator of transcription. J. Biol.
Chem. 271, 13221-13227.
Camins, A., Jimenez, A., Sureda, F.X., Pallas, M., Escubedo, E., and Camarasa, J.
(1996). Characterization of nitrobenzylthioinosine binding sites in the mitochondrial
236
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
fraction of rat testis. Life Sci. 58, 753-759.
Cass, C.E., Young, J.D., Baldwin, S.A., Cabrita, M.A., Graham, K.A., Griffiths, M.,
Jennings, L.L., Mackey, J.R., Ng, A.M., Ritzel, M.W., Vickers, M.F., and Yao, S.Y.
(1999). Nucleoside transporters of mammalian cells. Pharm. Biotechnol. 12, 313-352.
Chandrasena, G., Giltay, R., Patil, S.D., Bakken, A., and Unadkat, J.D. (1997).
Functional expression of human intestinal Na+ -dependent and Na+ -independent
nucleoside transporters in Xenopus laevis oocytes. Biochem. Pharmacol. 5 3 ,1909-1918.
Chang, X.B., Hou, Y.X., Jensen, T.J., and Riordan, J.R. (1994). Mapping of cystic
fibrosis transmembrane conductance regulator membrane topology by glycosylation site
insertion. J. Biol. Chem. 269,18572-18575.
Charon, J.P., McCormick, J., Mehta, A., and Kemp, P.J. (1994). Characterization of
sodium-dependent glucose transport in sheep tracheal epithelium. Am. J. Physiol. 267,
L390-L397.
Che, M., Nishida, T., Gatmaitan, Z., and Arias, I.M. (1992). A nucleoside transporter is
functionally linked to ectonucleotidases in rat liver canalicular membrane. J. Biol. Chem.
267, 9684-9688.
Che, M., Ortiz, D.F., and Arias, I.M. (1995). Primary structure and functional expression
of a cDNA encoding the bile canalicular, purine-specific Na+ -nucleoside cotransporter. J.
Biol. Chem. 270,13596-13599.
Che, M., Gatmaitan, Z., and Arias, I.M. (1997). Ectonucleotidases, purine nucleoside
transporter, and function of the bile canalicular plasma membrane of the hepatocyte.
FASEB J. 11, 101-108.
Chesney, R.W., Zelikovic, I., Jones, D.P., Budreau, A., and Jolly, K. (1990). The renal
transport of taurine and the regulation of renal sodium-chloride-dependent transporter
activity. Pediatr. Nephrol. 4, 399-407.
Chiang, C.W., Carter, N., Sullivan, W.J., Jr., Donald, R.G., Roos, D.S., Naguib, F.N., el
Kouni, M.H., Ullman, B., and Wilson, C.M. (1999). The adenosine transporter of
Toxoplasma gondii. Identification by insertional mutagenesis, cloning, and recombinant
expression. J. Biol. Chem. 274, 35255-35261.
Choi, D.S., Handa, M., Young, H., Gordon, A.S., Diamond, I., and Messing, R.O. (2000).
Genomic organization and expression of the mouse equilibrative, nitrobenzylthioinosine-
sensitive nucleoside transporter 1 (ENT1) gene. Biochem. Biophys. Res. Commun. 277,
200-208.
Chunn, J.L., Young, H.W., Banerjee, S.K., Colasurdo, G.N., and Blackburn, M.R.
(2001). Adenosine-dependent airway inflammation and hyperresponsiveness in partially
adenosine deaminase-deficient mice. J. Immunol. 167, 4676-4685.
237
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Closs, E.I., Lyons, C.R., Kelly, C., and Cunningham, J.M. (1993). Characterization of the
third member of the MCAT family of cationic amino acid transporters. Identification of a
domain that determines the transport properties of the MCAT proteins. J. Biol. Chem.
268, 20796-20800.
Coe, I.R., Griffiths, M., Young, J.D., Baldwin, S.A., and Cass, C.E. (1997). Assignment
of the human equilibrative nucleoside transporter (hENTl) to 6p21.1-p21.2. Genomics
45,459-460.
Covitz, K.M., Amidon, G.L., and Sadee, W. (1998). Membrane topology of the human
dipeptide transporter, hPEPTl, determined by epitope insertions. Biochemistry 37,
15214-15221.
Craig, C.G. and White, T.D. (1993). N-methyl-D-aspartate- and non-N-methyl-D-
aspartate-evoked adenosine release from rat cortical slices: distinct purinergic sources
and mechanisms of release. J. Neurochem. 60,1073-1080.
Craig, J.E., Zhang, Y., and Gallagher, M.P. (1994). Cloning of the nupC gene of
Escherichia coli encoding a nucleoside transport system, and identification of an adjacent
insertion element, IS 186. Mol. Microbiol. 11, 1159-1168.
Crawford, C.R., Patel, D.H., Naeve, C., and Belt, J.A. (1998). Cloning of the human
equilibrative, nitrobenzylmercaptopurine riboside (NBMPR)-insensitive nucleoside
transporter ei by functional expression in a transport-deficient cell line. J. Biol. Chem.
273, 5288-5293.
Cronstein, B.N. (1994). Adenosine, an endogenous anti-inflammatory agent. J. Appl.
Physiol. 76, 5-13.
Dagnino, L. and Paterson, A.R. (1990). Sodium-dependent and equilibrative nucleoside
transport systems in L1210 mouse leukemia cells: effect of inhibitors of equilibrative
systems on the content and retention of nucleosides. Cancer Res. 50, 6549-6553.
Dagnino, L., Bennett, L.L., Jr., and Paterson, A.R. (1991). Sodium-dependent nucleoside
transport in mouse leukemia L1210 cells. J. Biol. Chem. 266, 6308-6311.
Dancis, J., Lee, J.D., Mendoza, S., and Liebes, L. (1993). Transfer and metabolism of
dideoxyinosine by the perfused human placenta. J. Acquir. Immune Defic. Syndr. 6, 2-6.
Davis, K.E., Straff, D.J., Weinstein, E.A., Bannerman, P.G., Correale, D.M., Rothstein,
J.D., and Robinson, M.B. (1998). Multiple signaling pathways regulate cell surface
expression and activity of the excitatory amino acid carrier 1 subtype of Glu transporter
in C6 glioma. J. Neurosci. 18, 2475-2485.
del Santo, B., Valdes, R., Mata, J., Felipe, A., Casado, F.J., and Pastor-Anglada, M.
(1998). Differential expression and regulation of nucleoside transport systems in rat liver
parenchymal and hepatoma cells. Hepatology 2 8 ,1504-1511.
238
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Dresser, M.J., Gerstin, K.M., Gray, A.T., Loo, D.D., and Giacomini, K.M. (2000).
Electrophysiological analysis of the substrate selectivity of a sodium-coupled nucleoside
transporter (rCNTl) expressed in Xenopus laevis oocytes. Drug Metab. Dispos. 28, 1135-
1140.
Dyer, J., Hosie, K.B., and Shirazi-Beechey, S.P. (1997). Nutrient regulation of human
intestinal sugar transporter (SGLT1) expression. Gut 41, 56-59.
Eckhardt, M., Gotza, B., and Gerardy-Schahn, R. (1999). Membrane topology of the
mammalian CMP-sialic acid transporter. J. Biol. Chem. 274, 8779-8787.
Ellgaard, L., Molinari, M., and Helenius, A. (1999). Setting the standards: quality control
in the secretory pathway. Science 286,1882-1888.
Felipe, A., Valdes, R., Santo, B., Lloberas, J., Casado, J., and Pastor-Anglada, M. (1998).
Na+ -dependent nucleoside transport in liver: two different isoforms from the same gene
family are expressed in liver cells. Biochem. J. 330, 997-1001.
Feoktistov, I. and Biaggioni, I. (1997). Adenosine A2B receptors. Pharmacol. Rev. 49,
381-402.
Fideu, M.D. and Miras-Portugal, M.T. (1992). Long term regulation of nucleoside
transport by thyroid hormone (T3) in cultured chromaffin cells. Neurochem. Res. 17,
1099-1104.
Flanagan, S.A. and Meckling-Gill, K.A. (1997). Characterization of a novel Na+ -
dependent, guanosine-specific, nitrobenzylthioinosine-sensitive transporter in acute
promyelocytic leukemia cells. J. Biol. Chem. 272,18026-18032.
Foga, I.O., Geiger, J.D., and Parkinson, F.E. (1996). Nucleoside transporter-mediated
uptake and release of [3 H]L-adenosine in DDT1 MF-2 smooth muscle cells. Eur. J.
Pharmacol. 318, 455-460.
Fowler, J.C. (1993). Changes in extracellular adenosine levels and population spike
amplitude during graded hypoxia in the rat hippocampal slice. Naunyn Schmiedebergs
Arch. Pharmacol. 347, 73-78.
Fozard, J.R. and Hannon, J.P. (1999). Adenosine receptor ligands: potential as
therapeutic agents in asthma and COPD. Pulm. Pharmacol. Ther. 12, 111-114.
Freitag, A., Wessler, I., and Racke, K. (1996). Adrenoceptor- and cholinoceptor-mediated
mechanisms in the regulation of 5-hydroxytryptamine release from isolated tracheae of
newborn rabbits. Br. J. Pharmacol. 119, 91-98.
Fumagalli, F., Gainetdinov, R.R., Wang, Y.M., Valenzano, K.J., Miller, G.W., and
Caron, M.G. (1999). Increased methamphetamine neurotoxicity in heterozygous vesicular
monoamine transporter 2 knock-out mice. J. Neurosci. 19, 2424-2431.
239
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Gainetdinov, R.R., Jones, S.R., and Caron, M.G. (1999). Functional hyperdopaminergia
in dopamine transporter knock-out mice. Biol. Psychiatry 46, 303-311.
Gati, W.P., Paterson, A.R., Belch, A.R., Chlumecky, V., Larratt, L.M., Mant, M.J., and
Turner, A.R. (1998). es nucleoside transporter content of acute leukemia cells: role in cell
sensitivity to cytarabine (araC). Leuk. Lymphoma 32,45-54.
Gilbertsen, R.B. (1987). Adenosine and adenosine receptors in immune function.
Minireview and meeting report. Agents Actions 22, 91-98.
Gilstrap, L.C., Bawdon, R.E., Roberts, S.W., and Sobhi, S. (1994). The transfer of the
nucleoside analog ganciclovir across the perfused human placenta. Am. J. Obstet.
Gynecol. 170, 967-972.
Gomez-Angelats, M., del Santo, B., Mercader, J., Ferrer-Martinez, A., Felipe, A.,
Casado, J., and Pastor-Anglada, M. (1996). Hormonal regulation of concentrative
nucleoside transport in liver parenchymal cells. Biochem. J. 313, 915-920.
Griffith, D.A., Conant, A.R., and Jarvis, S.M. (1990). Differential inhibition of
nucleoside transport systems in mammalian cells by a new series of compounds related to
lidoflazine and mioflazine. Biochem. Pharmacol. 40, 2297-2303.
Griffith, D.A. and Jarvis, S.M. (1996). Nucleoside and nucleobase transport systems of
mammalian cells. Biochim. Biophys. Acta 1286,153-181.
Griffiths, M., Beaumont, N., Yao, S.Y., Sundaram, M., Boumah, C.E., Davies, A.,
Kwong, F.Y., Coe, I, Cass, C.E., Young, J.D., and Baldwin, S.A. (1997a). Cloning of a
human nucleoside transporter implicated in the cellular uptake of adenosine and
chemotherapeutic drugs. Nat. Med. 3, 89-93.
Griffiths, M., Yao, S.Y., Abidi, F., Phillips, S.E., Cass, C.E., Young, J.D., and
Baldwin,S.A. (1997b). Molecular cloning and characterization of a
nitrobenzylthioinosine-insensitive (ei) equilibrative nucleoside transporter from human
placenta. Biochem. J. 328, 739-743.
Gu, J.G., Foga, I.O., Parkinson, F.E., and Geiger, J.D. (1995). Involvement of
bidirectional adenosine transporters in the release of L-[3 H] adenosine from rat brain
synaptosomal preparations. J. Neurochem. 64, 2105-2110.
Gutierrez, M.M., and Giacomini, K.M. (1993). Substrate selectivity, potential sensitivity
and stoichiometry of Na+ -nucleoside transport in brush border membrane vesicles from
human kidney. Biochim. Biophys. Acta 1149, 202-208.
Hamilton, S.R., Yao, S.Y., Ingram, J.C., Hadden, D.A., Ritzel, M.W., Gallagher, M.P.,
Henderson, P.J., Cass, C.E., Young, J.D., and Baldwin, S.A. (2001). Subcellular
distribution and membrane topology of the mammalian concentrative Na+ -nucleoside
cotransporter rCNTl. J. Biol. Chem. 276, 27981-27988.
240
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hammond, J.R. (2000). Interaction of a series of draflazine analogues with equilibrative
nucleoside transporters: species differences and transporter subtype selectivity. Naunyn
Schmiedebergs Arch. Pharmacol. 361, 373-382.
Handa, M., Choi, D., Caldeiro, R.M., Messing, R.O., Gordon, A.S., and Diamond, I., I
(2001). Cloning of a novel isoform of the mouse NBMPR-sensitive equilibrative
nucleoside transporter (ENT1) lacking a putative phosphorylation site. Gene 262, 301-
307.
Harkema, J.R., Mariassy, A.T., St. George, J., Hyde, D.M., and Plopper, C.G. (1991).
Epithelial cells of the conducting airway. In the airway epithelium. Farmers, S.G. and
Hay, D.W., eds. New York: Marcel Dekker, 3-40.
Hasko, G., Szabo, C., Nemeth, Z.H., Kvetan, V., Pastores, S.M., and Vizi, E.S. (1996).
Adenosine receptor agonists differentially regulate IL-10, TNF-alpha, and nitric oxide
production in RAW 264.7 macrophages and in endotoxemic mice. J. Immunol. 157,
4634-4640.
Hille, B. (1992). Ionic channels of excitable membranes. Introduction. Sinauer
Associates, Inc., Sunderland, 1-20.
Heidmann, D.E., Szot, P., Kohen, R., and Hamblin, M.W. (1998). Function and
distribution of three rat 5-hydroxytryptamine7 (5-HT7) receptor isoforms produced by
alternative splicing. Neuropharmacol. 37, 1621-1632.
Hofmann, K. and Stoffel, W. (1993). TMbase - A database of membrane spanning
protein segments. Biol. Chem. Hoppe Seyler 374, 166.
Hollmann, M., Maron, C., and Heinemann, S. (1994). N-glycosylation site tagging
suggests a three transmembrane domain topology for the glutamate receptor GluRl.
Neuron 13,1331-1343.
Hresko, R.C., Kruse, M., Strube, M., and Mueckler, M. (1994). Topology of the Glut 1
glucose transporter deduced from glycosylation scanning mutagenesis. J. Biol. Chem.
269, 20482-20488.
Huang, Q.Q., Yao, S.Y., Ritzel, M.W., Paterson, A.R., Cass, C.E., and Young, J.D.
(1994). Cloning and functional expression of a complementary DNA encoding a
mammalian nucleoside transport protein. J. Biol. Chem. 269,17757-17760.
Hurtley, S.M. and Helenius, A. (1989). Protein oligomerization in the endoplasmic
reticulum. Annu. Rev. Cell Biol. 5, 277-307.
Hyatt, S.L., Aulak, K.S., Malandro, M., Kilberg, M.S., and Hatzoglou, M. (1997).
Adaptive regulation of the cationic amino acid transporter-1 (Cat-1) in FAO cells. J. Biol.
Chem. 272, 19951-19957.
241
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hyde, R.J., Cass, C.E., Young, J.D., and Baldwin, S.A. (2001). The ENT family of
eukaryote nucleoside and nucleobase transporters: recent advances in the investigation of
structure/function relationships and the identification of novel isoforms. Mol. Membr.
Biol. 18, 53-63.
Ihara, T., Tsujikawa, T., Fujiyama, Y., and Bamba, T. (2000). Regulation of PepTl
peptide transporter expression in the rat small intestine under malnourished conditions.
Digestion 61, 59-67.
Ivanova, L., Barmada, S., Kummer, T., and Harris, D.A. (2001). Mutant prion proteins
are partially retained in the endoplasmic reticulum. J. Biol. Chem. 276,42409-42421.
Jakobs, E.S. and Paterson, A.R. (1986). Sodium-dependent, concentrative nucleoside
transport in cultured intestinal epithelial cells. Biochem. Biophys. Res. Commun. 140,
1028-1035.
Jarvis, S.M., McBride,D., and Young, J.D. (1982). Erythrocyte nucleoside transport:
asymmetrical binding of nitrobenzylthioinosine to nucleoside permeation sites. J. Physiol.
324, 31-46.
Jarvis, S.M., Janmohamed, S.N., and Young, J.D. (1983). Kinetics of
nitrobenzylthioinosine binding to the human erythrocyte nucleoside transporter.
Biochem. J. 216, 661-667.
Jarvis, S.M. and Young, J.D. (1986). Nucleoside transport in rat erythrocytes: two
components with differences in sensitivity to inhibition by nitrobenzylthioinosine and p-
chloromercuriphenyl sulfonate. J. Membr. Biol. 93, 1-10.
Jennings, L.L., Hao, C., Cabrita, M.A., Vickers, M.F., Baldwin, S.A., Young, J.D., and
Cass, C.E. (2001). Distinct regional distribution of human equilibrative nucleoside
transporter proteins 1 and 2 (hENTl and hENT2) in the central nervous system.
Neuropharmacol. 40, 722-731.
Jones, K.W., Rylett, R.J., and Hammond, J.R. (1994). Effect of cellular differentiation on
nucleoside transport in human neuroblastoma cells. Brain Res. 660,104-112.
Kelly, M.A., Rubinstein, M., Phillips, T.J., Lessov, C.N., Burkhart-Kasch, S., Zhang, G.,
Bunzow, J.R., Fang, Y., Gerhardt, G.A., Grandy, D.K., and Low, M.J. (1998). Locomotor
activity in D2 dopamine receptor-deficient mice is determined by gene dosage, genetic
background, and developmental adaptations. J. Neurosci. 18, 3470-3479.
Kilberg, M.S., Stevens, B.R., and Novak, D.A. (1993). Recent advances in mammalian
amino acid transport. Annu. Rev. Nutr. 13, 137-165.
Kim, K.M., Kingsmore, S.F., Han, H., Yang-Feng, T.L., Godinot, N., Seldin, M.F.,
Caron, M.G., and Giros, B. (1994). Cloning of the human glycine transporter type 1:
molecular and pharmacological characterization of novel isoform variants and
242
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
chromosomal localization of the gene in the human and mouse genomes. Mol.
Pharmacol. 45, 608-617.
Kiss, A., Far ah, K., Kim, J., Ganiock, R.J., Drysdale, T.A., and Hammond, J.R. (2000).
Molecular cloning and functional characterization of inhibitor-sensitive (mENTl) and
inhibitor-resistant (mENT2) equilibrative nucleoside transporters from mouse brain.
Biochem. J. 352, 363-372.
Knight, D., Zheng, X., Rocchini, C., Jacobson, M., Bai, T., and Walker, B. (1997).
Adenosine A3 receptor stimulation inhibits migration of human eosinophils. J. Leukoc.
Biol. 62, 465-468.
Kobayashi, S., Conforti, L., Pun, R.Y., and Millhom, D.E. (1998). Adenosine modulates
hypoxia-induced responses in rat PC 12 cells via the A2A receptor. J. Physiol. 508, 95-
107.
Kwong, F.Y., Wu, J.S., Shi, M.M., Fincham, H.E., Davies, A., Henderson, P.J., Baldwin,
S.A., and Young, J.D. (1993). Enzymic cleavage as a probe of the molecular structures of
mammalian equilibrative nucleoside transporters. J. Biol. Chem. 268, 22127-22134.
Latini, S. and Pedata, F. (2001). Adenosine in the central nervous system: release
mechanisms and extracellular concentrations. J. Neurochem. 79,463-484.
Leung, G.P., Ward, J.L., Wong, P.Y., and Tse, C.M. (2001). Characterization of
nucleoside transport systems in cultured rat epididymal epithelium. Am. J. Physiol. Cell
Physiol. 280, C1076-C1082.
Liedtke, C.M. (1992). Electrolyte transport in the epithelium of pulmonary segments of
normal and cystic fibrosis lung. FASEB J. 6, 3076-3084.
Linden, J. (2001). Molecular approach to adenosine receptors: receptor-mediated
mechanisms of tissue protection. Annu. Rev. Pharmacol. Toxicol. 41, 775-787.
Lippincott-Schwartz, J., Roberts, T.H., and Hirschberg, K. (2000). Secretory protein
trafficking and organelle dynamics in living cells. Annu. Rev. Cell Dev. Biol. 16, 557-
589.
Lopez-Navarro, A.T., Gil, A., and Sanchez-Pozo, A. (1995). Deprivation of dietary
nucleotides results in a transient decrease in acid-soluble nucleotides and RNA
concentration in rat liver. J. Nutr. 125, 2090-2095.
Lopez-Navarro, A.T., Ortega, M.A., Peragon, J., Bueno, J.D., Gil, A., and Sanchez-
Pozo,A. (1996). Deprivation of dietary nucleotides decreases protein synthesis in the liver
and small intestine in rats. Gastroenterol. 110,1760-1769.
Lostao, M.P., Mata, J.F., Larrayoz, I.M., Inzillo, S.M., Casado, F.J., and Pastor-Anglada,
M. (2000). Electrogenic uptake of nucleosides and nucleoside-derived drugs by the
243
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
human nucleoside transporter 1 (hCNTl) expressed in Xenopus laevis oocytes. FEBS
Lett. 481,137-140.
Lum, P.Y., Ngo, L.Y., Bakken, A.H., and Unadkat, J.D . (2000). Human intestinal es
nucleoside transporter: molecular characterization and nucleoside inhibitory profiles.
Cancer Chemother. Pharmacol. 45, 273-278.
Mackey, J.R., Mani, R.S., Seiner, M., Mowles, D., Young, J.D., Belt, J.A., Crawford,
C.R., and Cass, C.E. (1998). Functional nucleoside transporters are required for
gemcitabine influx and manifestation of toxicity in cancer cell lines. Cancer Res. 58,
4349-4357.
Mackey, J.R., Yao, S.Y., Smith, K.M., Karpinski, E ., Baldwin, S.A., Cass, C .E ., and
Young, J.D. (1999). Gemcitabine transport in xenopus oocytes expressing recombinant
plasma membrane mammalian nucleoside transporters. J. Natl. Cancer Inst. 91,1876-
1881.
Mailliard, M.E., Stevens, B.R., and Mann, G.E. (1995). Amino acid transport by small
intestinal, hepatic, and pancreatic epithelia. Gastroenterol. 108, 888-910.
Mangravite, L.M., Lipschutz, J.H., Mostov, K.E., and Giacomini, K.M. (2001).
Localization of GFP-tagged concentrative nucleoside transporters in a renal polarized
epithelial cell line. Am. J. Physiol. Renal Physiol. 280, F879-F885.
Mani, R.S., Hammond, J.R., Maijan, J.M., Graham, K.A., Young, J.D., Baldwin, S.A.,
and Cass, C.E. (1998). Demonstration of equilibrative nucleoside transporters (hENTl
and hENT2) in nuclear envelopes of cultured human choriocarcinoma (BeWo) cells by
functional reconstitution in proteoliposomes. J. Biol. Chem. 273, 30818-30825.
Marquardt, D.L., Walker, L.L., and Heinemann, S. (1994). Cloning of two adenosine
receptor subtypes from mouse bone marrow-derived mast cells. J. Immunol. 152, 4508-
4515.
Mathias, N.R., Kim, K.J., Robinson, T.W., and Lee, V.H.L. (1995). Development and
characterization of rabbit tracheal epithelial cell monolayer models for drug transport
studies. Pharm. Res. 12,1499-1505.
Mathias, N.R., Kim, K.-J., and Lee, V.H.L. (1996a). Nucleoside and nucleoside analog
transport in airway epithelial cell culture monolayers. Pharm.Res. 13, S-513 (Abstract).
Mathias, N.R., Kim, K.-J., and Lee, V.H.L. (1996b). Targeted drug delivery to the
respiratory tract: solute permeability of air-interface cultured rabbit tracheal epithelial cell
monolayers. J. Drug Targeting 4, 79-86.
McManaman, J. and Tam, D.A. (1999). Gabapentin for self-injurious behavior in Lesch-
Nyhan syndrome. Pediatr. Neurol. 20, 381-382.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Meisner, D., Pringle, J., and Mezei, M. (1989). Liposomal pulmonary drug delivery. I. In
vivo disposition of atropine base in solution and liposomal form following endotracheal
instillation to the rabbit lung. J . Microencapsul. 6, 379-387.
Melikian, H .E . and Buckley, K.M. (1999). Membrane trafficking regulates the activity of
the human dopamine transporter. J. Neurosci. 19, 7699-7710.
Moron, J.A ., Brockington, A., Wise, R.A., Rocha, B.A., and Hope, B.T. (2002).
Dopamine uptake through the norepinephrine transporter in brain regions with low levels
of the dopamine transporter: evidence from knock-out mouse lines. J. Neurosci. 22,389-
395.
Mun, E.C., Tally, K .J., and Matthews, J.B. (1998). Characterization and regulation of
adenosine transport in T84 intestinal epithelial cells. Am. J. Physiol. 274, G261-G269.
Murray, A.W. (1971). The biological significance of purine salvage. Annu. Rev.
Biochem. 40, 811-826.
Newby, A.C. (1986). How does dipyridamole elevate extracellular adenosine
concentration? Predictions from a three-compartment model of adenosine formation and
inactivation. Biochem. J. 237, 845-851.
Nordeng, T.W. and Bakke, O. (1999). Overexpression of proteins containing tyrosine- or
leucine-based sorting signals affects transferrin receptor trafficking. J . Biol. Chem. 274,
21139-21148.
Olivares, L., Aragon, C., Gimenez, C., and Zafra, F. (1997). Analysis of the
transmembrane topology of the glycine transporter GLYT1. J. Biol. Chem. 272,1211-
1217.
Olsen, M.A., Nawoschik, S.P., Schurman, B.R., Schmitt, H.L., Bumo, M., Smith, D.L.,
and Schechter, L.E. (1999). Identification of a human 5-HT6 receptor variant produced
by alternative splicing. Brain Res. Mol. Brain Res. 64, 255-263.
Osses, N., Pearson, J.D., Yudilevich, D.L., and Jarvis, S.M. (1996). Hypoxanthine enters
human vascular endothelial cells (ECV 304) via the nitrobenzylthioinosine-insensitive
equilibrative nucleoside transporter. Biochem. J. 317, 843-848.
Pagani, F., Buratti, E., Stuani, C., Romano, M., Zuccato, E., Niksic, M., Giglio, L.,
Faraguna, D., and Baralle, F.E. (2000). Splicing factors induce cystic fibrosis
transmembrane regulator exon 9 skipping through a nonevolutionary conserved intronic
element. J. Biol. Chem. 275, 21041-21047.
Pajor, A.M. (1998). Sequence of a pyrimidine-selective Na+ /nucleoside cotransporter
from pig kidney, pkCNTl. Biochim. Biophys. Acta 1415, 266-269.
Parker, M.D., Ourmozdi, E.P., and Tanner, M.J. (2001). Human BTR1, a new
245
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
bicarbonate transporter superfamily member and human AE4 from kidney. Biochem.
Biophys. Res. Commun. 282,1103-1109.
Parkinson, F.E. and Geiger, J.D. (1996). Effects of iodotubercidin on adenosine kinase
activity and nucleoside transport in DDT1 MF-2 smooth muscle cells. J. Pharmacol. Exp.
Ther. 277, 1397-1401.
Pastor-Anglada, M., Felipe, A., and Casado, F.J. (1998). Transport and mode of action of
nucleoside derivatives used in chemical and antiviral therapies. Trends Pharmacol. Sci.
19,424-430.
Pastor-Anglada, M., Casado, F.J., Valdes, R., Mata, J., Garcia-Manteiga, J., and Molina,
M. (2001). Complex regulation of nucleoside transporter expression in epithelial and
immune system cells. Mol. Membr. Biol. 18, 81-85.
Patel, D.H., Crawford, C.R., Naeve, C.W., and Belt, J.A. (2000) Cloning, genomic
organization and chromosomal localization of the gene encoding the murine sodium-
dependent, purine-selective, concentrative nucleoside transporter (CNT2). Gene 242, 51-
58.
Patil, S.D., Ngo, L.Y., and Unadkat, J.D. (2000). Structure-inhibitory profiles of
nucleosides for the human intestinal N1 and N2 Na+ -nucleoside transporters. Cancer
Chemother. Pharmacol. 46, 394-402.
Paul, B., Chen, M.F., and Paterson, A.R. (1975). Inhibitors of nucleoside transport. A
structure-activity study using human erythrocytes. J. Med. Chem. 18, 968-973.
Penn, R.B., Parent, J.L., Pronin, A.N., Panettieri, R.A., Jr., and Benovic, J.L. (1999).
Pharmacological inhibition of protein kinases in intact cells: antagonism of (3-adrenergic
receptor ligand binding by H-89 reveals limitations of usefulness. J. Pharmacol. Exp.
Ther. 288,428-437.
Phillis, J.W., O'Regan, M.H., and Walter, G.A. (1989). Effects of two nucleoside
transport inhibitors, dipyridamole and soluflazine, on purine release from the rat cerebral
cortex. Brain Res. 481, 309-316.
Pisoni, R.L. and Thoene, J.G. (1989). Detection and characterization of a nucleoside
transport system in human fibroblast lysosomes. J. Biol. Chem. 264,4850-4856.
Plopper, C.G., Mariassy, A.T., Wilson, D.W., Alley, J.L., Nishio, S.J., and Nettesheim, P.
(1983). Comparison of nonciliated tracheal epithelial cells in six mammalian species:
ultrastructure and population densities. Exp. Lung Res. 5, 281-94
Pressacco, J., Mitrovski, B., Erlichman, C., and Hedley, D.W. (1995a). Effects of
thymidylate synthase inhibition on thymidine kinase activity and nucleoside transporter
expression. Cancer Res. 55, 1505-1508.
246
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Pressacco, J., Wiley, J.S., Jamieson, G.P., Erlichman, C., and Hedley, D.W. (1995b).
Modulation of the equilibrative nucleoside transporter by inhibitors of DNA synthesis.
Br. J. Cancer 72, 939-942.
Puchelle, E. (2000). Airway epithelium wound repair and regeneration after injury. Acta
Otorhinolaryngol. Belg. 54,263-270.
Puri, R.N. and Colman, R.W. (1998). Purinergic receptors in human blood platelets:
chemical modification and cloning investigations. J. Protein Chem. 17,429-451.
Rauchwerger, D.R., Firby, P.S., Hedley, D.W., and Moore, M.J. (2000). Equilibrative-
sensitive nucleoside transporter and its role in gemcitabine sensitivity. Cancer Res. 60,
6075-6079.
Reinheimer, T., Mohlig, T., Zimmermann, S., Hohle, K.D., and Wessler, I. (2000).
Muscarinic control of histamine release from airways. Inhibitory Ml-receptors in human
bronchi but absence in rat trachea. Am. J. Respir. Crit Care Med. 162, 534-538.
Renaud, W., Merten, M., and Figarella, C. (1994). Increased coexpression of CFTR and
S100 calcium binding proteins MRP8 and MRP14 mRNAs in cystic fibrosis human
tracheal gland cells. Biochem. Biophys. Res. Commun. 201,1518-1525.
Rennard, S.I., Beckmann, J.D., and Robbins, R.A. (1991). Biology of the airway. The
lung: scientific foundations. Crystal, R.G. and West, J.B., eds., New York: Raven Press,
157-167.
Rietschel, M., Krauss, H., Muller, D.J., Schulze, T.G., Knapp, M., Marwinski, K.,
Maroldt, A.O., Paus, S., Grunhage, F., Propping, P., Maier, W., Held, T., and Nothen,
M.M. (2000). Dopamine D3 receptor variant and tardive dyskinesia. Eur. Arch.
Psychiatry Clin. Neurosci. 250, 31-35.
Ritzel, M.W., Yao, S.Y., Huang, M.Y., Elliott, J.F., Cass, C.E., and Young, J.D. (1997).
Molecular cloning and functional expression of cDNAs encoding a human Na+ -
nucleoside cotransporter (hCNTl). Am. J. Physiol. 272, C707-C714.
Ritzel, M.W., Yao, S.Y., Ng, A.M., Mackey, J.R., Cass, C.E., and Young, J.D. (1998).
Molecular cloning, functional expression and chromosomal localization of a cDNA
encoding a human Na+ /nucleoside cotransporter (hCNT2) selective for purine
nucleosides and uridine. Mol. Membr. Biol. 15, 203-211.
Ritzel, M.W., Ng, A.M., Yao, S.Y., Graham, K., Loewen, S.K., Smith, K.M., Hyde, R.J.,
Karpinski, E., Cass, C.E., Baldwin, S.A., and Young, J.D. (2001a). Recent molecular
advances in studies of the concentrative Na+ -dependent nucleoside transporter (CNT)
family: identification and characterization of novel human and mouse proteins (hCNT3
and mCNT3) broadly selective for purine and pyrimidine nucleosides (system cib). Mol.
Membr. Biol. 18, 65-72.
247
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Ritzel, M.W., Ng, A.M., Yao, S.Y., Graham, K., Loewen, S.K., Smith, K.M., Ritzel,
R.G., Mowles, D.A., Carpenter, P., Chen, X.Z., Karpinski, E., Hyde, R.J., Baldwin, S.A.,
Cass, C.E., and Young, J.D. (2001b). Molecular identification and characterization of
novel human and mouse concentrative Na+ -nucleoside cotransporter proteins (hCNT3
and mCNT3) broadly selective for purine and pyrimidine nucleosides (system cib). J.
Biol. Chem. 276, 2914-2927.
Rongen, G.A., Smits, P., and Thien, T. (1994). Characterization of ATP-induced
vasodilation in the human forearm vascular bed. Circulation 9 0 ,1891-1898.
Ross, R.J., Esposito, N., Shen, X.Y., Von Laue, S., Chew, S.L., Dobson, P.R., Postel-
Vinay, M.C., and Finidori, J. (1997). A short isoform of the human growth hormone
receptor functions as a dominant negative inhibitor of the full-length receptor and
generates large amounts of binding protein. Mol. Endocrinol. 11,265-273.
Rowlands, M.G., Budworth, J., Jarman, M., Hardcastle, I.R., McCague, R., and Gescher,
A. (1995). Comparison between inhibition of protein kinase C and antagonism of
calmodulin by tamoxifen analogues. Biochem. Pharmacol. 50, 723-726.
Saito, H., Motohashi, H., Mukai, M., and Inui, K. (1997). Cloning and characterization of
a pH-sensing regulatory factor that modulates transport activity of the human H+ /peptide
cotransporter, PEPT1. Biochem. Biophys. Res. Commun. 237, 577-582.
Sandoval, I.V., Arredondo, J.J., Alcalde, J., Gonzalez, N.A., Vandekerckhove, J.,
Jimenez, M.A., and Rico, M. (1994). The residues Leu(Ile)475-Ile(Leu, Val, Ala)476,
contained in the extended carboxyl cytoplasmic tail, are critical for targeting of the
resident lysosomal membrane protein LIMP II to lysosomes. J. Biol. Chem. 269,6622-
6631.
Schwartzman, M. and Raz, A. (1982). Purinergic vs peptidergic stimulation of lipolysis
and prostaglandin generation in the perfused rabbit kidney. Biochem. Pharmacol. 31,
2453-2458.
Sen, R.P., Delicado, E.G., Alvarez, A., Brocklebank, A.M., Wiley, J.S., and Miras-
Portugal, M.T. (1998). Flow cytometric studies of nucleoside transport regulation in
single chromaffin cells. FEBS Lett. 422, 368-372.
Sen, R.P., Delicado, E.G., and Miras-Portugal, M.T. (1999). Differential modulation of
nucleoside transport types in neuroblastoma cells by protein kinase activation.
Neuropharmacol. 38, 1009-1015.
Seubert, C.N., Morey, T.E., Martynyuk, A.E., Cucchiara, R.F., and Dennis, D.M. (2000).
Midazolam selectively potentiates the A(2A) - but not A1-receptor — mediated effects of
adenosine: role of nucleoside transport inhibition and clinical implications. Anesthesiol.
92, 567-577.
Sloan, J.L. and Mager, S. (1999). Cloning and functional expression of a human Na+ and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CX-dependent neutral and cationic amino acid transporter B0 + . J. Biol. Chem. 274, 23740-
23745.
Soderback, U., Sollevi, A., Wallen, N.H., Larsson, P.T., and Hjemdahl, P. (1991). Anti
aggregatory effects of physiological concentrations of adenosine in human whole blood
as assessed by filtragometry. Clin. Sci. 81, 691-694.
Soler, C., Felipe, A., Mata, J.F., Casado, F.J., Celada, A., and Pastor-Anglada, M. (1998).
Regulation of nucleoside transport by lipopolysaccharide, phorbol esters, and tumor
necrosis factor-alpha in human B-lymphocytes. J. Biol. Chem. 273, 26939-26945.
Soler, C., Felipe, A., Casado, F.J., Celada, A., and Pastor-Anglada, M. (2000). Nitric
oxide regulates nucleoside transport in activated B lymphocytes. J. Leukoc. Biol. 67,
345-349.
Soler, C., Garcia-Manteiga, J., Valdes, R., Xaus, J., Comalada, M., Casado, F.J., Pastor-
Anglada, M., Celada, A., and Felipe, A. (2001a). Macrophages require different
nucleoside transport systems for proliferation and activation. FASEB J. 15,1979-1988.
Soler, C., Valdes, R., Garcia-Manteiga, I., Xaus, J., Comalada, M., Casado, F.J.,
Modolell, M., Nicholson, B., MacLeod, C., Felipe, A., Celada, A., and Pastor-Anglada,
M. (2001b). Lipopolysaccharide (LPS)-induced apoptosis of macrophages determines the
up-regulation of concentrative nucleoside transporters CNT1 and CNT2 through TNF-
alpha-dependent and -independent mechanisms. J. Biol. Chem. 276, 30043-30049.
Sollevi, A. (1986). Cardiovascular effects of adenosine in man; possible clinical
implications. Prog. Neurobiol. 27, 319-349.
Sonnhammer, E.L., von Heijne, G., and Krogh, A. (1998). A hidden Markov model for
predicting transmembrane helices in protein sequences. Proc. Int. Conf. Intell. Syst. Mol.
Biol. 6 ,175-182.
Stevens, B.R., Tellier, M., Harvey, W., Feldman, D.H., and Bosworth, J. (2000).
Interleukin-2 and concanavalin A upregulate a cat2 isoform encoding a high affinity L-
arginine transporter in feline lymphocytes. Can. J. Vet. Res. 6 4 ,187-191.
Stiles, G.L. (1992). Adenosine receptors. J. Biol. Chem. 267, 6451-6454.
Sundaram, M., Yao, S.Y., Ng, A.M., Griffiths, M., Cass, C.E., Baldwin, S.A., and
Young, J.D. (1998). Chimeric constructs between human and rat equilibrative nucleoside
transporters (hENTl and rENTl) reveal hENTl structural domains interacting with
coronary vasoactive drugs. J. Biol. Chem. 273, 21519-21525.
Sundaram, M., Yao, S.Y., Ingram, J.C., Berry, Z.A., Abidi, F., Cass, C.E., Baldwin, S.A.,
and Young, J.D. (2001a). Topology of a Human Equilibrative, Nitrobenzylthioinosine
(NBMPR)-sensitive Nucleoside Transporter (hENTl) Implicated in the Cellular Uptake
of Adenosine and Anti-cancer Drugs. J. Biol. Chem. 276,45270-45275.
249
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Sundaram, M., Yao, S.Y., Ng, A.M., Cass, C.E., Baldwin, S.A., and Young, J.D. (2001b).
Equilibrative nucleoside transporters: mapping regions of interaction for the substrate
analogue nitrobenzylthioinosine (NBMPR) using rat chimeric proteins. Biochemistry 40,
8146-8151.
Sweet, D.H., Miller, D.S., and Pritchard, J.B. (2000). Basolateral localization of organic
cation transporter 2 in intact renal proximal tubules. Am. J. Physiol. Renal Physiol. 279,
F826-F834.
Tai, Y.H., Flick, J., Levine, S.A., Madara, J.L., Sharp, G.W., and Donowitz, M. (1996).
Regulation of tight junction resistance in T84 monolayers by elevation in intracellular
Ca2 + : a protein kinase C effect. J. Membr. Biol. 149,71-79.
Takata, K. (1996). Glucose transporters in the transepithelial transport of glucose. J.
Electron Microsc. 45, 275-284.
Tamai, I, Yabuuchi, EL, Nezu, J., Sai, Y., Oku, A., Shimane, M., and Tsuji, A. (1997).
Cloning and characterization of a novel human pH-dependent organic cation transporter,
OCTN1. FEBS Lett. 419,107-111.
Terada, T., Sawada, K., Saito, H., Hashimoto, Y., and Inui, K. (1999). Functional
characteristics of basolateral peptide transporter in the human intestinal cell line Caco-2.
Am. J. Physiol. 276, G1435-G1441.
Thompson, J.D., Higgins, D.G., and Gibson, T.J. (1994). CLUSTAL W: improving the
sensitivity of progressive multiple sequence alignment through sequence weighting,
position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673-
4680.
Thomson, A.B. and Wild, G. (1997). Adaptation of intestinal nutrient transport in health
and disease. Part I. Dig. Dis. Sci. 42,453-469.
Tsuji, A. and Tamai, I. (1996). Carrier-mediated intestinal transport of drugs. Pharm. Res.
13, 963-977.
Turk, E., Kemer, C.J., Lostao, M.P., and Wright, E.M. (1996). Membrane topology of the
human Na+ /glucose cotransporter SGLT1. J. Biol. Chem. 271, 1925-1934.
Uauy, R., Stringel, G., Thomas, R., and Quan, R. (1990). Effect of dietary nucleosides on
growth and maturation of the developing gut in the rat. J. Pediatr. Gastroenterol. Nutr. 10,
497-503.
Uauy, R., Quan, R., and Gil, A. (1994). Role of nucleotides in intestinal development and
repair: implications for infant nutrition. J. Nutr. 124,1436S-1441S.
Valdes, R., Ortega, M.A., Casado, F.J., Felipe, A., Gil, A., Sanchez-Pozo, A., and Pastor-
Anglada, M. (2000). Nutritional regulation of nucleoside transporter expression in rat
250
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
small intestine. Gastroenterol. 119, 1623-1630.
Varoqui, H. and Erickson, J.D. (2002). Selective up-regulation of system A transporter
mRNA in diabetic liver. Biochem. Biophys. Res. Commun. 290, 903-908.
Vasudevan, G., Carter, N.S., Drew, M.E., Beverley, S.M., Sanchez, M.A., Seyfang, A.,
Ullman, B., and Landfear, S.M. (1998). Cloning of Leishmania nucleoside transporter
genes by rescue of a transport-deficient mutant. Proc. Natl. Acad. Sci. U. S. A. 95, 9873-
9878.
Vickers, M.F., Mani, R.S., Sundaram, M., Hogue, D.L., Young, J.D., Baldwin, S.A., and
Cass, C.E. (1999). Functional production and reconstitution of the human equilibrative
nucleoside transporter (hENTl) in Saccharomyces cerevisiae. Interaction of inhibitors of
nucleoside transport with recombinant hENTl and a glycosylation-defective derivative
(hENTl/N48Q). Biochem. J. 339, 21-32.
Vickers, M.F., Yao, S.Y., Baldwin, S.A., Young, J.D., and Cass, C.E. (2000). Nucleoside
transporter proteins of Saccharomyces cerevisiae. Demonstration of a transporter (FU11)
with high uridine selectivity in plasma membranes and a transporter (FUN26) with broad
nucleoside selectivity in intracellular membranes. J. Biol. Chem. 275, 25931-25938.
Von Lubitz, D.K., Carter, M.F., Beenhakker, M., Lin, R.C., and Jacobson, K.A. (1995).
Adenosine: a prototherapeutic concept in neurodegeneration. Ann. N. Y. Acad. Sci. 765,
163-178.
Walker, D., Thwaites, D.T., Simmons, N.L., Gilbert, H.J., and Hirst, B.H. (1998).
Substrate upregulation of the human small intestinal peptide transporter, hPepTl. J.
Physiol. 507, 697-706.
Wallman-Johansson, A. and Fredholm, B.B. (1994). Release of adenosine and other
purines from hippocampal slices stimulated electrically or by hypoxia/hypoglycemia.
Effect of chlormethiazole. Life Sci. 55, 721-728.
Wang, W., Takezawa, D., Narasimhulu, S.B., Reddy, A.S., and Poovaiah, B.W. (1996).
A novel kinesin-like protein with a calmodulin-binding domain. Plant Mol. Biol. 31, 87-
100.
Wang, J., Su, S.F., Dresser, M.J., Schaner, M.E., Washington, C.B., and Giacomini, K.M.
(1997). Na+ -dependent purine nucleoside transporter from human kidney: cloning and
functional characterization. Am. J. Physiol. 273, F1058-F1065.
Ward, J.L. and Tse, C.M. (1999). Nucleoside transport in human colonic epithelial cell
lines: evidence for two Na+ -independent transport systems in T84 and Caco-2 cells.
Biochim. Biophys. Acta 1419,15-22.
Ward, J.L., Sherali, A., Mo, Z.P., and Tse, C.M. (2000). Kinetic and pharmacological
properties of cloned human equilibrative nucleoside transporters, ENT1 and ENT2, stably
expressed in nucleoside transporter-deficient PK15 cells. ENT2 exhibits a low affinity
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
for guanosine and cytidine but a high affinity for inosine. J. Biol. Chem. 275, 8375-8381.
Wasa, M., Bode, B.P., and Souba, W.W. (1996). Adaptive regulation of amino acid
transport in nutrient-deprived human hepatomas. Am. J. Surg. 171,163-169.
Watkins, L.F. and Lewis, R.A. (1987). The metabolism of deoxyguanosine in
mitochondria. Characterization of the uptake process. Mol. Cell Biochem. 77, 71-77.
Wenzel, U., Diehl, D., Herget, M., Kuntz, S., and Daniel, H. (1999). Regulation of the
high-affinity HVpeptide cotransporter in renal LLC-PK1 cells. J. Cell Physiol. 178, 341-
348.
Westh Hansen, S.E., Jensen, N., and Munch-Petersen, A. (1987). Studies on the sequence
and structure of the Escherichia coli K-12 nupG gene, encoding a nucleoside-transport
system. Eur. J. Biochem. 168, 385-391.
Wickner, R.B., Edskes, H.K., Maddelein, M.L., Taylor, K.L., and Moriyama, H. (1999).
Prions of yeast and fungi. Proteins as genetic material. J. Biol. Chem. 274, 555-558.
Williams, T.C. and Jarvis, S.M. (1991). Multiple sodium-dependent nucleoside transport
systems in bovine renal brush-border membrane vesicles. Biochem. J. 274, 27-33.
Witte, D.P., Wiginton, D.A., Hutton, J.J., and Aronow, B.J. (1991). Coordinate
developmental regulation of purine catabolic enzyme expression in gastrointestinal and
postimplantation reproductive tracts. J. Cell Biol. 115, 179-190.
Yao, S.Y., Ng, A.M., Ritzel, M.W., Gati, W.P., Cass, C.E., and Young, J.D. (1996).
Transport of adenosine by recombinant purine- and pyrimidine-selective
sodium/nucleoside cotransporters from rat jejunum expressed in Xenopus laevis oocytes.
Mol. Pharmacol. 5 0 ,1529-1535.
Yao, S.Y., Ng, A.M., Muzyka, W.R., Griffiths, M., Cass, C.E., Baldwin, S.A., and
Young, J.D. (1997). Molecular cloning and functional characterization of
nitrobenzylthioinosine (NBMPR)-sensitive (es) and NBMPR-insensitive (ei) equilibrative
nucleoside transporter proteins (rENTl and rENT2) from rat tissues. J. Biol. Chem. 272,
28423-28430.
Yao, S.Y., Ng, A.M., Sundaram, M., Cass, C.E., Baldwin, S.A., and Young, J.D. (2001a).
Transport of antiviral 3'-deoxy-nucleoside drugs by recombinant human and rat
equilibrative, nitrobenzylthioinosine (NBMPR)-insensitive (ENT2) nucleoside
transporter proteins produced in Xenopus oocytes. Mol. Membr. Biol. 18, 161-167.
Yao, S.Y., Sundaram, M., Chomey, E.G., Cass, C.E., Baldwin, S.A., and Young, J.D.
(2001b). Identification of Cysl40 in helix 4 as an exofacial cysteine residue within the
substrate-translocation channel of rat equilibrative nitrobenzylthioinosine (NBMPR)-
insensitive nucleoside transporter rENT2. Biochem. J. 353, 387-393.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Zhen, Y.S., Taniki, T., and Weber, G. (1992). Azidothymidine and dipyridamole as
biochemical response modifiers: synergism with methotrexate and 5-fluorouracil in
human colon and pancreatic carcinoma cells. Oncol. Res. 4 ,73-78.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
An air-interface rabbit tracheal epithelial cell culture model for the characterization of nucleoside and nucleoside drug transport
PDF
Drug efflux pumps in rabbit conjunctival epithelial cells
PDF
Drug transport mechanisms in the rabbit corneal epithelium: In vitro-in vivo correlation
PDF
Characterization of lacrimal gland acinar cell cytoskeletal structure and its participation in stimulated secretion
PDF
Biochemical, molecular biological and computer modeling studies of the conjunctival dipeptide transporter isoforms
PDF
Characterization of microtubule motors in the apical trafficking of hydrogen,potassium-ATPase in oxyntic cells
PDF
Homeostatic regulation of intracellular signaling networks by Etk /Bmx
PDF
Cytochrome P450 3A metabolism in the rabbit lacrimal gland and conjunctiva
PDF
Etk/Bmx activation modulates barrier function in epithelial cells
PDF
EPR analysis of soluble and membrane -bound annexin A2
PDF
Alteration of the in vitro and in vivo processing of a polypeptide, BBI, through conjugation with palmitic acid
PDF
Insulin-transferrin conjugate as an oral hypoglycemic agent
PDF
Drug delivery to the posterior region of the eye: Characterization of organic cation transporters
PDF
Conjunctival epithelial uptake of biodegradable nanoparticles: Mechanism, intracellular distribution, and absorption enhancement
PDF
A dynamic apical actin cytoskeleton facilitates exocytosis of tear proteins in rabbit lacrimal acinar epithelial cells
PDF
Characterization of dynamin II and clathrin adaptor complex in the regulation of gastric hydrogen,potassium-ATPase recycling in oxyntic cells
PDF
Computational prediction of peptide-MHC class I binding interactions
PDF
Cytokine regulation of retinal pigment epithelial cell function and behavior
PDF
How cells cry: The cytoskeleton as a facilitator of regulated secretion and membrane trafficking in lacrimal gland acinar cells
PDF
An in vivo study of G protein coupled receptor mediated signaling
Asset Metadata
Creator
Wu, Sharon Kuang-Hui
(author)
Core Title
Basolateral transport of nucleosides in rabbit tracheal epithelial cells: An equilibrative nucleoside transporter outcome
School
Graduate School
Degree
Doctor of Philosophy
Degree Program
Pharmaceutical Sciences
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
biology, molecular,biophysics, general,Health Sciences, Pharmacy,OAI-PMH Harvest
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Lee, Vincent H.L. (
committee chair
), Ann, David (
committee member
), Haworth, Ian S. (
committee member
), Kim, Kwang-Jin (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-553536
Unique identifier
UC11339808
Identifier
3094418.pdf (filename),usctheses-c16-553536 (legacy record id)
Legacy Identifier
3094418.pdf
Dmrecord
553536
Document Type
Dissertation
Rights
Wu, Sharon Kuang-Hui
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
biology, molecular
biophysics, general
Health Sciences, Pharmacy