Localization of human equilibrative nucleoside transporters,hENT1 and hENT2, in - 《American Journal of Physiology》美国生理学杂志干细胞之家



免疫细胞治疗专区	欢迎关注干细胞微信公众号

12 3 4 5 6 7 8 9 10 ... 26 下一页

返回列表

查看: 488111\|回复: 251	go Localization of human equilibrative nucleoside transporters,hENT1 and hENT2, in [复制链接]

轻羽

新手上路

Rank: 1

积分: 0
威望: 0
包包: 0

楼主

发表于 2009-4-21 13:36 |只看该作者 |倒序浏览 |打印

作者：Lara M.Mangravite, GuangqingXiao,  KathleenM.Giacomini作者单位：Department of Biopharmaceutical Sciences, University ofCalifornia, San Francisco, San Francisco, California 94143-0446
   - m2 k  }3 M+ s5 x* q( P- z3 c! J0 P
   0 n* i9 ^: e# P5 _; n1 A, @2 q
      . H. t9 V4 v' @. Q! D
                     9 K! m# x0 ^8 u: _" M
         % p1 ~2 d0 e* \& a. f

         ( f. d% q' u- J# y, j
         ; X) P2 K8 r# [* _, M



      " V7 k" ~9 [) d3 m+ w0 A' M
      【摘要】
   Nucleoside transporters areimportant in the disposition of nucleosides and nucleoside analogs inthe kidney. Two human equilibrative nucleoside transporters have beencloned and characterized, hENT1 and hENT2. The primary goal of thisstudy was to localize these transporters in polarized renal epithelia.hENT1 and hENT2 were tagged with green fluorescence protein, stablyexpressed in renal epithelial cells, and localized byimmunofluorescence and functional analysis. Our data demonstrated thatboth transporters are expressed on the basolateral membrane. hENT1 isalso present on the apical membrane. Additionally, we examined theimportance to basolateral targeting of two COOH-terminal targetingmotifs: a RXXV motif for hENT1 and a dileucine repeat for hENT2.Neither motif appeared to affect targeting, but the dileucine repeatwas implicated in surface expression of hENT2. In addition, a splicevariant of hENT2 was identified that is predicted to result in a156-residue COOH-terminal truncation. This variant had a tissuedistribution similar to wild-type hENT2 but was retainedintracellularly. These data suggest that hENT1 and hENT2 on thebasolateral membrane function with concentrative nucleosidetransporters on the apical membrane to mediate active reabsorption ofnucleosides within the kidney. 5 T- N. l5 M# u+ V7 J7 r8 K! a
      【关键词】 equilibrative nucleoside transporter MadinDarby canine kidney dileucine splice variant; y1 `9 ?1 e/ A5 N: u
   INTRODUCTION

NUCLEOSIDE TRANSPORTERS ARE polytopic membrane proteins that mediate boththe uptake and release of hydrophilic nucleosides across lipophilicmembranes. Nucleoside transporters are essential for cellular uptake ofmany clinically relevant nucleoside analogs used in the treatment ofcancers and viral infections. Additionally, they are highly abundant inthe kidney where they are hypothesized to play a major role in thesalvage of endogenous nucleosides used for nucleotide synthesis.
# Z, A0 n% E+ b
Two major classes of nucleoside transporters, equilibrative nucleosidetransporters (ENT, SLC29) and concentrative nucleoside transporters(CNT, SLC28), have been characterized from a variety of species,including humans and rats ( 12, 13, 33, 34, 40, 47, 48 ).The CNT family are secondary active transporters that couple cellulartransport of nucleosides to an internally directed sodium or protongradient ( 10, 33, 47 ). In contrast, the ENT familymediates passive transport of nucleosides. Classically, the ENT familycan be further subdivided into two types of transporters ( es and ei ) based on their sensitivity to inhibition bynitrobenzylthioinosine (NBMPR); Es -type transport issensitive to NBMPR, whereas ei -type transport is not( 2, 48 ). Two members of the ENT family have been clonedand functionally characterized: ENT1, which mediates es -typetransport, and ENT2, which mediates ei -type transport ( 15, 19, 48 ).$ u% {1 t- O! v% @1 ~( s4 c
! Q+ x9 Y* ?" y# |. f0 @
Members of both the CNT and ENT family are present in renal epitheliumthat forms the barrier between the tubule lumen and the circulatorysystem ( 11, 24, 25, 45, 46 ). These transporters arehypothesized to act in series to mediate the vectorial flux ofnucleosides through this epithelium in a reabsorptive direction, providing a means to salvage nucleosides from the filtrate. The abilityof the epithelial cells to perform this function depends on theasymmetric cellular distribution of nucleoside transporters to theapical and basolateral membrane. Early studies using apical andbasolateral membrane vesicles from renal epithelium in animal modelsindicate that transport at the apical membrane is predominantly concentrative while basolateral transport is predominantlyequilibrative ( 3, 23, 25, 31, 37, 39, 46 ). Some studiesadditionally report equilibrative nucleoside transport activity onthe apical membrane ( 6, 9 ). Molecular localization studiesin our laboratory provided the first direct evidence that CNT1 and CNT2are localized to the apical membrane in cultured renal cells( 27 ). This result is supported by recentimmunohistochemical studies demonstrating apical expression of CNT1 inepithelia using rat kidney tissue ( 14 ). Recent work by Laiet al. ( 22 ) localized ENT1 predominantly to thebasolateral membrane of differentiated renal epithelial cells.Immunofluorescence studies visualized ENT1 entirely on the basolateralmembrane, but functional assays indicated low levels of ENT1-mediatedtransport on the apical membrane as well ( 22 ). To date,there is no information regarding intracellular localization of ENT2 inrenal epithelial cells. Knowledge of the localization of thesetransporters will enhance our understanding of how ENT1 and ENT2 workin concert with the CNT family to mediate transepithelial flux ofnucleosides and nucleoside analogs within the kidney. Furthermore, thisinformation will contribute to understanding the differential functionsof these two transporters.2 j9 |) w, a- X5 e9 I7 ~

In polarized cells such as renal epithelium, plasma membrane proteinsare sorted in the trans -Golgi network and specifically sentto either the apical or basolateral membrane ( 1 ).Basolateral targeting appears to be triggered by distinct amino acidsequences (targeting motifs) within the protein itself, which interactwith the cellular sorting machinery ( 50 ). Some of thesetargeting motifs [such as the tyrosine motif (NPXY) or dileucinerepeat] are related to signals for clathrin-coated pit localization.These signals overlap with those used for endosomal recycling andendocytosis ( 1 ). Some proteins contain basolateraltargeting motifs unrelated to clathrin-coated pits such as the R/HXXVmotif seen in the cation-dependent mannose 6-phosphate receptor(CD-MPR; see Ref. 8 ). Although the exact mechanisms ofaction of these unrelated motifs are unknown, their structuralorientation appears to allow for interaction with the basolateralsorting machinery. Apical targeting is less well understood but appearsto be based on segregation of apical proteins into vesicles or raftsenriched with lipids preferentially delivered to the apical membrane.For some proteins, it appears that incorporation into these rafts isbased on glycosylation motifs or glycosylphosphatidylinositol anchors( 1 )." \& \' X- U; C3 t4 Q0 a0 z' ~9 `

The goal of this study was to determine the localization of both human(h) ENT1 and hENT2 within renal epithelial cells. We used Madin-Darbycanine kidney (MDCK) cells, which have been successfully used to studyin vivo intracellular localization of a variety of renal transporters( 4, 28, 29, 38 ). Additionally, we sought to examine theimportance of two targeting motifs in basolateral targeting anddistribution of hENT1 and hENT2.3 h/ q. N  D2 ^) _

MATERIALS AND METHODS- f/ n' w% R. i9 Y
+ r: w2 W9 V# N8 e# X* C
Materials. Cell culture media and supplements were purchased from the Universityof California, San Francisco Cell Culture Facility (San Francisco, CA).pEGFP-C1 was purchased from Clontech (Palo Alto, CA), and vector pOXwas a gift from Andrew T. Gray (University of California, SanFrancisco). The EMBL MDCK II strain was a gift from Dr. Karl Matlin.Texas red-conjugated phalloidin was purchased from Molecular Probes(Eugene, OR). Transwell polycarbonate cell culture filters andpolycarbonate cell culture plates were purchased from Corning Costar(Corning, NY). Bradford reagent was purchased from Bio-Rad (Hercules,CA). Radiolabeled adenosine and thymidine were from MoravekBiochemicals (Brea, CA). All other chemicals were supplied by Sigma(St. Louis, MO)./ @" I% v2 A7 I' I; J- J* W# l

Plasmid construction. hENT1, hENT2, and the splice variant hENT2A were cloned by PCR usingprimers flanking the open reading frame (ORF) of hENT1 and hENT2. Theprimers were designed based on published hENT1 and hENT2 cDNA sequences( 7, 12 ). For hENT1, the sense primer was5'-gggaaaaccgagaacaccatcaccatg-3'; the antisense primer was 5'-agtccttctgtccatcctttgtcacac-3'. For hENT2, the sense primer was5'-ggcgcatccgccgcggcggccatggcg-3', and the antisense primer was5'-gagcctggaggggccacttcagagcag-3'. hENT1 was then subcloned in frameinto pEGFP-C1 vector by adding a Sal I site to the 5' endand a Sac II site to the 3' end. hENT2 and hENT2A weresubcloned in frame into pEGFP-C1 vector by adding a Sal Isite to the 5' end and an Apa I site to the 3' end. Allplasmid constructions and DNA sequences were confirmed by enzymedigestion analyses and by automated sequencing at the BiomolecularResource Center at the University of California, San Francisco.

Site-directed mutagenesis. Mutations of hENT1 (R453A and RAIV) and of hENT2 (L455R and LL)were constructed with the QuickChange Site-directed Mutagenesis Kit(Stratagene, La Jolla, CA) using wild-type hENT1 cDNA and hENT2 cDNA asthe templates. The sequences of these mutants were confirmed by DNAsequencing at the Biomolecular Resource Center at the University ofCalifornia, San Francisco.
7 Y# I4 a0 l# `
Stable transfection of MDCK. MDCK cells were grown in MEM with Earle's BSS supplement, 5%heat-inactivated FBS, 100 U/ml penicillin, and 100 U/ml streptomycin ina humidified atmosphere of 5% CO 2 -95% air at 37°C.Cells were transfected with 1 µg DNA and 16 µg Effectene (Qiagen,Valencia, CA). Cells were grown for 48 h and then diluted in mediasupplemented with 700 µg/ml G418. Clones were picked after 2 wk ofgrowth in selection media, and positive clones were chosen by Westernblot, confocal microscopy, and functional uptake of 3 H-labeled nucleoside.; z  X* W  e' b& V  W5 e, R

Confocal microscopy. Samples were prepared for confocal microscopy as described previously( 27 ). Samples were grown for 4-7 days on permeable support and then fixed with 4-8% paraformaldehyde, permeabilized with 0.025% (wt/vol) saponin in PBS, stained with Texas red-conjugated phalloidin for visualization of actin, and mounted on slides in Vectashield mounting medium (Vector Laboratories, Burlingame, CA).Samples were analyzed using a Bio-Rad MRC-1024 laser scanning confocal microscope.
# A( U8 G$ y7 t9 K) B
Functional uptake in MDCK. Stably transfected MDCK were grown for 5-7 days on permeablesupport and then assayed for membrane specific functionality, asdescribed previously ( 27 ). Briefly, cells were treatedwith 0.1 µM [ 3 H]inosine in choline buffer (in mM: 128 choline, 4.73 KCl, 1.25 CaCl 2, 1.25 MgSO 4, and5 HEPES, pH 7.4) in the absence or presence of 1 mM inosine. Allbuffers in hENT2 experiments also contained 10 µM NBMPR to reducebackground levels of endogenous es -type function. Reactionmix was applied to either the apical or basolateral membrane for 2 minand removed, and cells were washed three times in ice-cold cholinebuffer to terminate the reaction. Cellular uptake of[ 3 H]inosine was measured by lysing cells and counting ina Beckman Scintillation Counter. All experiments were repeated induplicate on three separate occasions.# q6 H2 b3 S- ]3 X

Expression and transport assay in Xenopus laevis oocytes. To study the function of wild-type and mutant hENTs and greenfluorescence protein (GFP)-tagged hENTs, DNA of these transporters wassubcloned in pOX vector by adding a Sal I site to the 5' end and an Xbal I site to the 3' end. pOX contains the 5' and 3'untranslated regions of the Xenopus -globin gene flankingthe insert ( 17 ). hENT- and GFP-tagged hENT cRNA wassynthesized using T3 polymerase (Stratagene) following themanufacturer's protocol. Oocytes were harvested and treated asdescribed previously ( 47 ). Fifty nanoliters of cRNA(~0.4 ng/nl) or water was injected individually in defolliculated oocytes. Oocytes were incubated at 18°C for 30-40 h, and then uptake assays were performed for 40 min at 25°C in 100 µl oftransport buffer (2 mM KCl, 1 mM CaCl 2, and 10 mM HEPES)containing various concentrations of 3 H-labeled nucleosides(Moravek Biochemicals). The reaction was terminated by washing oocytesfive times in 3 ml ice-cold choline buffer. Oocytes were lysedindividually in 10% SDS, and the amount of radiolabeled nucleosidetransported in each oocyte was determined by liquid scintillation counting.

Statistics and data analysis. Groups of 8-10 cRNA-injected or water-injected oocytes were usedfor each experiment. Uptake values are expressed as means ± SE.For kinetic studies, uptake rates ( V ) determined atdifferent substrate concentrations (S) were fit to the Michaelis-Menten equation: V = V max × S/( K m S), where V max is the maximal uptake rate, and K m is the Michaelis-Menten constant (thesubstrate concentration at V max /2). Fits werecarried out using a nonlinear least-squares regression-fitting program(Kaleidagraph, V.3.0; Abelbeck/Synergy Software, Reading, PA). Kineticexperiments were repeated several times in different batches ofoocytes; data for one representative experiment are presented in thisstudy. Statistical analysis was carried out by comparing the uptakesfrom tested compounds with those from controls in the same experimentsusing a two-tailed, two-sample equal-variance t -test.Results with the probability of P significant.

RESULTS* J7 l! q' Y+ e4 T0 A
2 d+ _( e/ |, ^8 M; a+ N# f7 \
Localization of hENT1 and hENT2 in polarized renal epithelialcells. To visualize hENT1 and hENT2 in the absence of protein-specificantibodies, we tagged the NH 2 terminus of hENT1 and hENT2 with GFP. Kinetic studies in Xenopus laevis oocytesindicated that there were no significant differences in the uptake ofadenosine or thymidine between tagged and untagged transporters (Table 1 ), suggesting that the GFP tag does notkinetically alter the function of these transporters.1 W7 x( l7 ?, E: G

Table 1. Apparent K m and V max of adenosine: R. K9 F' p5 S+ X
9 A6 T% U# r. x8 V  X" R$ Y. |
Immunofluorescent analysis of tagged hENT1 and hENT2 stably transfectedin MDCK indicated that both transporters localized predominantly to thebasolateral membrane (Fig. 1 ). Verticaloptical sections indicated the additional presence of a small portion of hENT1 on the apical membrane (Fig. 1 A, i-ii ).Apical localization was evident upon closer examination of the apicalsurface using xy -slices (Fig. 1 A, iii-v ). The apical presence of hENT2 was not observed (Fig. 1 B ). hENT1-mediated transport of inosine was observed atboth the apical and basolateral membranes, whereas hENT2-mediated transport was isolated to the basolateral membrane (Fig. 2 ). Results were replicated usingmultiple positive stable clones for each transporter./ w! d6 X: r% Q

Fig. 1. Immunofluorescent localization of wild-type equilibrativenucleoside transporter (ENT) 1 and ENT2 in Madin-Darby caninekidney (MDCK) cells. MDCK stably transfected with ENT1-greenfluorescence protein (GFP; A ) or ENT2-GFP ( B )were fixed, permeabilized, stained with Texas red-conjugatedphalloidin, and visualized by confocal fluorescence microscopy.Vertical optical sections ( i and ii ) with apicalmembrane on top (bar = 10 µm). Slices through the xy -plane are shown in iii - v. Distancesfrom plastic support (below basolateral membrane) are indicated inµm. Images iii and iv show apical membrane. Images i and iii are shown with GFP-tagged protein in greenand phalloidin-stained F-actin in red. All other images display onlythe GFP-tagged protein. Bar = 10 µm.
% S" C4 r: T, s
Fig. 2. Functional localization of GFP-tagged wild-type human (h)ENT1 and hENT2 in MDCK. hENT1-GFP ( A ), hENT2-GFP( B ), or GFP ( A and B ) stablytransfected cells were polarized by growth on permeabilized filters for5-7 days. Uptake of [ 3 H]inosine was measured for 2 min from either the apical (Ap) or basolateral (Bl) membrane in eitherthe absence (filled bars) or presence (open bars) of inosine (1 mM). Nosodium was present in any of the experiments. Nitrobenzylthioinosine(10 µM) was present in solutions used for functional analysis ofhENT2. Each experiment was repeated in duplicate on 3 or 4 separateoccasions.

Basolateral targeting is independent of the COOH-terminal tail ofboth hENT1 and hENT2. We were interested in investigating the molecular determinantsresponsible for polarized localization of hENT1 and hENT2. Based onhydropathy plot analysis and topology studies, hENT1 and hENT2 are eachpredicted to have eleven transmembrane domains with a four-amino-acidCOOH-terminal tail (Fig. 3 A;see Refs. 12-14 ). The COOH terminus of bothtransporters contained a motif implicated in basolateral targeting: anR/HXXV motif in hENT1 (RAIV) and a dileucine repeat in hENT2 (LL). Noother targeting motifs were obvious in either sequence at eitherterminus. We investigated the significance of these two sequences onpolarized trafficking of the proteins via mutagenesis studies.' X9 U4 N3 s  Z- g4 A, B/ V

Fig. 3. Expression and localization of mutagenized ENT1 and ENT2in MDCK cells. A : sequence analysis of COOH termini ofhENT1 and hENT2. Predicted transmembrane domains are underlined,targeting motifs are in bold, and single amino acids that were mutatedare in large bold letters. B : immunofluorescence ofpolarized MDCK stably transfected with ENT1 R453A ( i and ii ), ENT1 RAIV ( iii and iv ), ENT2L455R ( v and vi ), and ENT2 LL ( vii and viii ). Vertical optical images shown with apicalmembrane on top. Bar = 10 µm. i, iii, v,and vii show GFP-tagged protein as green andphalloidin-stained F-actin as red. ii, iv, vi, and viii show only the GFP-tagged protein from the image on left. C : immunofluorescent Z -series ofENT2 L455R; xy -sections spaced 3 µm apart are shown inseries. Position of each image relative to the plastic support belowthe basolateral membrane is indicated in each image in µm. Bar = 10 µm. Red indicates phalloidin-stained F-actin, and green indicatesGFP-tagged protein. Arrows indicate vesicular staining. D :immunofluorescent Z -series of ENT2 LL.' e, K, u# q+ \! ^+ M# P0 E- ?# [/ t, _
3 F6 @" Z8 v/ O0 O
The R/HXXV sequence in hENT1 was mutagenized in the following two ways: 1 ) a single mutation was made at position 453, removing thearginine (R453A), or 2 ) the entire COOH-terminal tail was truncated ( RAIV). Both mutants were stably transfected in MDCK andproduced full-length protein, as demonstrated by Western blot (data notshown). Neither mutation had a visible effect on localization of hENT1(Fig. 3 B ), nor was there an effect on hENT1-mediated transport of inosine at the basolateral membrane (data not shown).! v4 e# K* a: W# t( ~/ B

The dileucine repeat in hENT2 was also mutated by both pointmutation (L455R) and truncation ( LL; Fig. 3 A ) and stablytransfected in MDCK. Transfection efficiencies were extremely low( cells were transfected). L455R protein levels were too lowto be detected by Western blot. LL protein levels were detectable and yielded a single band smaller in size than wild-type hENT2 butequivalent to wild-type hENT1 (data not shown). Wild-type ENT2 clonesproduced a protein of abnormally large mass (~140 kDa) compared withENT1 (80 kDa). Evidence for large ENT2 has been seen in other studiesas well ( 18, 42 ).
3 ?( S* n. S# ~) K- r3 }" n. b
Both hENT2 L455R and LL trafficked exclusively to the basolateralmembrane with no apparent apical localization. However, surfaceexpression was reduced drastically (Fig. 3 B ). The L455R mutant displayed some vesicular staining in MDCK (Fig. 3 C ).The LL mutant displayed significant vesicular staining (Fig. 3 D ), indicating that the dileucine motif is important forsurface expression. This was also true when these proteins weretransfected in LLC-PK 1 cells, a renal epithelial cell lineoriginating from the proximal tubule (data not shown). Vesicularretention may occur by alteration of protein stability, surfacedelivery, or surface retention.$ ]1 p6 z( h; I: I. o3 j7 U
$ T+ @  J* J& E+ p* B
Functional localization studies could not be performed on the mutatedhENT2 stable MDCK clones because of low transfection efficiencies.Therefore, further studies analyzing the effect of these mutations onhENT2 function were carried out using heterologous expression inoocytes (Fig. 4 ). Neither truncation ofhENT1 ( RAIV) nor single mutation of hENT2 (L455R) altered thefunctional activity of these proteins. In contrast, the hENT2 truncatedmutant ( LL) showed significant reduction in function, suggestingthat the dileucine motif is essential for surface expression of hENT2in oocytes as well.

Fig. 4. Effect of mutations on hENT1- and hENT2-mediated uptakeof adenosine and thymidine in oocytes. Uptake of[ 3 H]adenosine (10 µM; A ) and[ 3 H]thymidine (10 µM; B ) was measured at25°C for 40 min in oocytes injected with H 2 O or cRNA forGFP-tagged hENT1 [wild-type (wt)], hENT1 RAIV ( R), or hENT2 LL ( ). Data are expressed asmeans ± SE of 6-8 oocytes. * Results that arestatistically different from GFP-tagged wild-type control( P9 W! O" ]6 c; i/ U% w& J+ e  `& J
: I) s0 u* G0 {
Identification of an hENT2 variant. In the process of cloning hENT2, using primers flanking the ORF ofthe published hENT2 cDNA sequence ( 7 ), we found avariant termed hENT2A (GenBank accession no. AF401235 ). We determined this to be a splice variant based on the genomic sequence of hENT2 (GenBank accession no. AF034102 ), which has 12 exons and 11 introns.hENT2A uses a different splicing site on the 5' end of exon 9, causinga 40-bp deletion (positions 1103-1142) in hENT2A mRNA (Fig. 5 ). This out-of-frame deletion introducesa premature stop codon in the ORF, encoding a truncated variant that is156 amino acids shorter than wild-type hENT2 and has an alternative COOH-terminal sequence (Fig. 6 ). RT-PCRanalysis of several tissues found that both wild-type and variant hENT2are expressed in skeletal muscle, liver, lung, brain, kidney, heart,pancreas, and placenta (data not shown)., @4 m# C8 u# g2 S: [3 x6 h4 e
0 |- b3 M6 V7 v- ^+ L
Fig. 5. Sequence analysis of hENT2 and hENT2A. A :schematic representation of hENT2 and hENT2A mRNA and geneorganization. Lines and boxes in hENT2 gene represent introns andexons, respectively. Exons are numbered. Gray boxes represent exon 9. Black boxes represent the 40-bp region in exon9 that is deleted in hENT2A mRNA. B : 5' and 3'exon-intronic splice sites of exon 9 in hENT2 and hENT2A.Exon region is in bold and italic, and the 40-bp region that is deletedin hENT2A mRNA is underlined.  I8 r5 _# `1 `. n! B% T  a2 o

Fig. 6. Protein sequences of hENT2 and hENT2A. Wild-type hENT2has 456 amino acid residues. Because of the premature stop codoninduced by the out-of-frame 40-bp deletion in the open reading frame ofhENT2, hENT2A has only 301 amino acid residues, and its COOH-terminalsequence is changed. Predicted transmembrane domains are overscored.Amino acid residues in hENT2A that differ from hENT2 are in boldtype.
7 L8 m; j9 A" d5 O
hENT2A was heterologously expressed in oocytes and functionallystudied. The variant did not take up adenosine or thymidine under ourexperimental conditions (Fig. 7 A ). This lack of activity maysuggest that the variant does not retain the domains necessary fornucleoside transport. Alternatively, it is possible that the variant isnot properly trafficking to the membrane. To further explore this,hENT2A was tagged with GFP and stably expressed in MDCK in the samemanner described earlier. hENT2A did not sort to the plasma membrane(Fig. 7, B and C ).- Z  [3 j% |+ U9 W$ L
- r) n, z6 |! o5 [4 z& ]
Fig. 7. Functional analysis and localization of hENT2A. A :uptake of 10 µM [ 3 H]adenosine was measured at 25°Cfor 40 min in oocytes injected with H 2 O or with 20 ng cRNA for hENT2 (wt), hENT2A [variant (v)], or both hENT2 andhENT2A (20 ng each; wt v). Data are expressed as means ± SE of 6-8 oocytes. * Results that are statistically differentfrom wild-type control ( P B :vertical optical image of hENT2A shown with apical membrane on top (bar = 10 µm). i : Merged image showingGFP-tagged hENT2A in green and phalloidin-stained F-actin in red. ii : Only GFP-tagged hENT2A is shown. C :immunofluorescence of hENT2A showing xy -sections in seriesstarting from the apical membrane and moving toward the basolateralmembrane. Positions of image relative to the plastic support below thebasolateral membrane are indicated in µm. Arrows indicate vesicularstaining. Bar = 10 µm.

Several splice variants of other membrane transporters have been foundto have dominant negative effects on the function of wild-typetransporters ( 20, 44 ). We tested the effect of expression of hENT2A on the function of wild-type hENT2 by coinjecting equal amounts of cRNA for both transporters in oocytes and measuring nucleoside uptake. Uptake in oocytes expressing hENT2 and hENT2A didnot differ from that observed in oocytes expressing wild-type hENT2alone (Fig. 7 A ).. N5 _1 j& x% r- j  Y* g

DISCUSSION" `# s. I- F, K% b, w
% c, @+ _1 J) j$ Z. P6 g
Past studies have attempted to localize equilibrativenucleoside transport within epithelia with conflicting results. Both the absence and presence of es -type transport activity(presumably ENT1) in brush-border membrane vesicles has been reported( 5, 6, 9, 26 ). In contrast, ei -type activity(presumably ENT2) was reported to reside only in basolateral membranevesicles ( 5 ). Because functional studies in isolatedplasma membrane vesicles may be confounded by the presence of multipletransport activities or contamination with other membranes, datalocalizing transporters using functional activity are difficult to interpret.$ |3 d( \  z  S, l

In this study, we directly examined the localization of GFP-taggedhENT1 and hENT2 in renal epithelial cells. Our data demonstrate thathENT1 and hENT2 are present and functional on the basolateral membrane(Figs. 1 and 2 ). Interestingly, hENT1 appears in small amounts on theapical membrane where it is also functional. The function of hENT1 onthe apical membrane in transfected MDCK was also demonstrated recentlyby Lai et al. ( 22 ). Previous data from this laboratorydemonstrated that the concentrative nucleoside transporters, CNT1 andCNT2, are predominantly localized to the apical membrane in renalepithelial cells ( 27 ). Together, these data provide apicture of asymmetrically localized CNTs and ENTs working in concert tosalvage nucleosides and nucleoside analogs from the tubular filtrate.In vivo studies showing that adenosine is reabsorbed in the kidneysupport this model ( 9 ).( `8 c. i4 N) K

We were additionally interested in examining the moleculardeterminants responsible for basolateral targeting of these two transporters. The COOH-terminal tail of hENT1 contained an R/HXXV motif(RAIV) that has been implicated in basolateral sorting of CD-MPR( 8, 30, 51 ). Neither mutation nor truncation of thissequence affected hENT1 levels on the basolateral membrane (Fig. 3 ). Incontrast, the COOH-terminal tail of hENT2 contained a dileucine repeat.This motif is implicated as a signal in both basolateral sorting andendosomal recycling of a large number of proteins( 16 ). Both mutation and truncation of the dileucine affected surface expression of hENT2, but, in both cases, all protein that reached the plasma membrane remained confined to thebasolateral membrane. This indicates that this motif is important formaintaining steady-state expression of hENT2 on the plasma membrane.Although this does not implicate the dileucine as a targeting motif,the repeat may be important in endosomal recycling or surfaceretention of hENT2.
" g: A( y6 G* {/ K! ?  x6 a8 }
Differential localization of hENT1 and hENT2 furthersubstantiates the idea that these two transporters are maintained and regulated by distinct mechanisms within the cell. hENT1 is found ubiquitously throughout the body and is thought to be the major transporter involved in uptake of nucleosides for DNA synthesis. hENT1is also implicated in terminating adenosine signals in the vicinity ofadenosine receptors ( 32 ). Within the renal epithelium, theA1 adenosine receptor, which also localizes to both membranes in MDCK,is thought to be the major receptor involved in adenosine signaling( 36 ). Conditions of chronic hypoxia selectivelydownregulate ENT1 function as a means to increase extracellularadenosine levels at the site of its receptor ( 21 ).Symbiosis between hENT1 and the A1 adenosine receptor may explain thepresence of hENT1 on the apical membrane., A( p7 u- Y  w- \$ h2 d. Z0 ]
0 v, I, u4 c4 i6 p  B
In contrast, hENT2 is expressed in far lower amounts in alltissues except skeletal muscle. It has a lower affinity for most physiological nucleosides, with the exception of inosine, an adenosine metabolite ( 32, 43 ). Recent data indicate that it alsointeracts with nucleoside bases, preferring the purinergic basehypoxanthine ( 49 ). For this reason, it has been proposedthat hENT2 is involved in mechanisms requiring heavy adenosinemetabolism, such as ATP depletion in skeletal muscle caused bystrenuous exercise ( 7 ). Within the kidney, its purelybasolateral localization suggests that its major role is to function inconcert with CNTs in the salvage of nucleosides from the filtrate.
7 s0 U/ }( O6 q3 W. R8 _
Confocal microscopy of MDCK expressing the GFP-tagged splicevariant hENT2A indicates that the variant is not expressed on thesurface of these cells (Fig. 7 ). Furthermore, our data demonstrated that the variant did not function when expressed in oocytes, suggesting that it is not functional or lacks expression on the plasma membrane. In addition, expression of hENT2A did not affect the function ofwild-type hENT2, a phenomenon that has been demonstrated for splicedisoforms of other membrane proteins ( 20, 35, 41, 44 ). Therole of hENT2A is unknown.
$ V' K. ^1 c# X$ q+ ]9 n' }% E
In summary, we report that cellular hENT2 is localized exclusively tothe basolateral membrane and that hENT1 is localized primarily to thebasolateral membrane in renal epithelial cells. The COOH-terminaldileucine motif in hENT2 is implicated in surface expression of thisprotein; however, neither the dileucine motif nor the RXXV motif inhENT1 appears to be important for basolateral targeting.3 V" x, |6 W2 [, X  W
0 c# U9 l8 K; M: f
ACKNOWLEDGEMENTS# j9 a: M( |% B. ^' r/ H3 b7 o8 A

This work was supported by General Medical Sciences Grant inPharmaceutical Chemistry, Pharmacology and Toxicology nos. GM-07175 (L. M. Mangravite) and GM-42230 and by a grant fromGlaxoSmithKline (G. Xiao).
      【参考文献】2 `9 k. ]) i, C0 V$ H  N  E
1. Aroeti, B,Okhrimenko H,Reich V,andOrzech E. Polarized trafficking of plasma membrane proteins: emerging roles for coats, SNAREs, GTPases and their link to the cytoskeleton. Biochim Biophys Acta 1376:57-90,1998 .
& k. ?0 R0 M7 C6 L! w
$ n5 S  ]" f" y: B4 X+ f
; h. U) \! k6 i  z, {1 [
2. Belt, JA,andNoel LD. Nucleoside transport in Walker 256 rat carcinosarcoma and S49 mouse lymphoma cells. Differences in sensitivity to nitrobenzylthioinosine and thiol reagents. Biochem J 232:681-688,1985  .: @: I. |5 Q% m
* h& _3 K: m( A' Q. u9 `* \$ N

! }% _: @4 {8 {6 H/ e: l6 G. K; m3 D
3. Betcher, SL,Forrest JN, Jr,Knickelbein RG,andDobbins JW. Sodium-adenosine cotransport in brush-border membranes from rabbit ileum. Am J Physiol Gastrointest Liver Physiol 259:G504-G510,1990 .6 w4 n1 o" }  y, c4 c

# `- K% }4 l8 P6 `4 [- [1 x
4. Brandsch, M,Ganapathy V,andLeibach FH. H( )-peptide cotransport in Madin-Darby canine kidney cells: expression and calmodulin-dependent regulation. Am J Physiol Renal Fluid Electrolyte Physiol 268:F391-F397,1995 .

  \. C5 h8 x% F6 z7 `

5. Chandrasena, G,Giltay R,Patil SD,Bakken A,andUnadkat JD. Functional expression of human intestinal Na -dependent and Na -independent nucleoside transporters in Xenopus laevis oocytes. Biochem Pharmacol 53:1909-1918,1997  .
; r5 B8 Q1 r( W; Q
5 [1 L* d4 ]) F; D6 e

6. Ciruela, F,Blanco J,Canela EI,Lluis C,Franco R,andMallol J. Solubilization and molecular characterization of the nitrobenzylthioinosine binding sites from pig kidney brush-border membranes. Biochim Biophys Acta 1191:94-102,1994 .
1 l. s- p3 T1 _- R5 M3 ~6 b+ q

7. Crawford, CR,Patel DH,Naeve C,andBelt JA. Cloning of the human equilibrative, nitrobenzylmercaptopurine riboside (NBMPR)-insensitive nucleoside transporter ei by functional expression in a transport-deficient cell line. J Biol Chem 273:5288-5293,1998 .

3 Y; F4 e# o& i& L2 }. |$ r" A0 }. c

8. Distel, B,Bauer U,Le Borgne R,andHoflack B. Basolateral sorting of the cation-dependent mannose 6-phosphate receptor in Madin-Darby canine kidney cells. Identification of a basolateral determinant unrelated to clathrin-coated pit localization signals. J Biol Chem 273:186-193,1998 .# W! ?3 i, b1 n( i
" H7 X8 L. x% q7 A5 K. K1 P

9. Franco, R,Centelles JJ,andKinne RK. Further characterization of adenosine transport in renal brush-border membranes. Biochim Biophys Acta 1024:241-248,1990 .

' G# l( c! G# y, x
10. Gerstin, KM,Dresser MJ,Wang J,andGiacomini KM. Molecular cloning of a Na -dependent nucleoside transporter from rabbit intestine. Pharm Res 17:906-910,2000 ." q  Q- `' R& ~- Z
3 r/ ^$ P/ {6 e- a7 T+ I5 N% y
! z5 G* z, l9 x7 k1 p4 m

11. Griffith, DA,Doherty AJ,andJarvis SM. Nucleoside transport in cultured LLC-PK1 epithelia. Biochim Biophys Acta 1106:303-310,1992 .; p0 f& n7 t3 \

12. Griffiths, M,Beaumont N,Yao SY,Sundaram M,Boumah CE,Davies A,Kwong FY,Coe I,Cass CE,Young JD,andBaldwin SA. Cloning of a human nucleoside transporter implicated in the cellular uptake of adenosine and chemotherapeutic drugs. Nat Med 3:89-93,1997  .: M% ]7 M2 G% |8 e3 ?( c$ c9 y
& M6 g3 f+ @5 I
  z1 c+ p1 _  F8 P* a" w, Z% L" W! {' x

13. Griffiths, M,Yao SY,Abidi F,Phillips SE,Cass CE,Young JD,andBaldwin SA. Molecular cloning and characterization of a nitrobenzylthioinosine-insensitive (ei) equilibrative nucleoside transporter from human placenta. Biochem J 328:739-743,1997.. h7 a: y* J9 r  ~4 p* P

# x% Q  i! o$ T' T  f9 {( A! b! {

14. Hamilton, SR,Yao SY,Ingram JC,Hadden DA,Ritzel MW,Gallagher MP,Henderson PJ,Cass CE,Young JD,andBaldwin SA. Sub-cellular distribution and membrane topology of the mammalian concentrative Na -nucleoside co-transporter rCNT1. J Biol Chem 25:25,2001.

, I# }: P2 t( S9 ?  U
6 w' p; E* P7 S$ e8 a  F/ M
15. Handa, M,Choi D,Caldeiro RM,Messing RO,Gordon AS,andDiamond, II Cloning of a novel isoform of the mouse NBMPR-sensitive equilibrative nucleoside transporter (ENT1) lacking a putative phosphorylation site. Gene 262:301-307,2001  .

16. Heilker, R,Spiess M,andCrottet P. Recognition of sorting signals by clathrin adaptors. Bioessays 21:558-567,1999  .! z) `1 L1 k! t0 w! S! U9 [) T2 l

17. Jegla, T,andSalkoff L. A novel subunit for shal K  channels radically alters activation and inactivation. J Neurosci 17:32-44,1997 .

18. Jennings, LL,Hao C,Cabrita MA,Vickers MF,Baldwin SA,Young JD,andCass CE. Distinct regional distribution of human equilibrative nucleoside transporter proteins 1 and 2 (hENT1 and hENT2) in the central nervous system. Neuropharmacology 40:722-731,2001  .2 P, R* p) l" Q: E  l# o6 U" i
; x' R( w. j. V6 p% }! `( t
' r; m3 S+ H& p$ A4 C

19. Kiss, A,Farah K,Kim J,Garriock RJ,Drysdale TA,andHammond JR. Molecular cloning and functional characterization of inhibitor-sensitive (mENT1) and inhibitor-resistant (mENT2) equilibrative nucleoside transporters from mouse brain. Biochem J 352:363-372,2000.

2 n2 I# ]/ Y8 i: k" L
" l, y  Y6 }3 N* V4 X2 U9 Y' l* Y, E
20. Kitayama, S,Ikeda T,Mitsuhata C,Sato T,Morita K,andDohi T. Dominant negative isoform of rat norepinephrine transporter produced by alternative RNA splicing. J Biol Chem 274:10731-10736,1999 .

) A% F3 W& N2 v" @: L
& d! g2 h8 }$ W0 k( _
21. Kobayashi, S,Zimmermann H,andMillhorn DE. Chronic hypoxia enhances adenosine release in rat PC12 cells by altering adenosine metabolism and membrane transport. J Neurochem 74:621-632,2000  .

; T" A5 [4 `% l% [( X% |
22. Lai, Y,Bakken AD,andUnadkat JD. Simultaneous expression of hCNT1-CFP and hENT1-YFP in Mardin-Darby canine kidney (MDCK) cells: localization and vectorial transport studies. J Biol Chem 277:37711-37717,2002 .7 w: F1 v# t' l( c! j

23. Lee, CW,Cheeseman CI,andJarvis SM. Na - and K -dependent uridine transport in rat renal brush-border membrane vesicles. Biochim Biophys Acta 942:139-149,1988 .
) ]+ B- d+ F% l2 A
4 |" U/ O8 Y: {

24. Lee, CW,Cheeseman Ci,andJarvis SM. Transport characteristics of renal brush border Na - and K -dependent uridine carriers. Am J Physiol Renal Fluid Electrolyte Physiol 258:F1203-F1210,1990 .% Q; A6 P' L; {: a  {3 f
5 @6 ]- {* \- N3 e7 o( h. z; H- a
9 ^8 K' s; o# u+ f9 i7 y# |9 D

25. Le Hir, M. Evidence for separate carriers for purine nucleosides and for pyrimidine nucleosides in the renal brush border membrane. Renal Physiol Biochem 13:154-161,1990  .
8 y( y1 H% O& ]$ z& ^

5 r. Q4 T/ W; q  `5 D
26. Le Hir, M,andDubach UC. Concentrative transport of purine nucleosides in brush border vesicles of the rat kidney. Eur J Clin Invest 15:121-127,1985  .

9 s/ c, E* T) i7 ?1 ~

27. Mangravite, LM,Lipschutz JH,Mostov KE,andGiacomini KM. Localization of GFP-tagged concentrative nucleoside transporters in a renal polarized epithelial cell line. Am J Physiol Renal Physiol 280:F879-F885,2001 .8 e% m. o% m/ k$ b" z' ?% G

7 r. M: m% T! r
* ]8 U) t5 K8 B4 a: \, z  i
28. Masuda, S,Saito H,Nonoguchi H,Tomita K,andInui K. mRNA distribution and membrane localization of the OAT-K1 organic anion transporter in rat renal tubules. FEBS Lett 407:127-131,1997  .
+ e/ X7 W8 D8 l4 h
$ Z& ]9 r, w* m$ j' d

29. Masuda, S,Takeuchi A,Saito H,Hashimoto Y,andInui K. Functional analysis of rat renal organic anion transporter OAT-K1: bidirectional methotrexate transport in apical membrane. FEBS Lett 459:128-132,1999  .# m3 H& C; [( r4 \$ V# K* s

0 h- Z& t/ b9 B: H
& Z8 F1 K2 m6 q5 h$ X5 O+ E
30. Mostov, KE,Verges M,andAltschuler Y. Membrane traffic in polarized epithelial cells. Curr Opin Cell Biol 12:483-490,2000  .8 P; X9 n, c. q1 U2 V* V/ R. ^
; b9 t  I0 U* Q$ F$ s7 v

31. Mun, EC,Tally KJ,andMatthews JB. Characterization and regulation of adenosine transport in T84 intestinal epithelial cells. Am J Physiol Gastrointest Liver Physiol 274:G261-G269,1998 .
. m  U' ^5 W: m$ |3 q
- [; w: _& D$ b* Z: a2 A

32. Pennycooke, M,Chaudary N,Shuralyova I,Zhang Y,andCoe IR. Differential expression of human nucleoside transporters in normal and tumor tissue. Biochem Biophys Res Commun 280:951-959,2001  .

# [; H' F  z5 j/ Q% B; Y$ p

33. Ritzel, MW,Ng AM,Yao SY,Graham K,Loewen SK,Smith KM,Ritzel RG,Mowles DA,Carpenter P,Chen XZ,Karpinski E,Hyde RJ,Baldwin SA,Cass CE,andYoung JD. 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 276:2914-2927,2000 .

: {$ s/ w  h. q. ]
3 n4 c. e, S3 O
34. Ritzel, MW,Yao SY,Huang MY,Elliott JF,Cass CE,andYoung JD. Molecular cloning and functional expression of cDNAs encoding a human Na -nucleoside cotransporter (hCNT1). Am J Physiol Cell Physiol 272:C707-C714,1997 .5 w  F0 E5 C% U& h. W" R8 b" X: u0 d

  a; _  ?' n; F+ a8 N: ^9 Y2 G' G
35. Sairam, MR,Jiang LG,Yarney TA,andKhan H. Follitropin signal transduction: alternative splicing of the FSH receptor gene produces a dominant negative form of receptor which inhibits hormone action. Biochem Biophys Res Commun 226:717-722,1996  ./ ?* P; l% t6 W* K2 v: U5 _
) o% t. [! A" H

, i3 |0 a  H$ _; l
36. Saunders, C,Keefer JR,Kennedy AP,Wells JN,andLimbird LE. Receptors coupled to pertussis toxin-sensitive G-proteins traffic to opposite surfaces in Madin-Darby canine kidney cells. A1 adenosine receptors achieve apical and alpha 2A adrenergic receptors achieve basolateral localization. J Biol Chem 271:995-1002,1996 .

5 M2 Q+ r( {% Y6 V! e: R* K. [
. F" k2 K* @# X/ T- q; h  X+ i5 H
37. Scharrer, E,andGrenacher B. Active intestinal absorption of nucleosides by Na -dependent transport across the brush border membrane in cows. J Dairy Sci 84:614-619,2001 .# u* e- B7 H2 C3 ], q( Y; l
; e* \! F0 \" k2 k
' {6 ~0 n) d. c& q: Q- x

38. Sweet, DH,Miller DS,andPritchard JB. Localization of an organic anion transporter-GFP fusion construct (rROAT1-GFP) in intact proximal tubules. Am J Physiol Renal Physiol 276:F864-F873,1999 .
# x+ f2 g& P5 D4 m5 y

39. Vijayalakshmi, D,andBelt JA. Sodium-dependent nucleoside transport in mouse intestinal epithelial cells. Two transport systems with differing substrate specificities. J Biol Chem 263:19419-19423,1988 .7 b! B& `* ?# Q0 D& m% l- i$ T6 @
7 p5 J0 v% s' m9 L/ n3 S) L
& U2 G# L( m9 g  c! Z
, `/ @* j4 m6 h1 F( v3 v
40. Wang, J,Su SF,Dresser MJ,Schaner ME,Washington CB,andGiacomini KM. Na -dependent purine nucleoside transporter from human kidney: cloning and functional characterization. Am J Physiol Renal Physiol 273:F1058-F1065,1997 .

2 s5 {" s% x/ M, K; M% K7 y! Q$ i& ?
& N) Y1 U, I! G, p9 s9 D; _
41. Wang, Y,andMiksicek RJ. Identification of a dominant negative form of the human estrogen receptor. Mol Endocrinol 5:1707-1715,1991 .
6 F, D1 M% o2 T
; f( P9 A$ j/ e6 e& c$ R- L
5 `6 O! f8 b4 R( m, V+ D
42. Ward, JL,Sherali A,Mo ZP,andTse CM. 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 for guanosine and cytidine but a high affinity for inosine. J Biol Chem 275:8375-8381,2000 .. D! p! l& u% S2 |8 \
2 C; C5 v+ S: r2 j: Z: ~7 M8 X% L- _

43. Ward, JL,andTse CM. Nucleoside transport in human colonic epithelial cell lines: evidence for two Na -independent transport systems in T84 and Caco-2 cells. Biochim Biophys Acta 1419:15-22,1999 .
( S2 Y8 Z6 a: _% i$ k1 s3 P

44. Watanabe, K,Fukuchi T,Hosoya H,Shirasawa T,Matuoka K,Miki H,andTakenawa T. Splicing isoforms of rat Ash/Grb2. Isolation and characterization of the cDNA and genomic DNA clones and implications for the physiological roles of the isoforms. J Biol Chem 270:13733-13739,1995 .+ _( _6 q: x( c& a) s/ p

' o7 I; z7 V5 I2 ]; N
45. Williams, TC,Doherty AJ,Griffith DA,andJarvis SM. Characterization of sodium-dependent and sodium-independent nucleoside transport systems in rabbit brush-border and basolateral plasma-membrane vesicles from the renal outer cortex. Biochem J 264:223-231,1989  .
7 L+ T* s- p- I- ^
" O7 I2 w% ~# a9 U. x7 G9 T" Q7 `
% S. H5 B  e" u4 }. }7 i5 l
46. Williams, TC,andJarvis SM. Multiple sodium-dependent nucleoside transport systems in bovine renal brush-border membrane vesicles. Biochem J 274:27-33,1991.

47. Xiao, G,Wang J,Tangen T,andGiacomini KM. A novel proton-dependent nucleoside transporter, CeCNT3, from Caenorhabditis elegans. Mol Pharmacol 59:339-348,2001 .5 E$ d5 `* M( ~$ v" f

48. Yao, SY,Ng AM,Muzyka WR,Griffiths M,Cass CE,Baldwin SA,andYoung JD. Molecular cloning and functional characterization of nitrobenzylthioinosine (NBMPR)-sensitive (es) and NBMPR-insensitive (ei) equilibrative nucleoside transporter proteins (rENT1 and rENT2) from rat tissues. J Biol Chem 272:28423-28430,1997 .' n2 K+ }; p0 M+ [: Y1 c8 U
% i7 X" p6 I- k) e6 J$ m: p

# c1 w4 V+ N# u% O
49. Yao, SY,Ng AM,Vickers MF,Sundaram M,Cass CE,Baldwin SA,andYoung JD. Functional and molecular characterization of nucleobase transport by recombinant human and rat equilibrative nucleoside transporters 1 and 2. Chimeric constructs reveal a role for the ENT2 helix 5-6 region in nucleobase translocation. J Biol Chem 277:24938-24948,2002 .

9 F7 C" [2 q& \% z9 v. T
50. Yeaman, C,Grindstaff KK,andNelson WJ. New perspectives on mechanisms involved in generating epithelial cell polarity. Physiol Rev 79:73-98,1999 .& s7 B/ h3 Y7 ^: T! p+ H
& j: k& b" J. X# ]! p
- c5 ^' E) t# P7 ^- `; I/ C
' Q, e5 R' k) C9 x, w
51. Zegers, MM,andHoekstra D. Mechanisms and functional features of polarized membrane traffic in epithelial and hepatic cells. Biochem J 336:257-269,1998.

收藏0 分享0 顶0 踩0