Role of glycosylation in the renal electrogenicNa -HCO - 《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 ... 23 下一页

返回列表

查看: 403173\|回复: 222	go Role of glycosylation in the renal electrogenicNa -HCO [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 13:38 |只看该作者 |倒序浏览 |打印

作者：InyeongChoi, LihuiHu, José D.Rojas, Bernhard M.Schmitt,  WalterF.Boron作者单位：Department of Cellular and Molecular Physiology, YaleUniversity School of Medicine, New Haven, Connecticut 06520 ; V; F6 \. m- s# v
   8 ^5 E1 A" G1 L5 i: Q


                     8 J6 p( l+ c4 c$ @' y
         4 r& n: k- M/ A% N! }; T4 ~

         # D) Z" `* N, \0 |# G
         6 s/ C; W  r3 O- o0 v6 z
                  # V# a; c' p0 h

      ' E  A! i; K$ H( i+ K; b6 w5 \- T

      【摘要】
   Theelectrogenic Na -HCO 3 − cotransporter NBCe1is important for the regulation of intracellular pH (pH i )and for epithelial HCO 3 − transport in many tissues,including kidney, pancreas, and brain. In the present study, weinvestigate glycosylation sites in NBCe1. Treatment of rat kidneymembrane extracts with peptide N-glycosidase F (PNGase F) shifted theapparent molecular weight (MW) of NBCe1 from 130 to 116, the MWpredicted from the deduced amino acid sequence. Treatment withendoglycosidase F 2 or H or O-glycosidase did not affect theMW of NBCe1. Lectin-binding studies, together with the enzyme data,suggest that the N-linked carbohydrates are of tri- or tetra-antennarytype. To localize glycosylation sites, we individually mutated theseven consensus N-glycosylation sites by replacing asparagine (N)with glutamine (Q) and assessing mutant transporters in Xenopuslaevis oocytes. Immunoblotting of oocyte membrane extracts treatedwith PNGase F indicates that NBCe1 is normally glycosylated at N597 andN617 (both on the third extracellular loop). However, N592 (on the sameloop) is glycosylated when the other two sites are mutated. Thetriple mutant (N592Q/N597Q/N617Q) is completely unglycosylated but,based on microelectrode measurements of membrane potential andpH i in oocytes, preserves the Na andHCO 3 − dependence and electrogenicity of wild-type NBCe1.
      【关键词】 transporter pH measurement acidbase mechanism3 o% m# ]- G: t+ Z1 b. s
   INTRODUCTION

THE MOST IMPORTANT pHbuffer system in the body is CO 2 /HCO 3 −.Transporters that carry HCO 3 − not only have atremendous impact on the regulation of intracellular pH(pH i ), but they also play an important role in the movement of NaHCO 3 and fluid across several epithelia. Inparticular, the electrogenic Na -HCO 3 − cotransporter plays key roles in both pH i regulation andepithelial transport. In astrocytes, for example, electrogenicNa -HCO 3 − cotransporters are a majormechanism of HCO 3 − uptake ( 5, 11, 16 ). Inthe kidney, electrogenic Na -HCO 3 − cotransporters in the proximal tubule ( 7 ) are central inthe reabsorption of ~85% of the HCO 3 − filtered inthe glomeruli ( 4 ). In pancreatic ducts, electrogenic Na -HCO 3 − cotransporters are keycomponents of the mechanism responsible for HCO 3 − secretion by the exocrine pancreas ( 21 ).

The key step in understanding the molecular physiology of electrogenicNa -HCO 3 − cotransporters was theexpression cloning from the salamander kidney of the electrogenicNa -HCO 3 − cotransporter NBC( 35 ). Together with the previously clonedClHCO 3 − exchangers or AEs ( 3, 12, 23 ),this first NBC defined a new bicarbonate transporter superfamily. Otherclosely related electrogenic NBCs were soon cloned from mammaliankidney ( 8, 34 ), pancreas ( 1 ), heart( 10 ), and brain ( 6 ). Indeed,immunocytochemistry studies localize NBCe1 to the basolateral membraneof the proximal tubule ( 38 ), the basolateral membrane ofpancreatic duct cells ( 27 ), and astrocytes and neurons inthe brain ( 37 ). It is now clear that these NBC clones,which we will refer to as NBCe1, are alternative splice products of thesame gene ( 2 ). NBCe1-A (also called kNBC1) is presentmainly in the kidney, NBCe1-B (or pNBC1) is widely distributed but ispresent at particularly high levels in the pancreas, and NBCe1-C (orbNBC1) is present mainly in the brain.1 S# W8 @$ v! Z7 J! x" L; \
% Y# \, T  i; y
The primary structure of rat NBCe1-A predicts a protein of 1,035 aminoacids that is 30-40% identical to the AEs. Topological studies onAE1 suggest that the protein has cytoplasmic NH 2 and COOHtermini and as many as 14 transmembrane (TM) segments ( 14, 32 ). AE1 is N-glycosylated once in the fourthextracellular loop, between TM7 and TM8, whereas AE2 and AE3 haveconsensus glycosylation sites in the third extracellular loop, betweenTM5 and TM6. The oligosaccharide on human AE1 takes two forms, along-chain polylactosaminyl structure or a shorter complex structure( 24 ). Mutation of the N-glycosylation site in AE1 reducestargeting the protein to the cell surface in Xenopus oocytestudies, but the nonglycosylated protein is still functional( 19 ). One role of N-glycosylation in AE1 may be to permitthe interaction of AE1 with calnexin, an endoplasmic reticulumchaperone; altering the N-glycosylation site by mutagenesis eliminatesthe interaction of AE1 with calnexin ( 31 ). AE2 is alsoglycosylated, but it lacks sialic acids ( 44 ). Theglycosylation status of AE3 is unknown.! J: ~/ K( R, w
2 B$ e8 @( I; x& Z( ~& b* B2 H
The deduced amino acid sequence predicts that rat NBCe1 has sevenputative consensus sites for N-linked glycosylation. On the basis of anAE1 topology model ( 14 ), three of the sites (N33, N199,and N208) are on the cytoplasmic NH 2 terminus, one (N497)is within TM3, and the remaining three (N592, N597, and N617) are onthe third extracellular loop. These last three sites are thuscandidates for glycosylation. Western blots show that NBCe1 migrates at130 kDa in the kidney ( 38 ), whereas the deduced amino acidsequence predicts a 116-kDa protein. Thus at least one of the consensussites may be glycosylated. In the present study, we investigated theoligosaccharide composition of NBCe1-A using a glycosidase enzyme assayand a lectin-binding assay. To locate the N-linked glycosylation sites,we individually mutated asparagine (N) to glutamine (Q) at all sevenputative glycosylation sites and also created double and triplemutants. We analyzed these mutants by expressing them in Xenopus oocytes and examining the molecular weight (MW) ofeach mutant transporter after treatment of peptide N-glycosidase F(PNGase F). We also performed microelectrode experiments to assess thefunction of mutant transporters. We found that NBCe1-A is normallyglycosylated at two sites on the third extracellular loop and that theunglycosylated protein has normal function.$ O, j# N! ]. o. Z4 N

MATERIALS AND METHODS

Extraction of crude membranes. A whole kidney from adult Sprague-Dawley rats or rabbits was removedunder anesthesia (intraperitoneal injection, 100 mg/kg pentobarbitalsodium). A frozen kidney from a cow was obtained commercially, anda kidney from the Ambistoma tigrium salamander was isolatedas described previously ( 38 ). Slices of the kidney fromeach species were placed in an ice-cold homogenization buffer (HB)-I(250 mM sucrose, 20 mM HEPES, pH 7.4, 100 mM NaCl, 2 mM EDTA, 1 mMPMSF, 1 µM leupeptin, 1 µM pepstatin) and homogenized using aPolytron. For extraction of crude plasma membranes of oocytesexpressing NBCe1, oocytes were homogenized in hypotonic HB-II (7.5 mMsodium phosphate, pH 7.4, 1 mM EDTA, 1 mM PMSF, 1 µM leupeptin, 1 µM pepstatin) as described previously ( 10 ). Homogenateswere centrifuged at 750 g in a microcentrifuge (Eppendorf, model 5415C) for 5 min at 4°C to remove cell debris; the supernatant was centrifuged for 30 min at 16,000 g at 4°C. The pellet,which contained most plasma and organellar membranes, was used for analysis., ~5 h# E. U, G$ N- H  {; P
5 \, w" f7 s) Q  {) Z$ V0 i
Deglycosylation. Membrane fractions (200 µg) isolated from rat tissues were heatdenatured at 70°C for 10 min in 0.2 M -mercaptoethanol, 0.5% SDSand protease-inhibitor cocktail. Some samples were then incubated overnight at 37°C in an incubation solution (40 mMNa 2 HPO 4, pH = 7.4, 10 mM EDTA, 1% -mercaptoethanol, and 0.6% Triton X-100) containing PNGase F (8,000 mU/ml) or endoglycosidase (Endo) F 2 (3.2 mU/ml). Othersamples were treated overnight at 37°C in 200 mM sodium acetatebuffer (pH = 5) containing either Endo H (40 mU/ml) orO-glycosidase (8 mU/ml). For membrane fractions extracted from oocytes,pellets were dissolved in 20 µl of Na 2 HPO 4 (40 mM, pH 7.0), and SDS was added to a final concentration of 1%.After the membrane fractions were denatured at 70°C for 10 min, they were incubated at 37°C for 6 h in the presence of 5 U ofN-glycosidase F in the incubation solution. The reaction was stopped byadding the SDS gel-loading buffer.
, S3 O5 x  p7 h2 X6 X5 v2 W
Lectin binding. We used the following lectins: concanavalin A (Con A), wheat germagglutinin (WGA), glycine max agglutinin (soy bean agglutinin; SBA),jacalin, Galanthus nivalis (GNA), Lens culinaris (LCA), Ulex europaeus (UEA-I), and Tetragonobulyspurpurea (LTA). Lectins coupled to agarose beads (Sigma, St.Louis, MO) were equilibrated at room temperature with thelectin-binding buffer, which contained (in mM) 150 NaCl, 20 Tris, pH7.5, 1 MgCl 2, 1 MnCl 2, 1 CaCl 2, and1 ZnSO 4, 0.5% Triton X-100, and protease-inhibitorcocktail. Agarose beads were then preabsorbed with 1% BSA in thelectin-binding buffer for 5 min at RT, followed by three 10-min washesin the lectin-binding buffer. Crude membrane proteins (50 µg) werethen added to the lectin-binding buffer containing lectin-agarose beads to a final volume of 200 µl. For competition experiments, sugars (methyl- - D -mannopyranoside for Con A and GNA; galactosefor SBA and jacalin; N -acetyl- D -glucosamine forWGA; and -mannose for LCA) were added to a final concentration of0.28 M. For UEA-1 and LTA, - L -fucose was used to a finalconcentration of 1 M. The reaction mixture, with or without competingsugar, was incubated overnight at 4°C in a rotating shaker.Afterward, the beads were washed three times with the lectin-bindingbuffer for 10 min.: Q: G8 H  N0 @0 w. V  Z
+ b; }) W% k! d! `5 g  Q, P+ m
Immunoblotting. Membrane extracts treated with glycosidases or material fromlectin-binding assays were mixed with Laemmli loading buffer and heatedfor 10 min at 70°C. The proteins were then separated on a 7.5%SDS-polyacrylamide gel and transferred overnight to a polyvinylidenedifluoride membrane (Immobilon-P, Millipore). Blots werepreincubated for 1 h in a blocking buffer containing 5% nonfatdry milk and 0.1% Tween 20 in Tris-saline (TBS; 50 mM Tris, pH 7.4;150 mM NaCl) and then incubated with an antibody (1:500 dilution),either NBC-5 ( 38 ) or K1A ( 6 ), whichrecognizes the COOH terminus of NBCe1-A. After several washes with TBScontaining 0.1% Tween 20, blots were incubated with anti-rabbit IgGconjugated to horseradish peroxidase (HRP; Chemicon) for 1 h(1:5,000 dilution). Blots were washed three times and developed by anHRP/hydrogen peroxide-catalyzed oxidation of luminol under alkalineconditions (Pierce, Rockford, IL).

Site-directed mutagenesis. The plasmid pTLN-NBCe1-A ( 35 ) served as a template forsite-specific mutagenesis as described by Pusch et al.( 33 ). Seven mutagenic primers were designed to replace thecodons for Asn (N) at each of the putative glycosylation sites (N33,N199, N208, N497, N592, N597, and N617). We created a double (ortriple) mutant with the same approach, but using a single (or double)mutant as the template. The final constructs were sequenced to verify the desired sequences.3 h+ j2 n! K; }6 ?4 d

Measurement of V m and pH i. Wild-type and mutant cDNA were transcribed in vitro using a mMessagemMachine kit (Ambion, Austin, TX) with SP6 RNA polymerase. Defolliculated Xenopus oocytes (Stage V-VI) were prepared asdescribed previously and injected with 20 ng of cRNA or 50 nl water and incubated in OR3 media (50% Leibovitz L-15 media with L -glutamine, 5 mM HEPES, pH 7.5) supplemented with 5 U/mlpenicillin/streptomycin ( 35 ). Injected oocytes weremaintained for 3-7 days at 18°C before use. Voltage and pHmicroelectrodes were prepared as described previously( 17 ). The pH electrode tip was filled with proton ionophore 1 cocktail B (Fluka Chemical, Ronkonkoma, NY) and back filledwith a pH 7 phosphate buffer. Electrodes were connected tohigh-impedance electrometers (model FD-223; WPI, Sarasota, FL), whichin turn were connected to the analog-to-digital converter of acomputer. In electrophysiological experiments, theCO 2 /HCO 3 − -free ND-96 solution contained(in mM) 96 NaCl, 2 KCl, 1.8 CaCl 2, 1 MgCl 2, and5 HEPES (pH 7.5); osmolality was adjusted to 196-200 mosmol/kgH 2 O. In solutions equilibrated with 1.5%CO 2 (pH 7.5), we replaced 10 mM NaCl with 10 mMNaHCO 3.: v1 y1 E1 G( P1 Q! i

Statistics. Data are reported as means ± SE. Levels of significance wereassessed using the unpaired, two-tailed Student's t -test. A P value of i change (dpH i /d t ) were fitted bya line using a least-square method.
$ {3 [. j; p5 H  A* x7 F2 b
RESULTS8 W. i* E7 q' b2 W. U

N-glycosylation of native NBCe1 in different species. To test whether the N-glycosylation of NBCe1 is a general phenomenonamong various species, we first prepared membrane extracts preparedfrom the kidneys of rabbit, rat, bovine, and salamander and thentreated these extracts with PNGase F. This enzyme digests the amidebond between the reducing end of N -acetylglucosamine and the -amide group of asparagine. We examined the effect of PNGase Ftreatment on the MW of the NBCe1 protein by immunoblot analysis, usingan antibody specific to the COOH terminus of the protein. Figure 1 shows that PNGase F treatment reducedthe apparent MW of NBCe1 in all species examined. The immunoreactivebands in rabbit, rat, and bovine decreased from ~130 to ~116, which is the size expected from the deduced amino acid sequence. Not shownare data demonstrating that PNGase F treatment also reduces theapparent MW of NBCe1 from ~130 to ~116 in mouse, human, and guineapig kidneys. These results suggest that N-glycosylation of NBCe1 is ageneral phenomenon in mammals. As noted previously, salamander NBCe1has an apparent MW of ~160 ( 38 ), which is 30 higher thanin mammalian species. Because PNGase F treatment reduced the apparentMW of salamander NBCe1 to ~130, rather than the predicted MW of 116, it is possible that salamander kidney NBCe1 has an additionalO-glycosylation site(s).
6 n/ G/ p" o- b7 x8 S2 `0 {
Fig. 1. Immunoblot showing N-linked glycosylation of renalelectrogenic Na -HCO 3 − cotransporter(NBCe1) from rabbit, rat, bovine, and salamander. The crude membranesextracted from kidneys were incubated in the absence ( ) or presence( ) of peptide N-glycosidase F (PNGase F). Membrane proteins wereseparated on a 7.5% SDS-polyacrylamide gel, transferred to a PVDFmembrane, and probed with an antibody to the COOH terminus of ratNBCe1.

Previous work on preparations not treated with PNGase F showed thatthe antibody recognizes other lower MW bands in some preparations ( 37, 38 ) but not in others ( 22, 42 ). It isinteresting to note that PNGase F treatment caused MWs of the faint 101 band of rabbit kidney to fall to 86, the 97 band of rat kidney to fall to 82, and the darker 103 band of bovine kidney to fall to 89, decreases of ~15 in each case. These results support the hypothesis that the lower MW bands represent COOH-terminal proteolytic products.! ]( f- @8 r9 e$ B* A1 ?( e+ Q

Glycosidase treatment of native rat NBCe1. To investigate carbohydrate components of the protein, we focused onrat kidney NBCe1, which has been characterized in terms of both itsfunctional properties ( 18, 34, 39 ) and its cellular localization ( 6, 28, 37, 38 ). In Fig. 2, lane 1 (no enzyme treatment) and lane 2 (PNGase F) confirm the rat data ofFig. 1, which indicate that kidney NBCe1 has an N-glycosylatedoligosaccharide. To verify directly the absence of an O-glycosylatedcarbohydrate-peptide linkage, we next treated rat kidney membraneextracts with an O-glycosidase, which digests the bond between thereducing end of N -acetyl-galactosamine (GalNAc) and thehydroxyl group of serine or threonine. We found that treatment with acombination of O-glycosidase and sialidase (separate experiments showedthat sialidase had no effect by itself) did not change the MW of theprotein, indicating absence of O-glycosylation (Fig. 2, compare lanes 2 and 3 ). To further characterize thecarbohydrate component of NBCe1, we treated the protein with EndoF 2, which cleaves biantennary complex-type oligosaccharides, or Endo H, which cleaves oligosaccharides of the highmannose and hybrid types. Because neither treatment shifted the MW ofNBCe1 (Fig. 2, lanes 4 and 5 ), we can concludethat NBCe1 has an N-linked oligosaccharide that is probably of the tri-or tetra-antennary type.$ C  j0 `, o. x! h; O2 \1 B6 d& r
' [: i. G: e3 Q2 i0 W. G! @
Fig. 2. Sensitivity of rat renal NBCe1 to various glycosidases.The enzymes used were PNGase F, O-glycosidase, endoglycosidase (Endo)F 2, and Endo H. The O-glycosidase preparation alsocontained sialidase, which by itself had no effect on apparentmolecular weight., ^+ k& |: Y1 [! N

Lectin binding of native rat NBCe1. Because a lectin affinity profile can provide some insight into thenature of the linked oligosaccharide, we performed lectin affinitychromatography with a series of lectins immobilized to agarose beads.We incubated crude membrane extracts with lectin-agarose beads, allowedglycoprotein to bind to the lectin, added either nothing or a competingmonosaccharide (see MATERIALS AND METHODS ) to elute boundglycoprotein from the beads, isolated the beads, eluted boundglycoprotein (if present) from the beads, and then performed animmunoblot on the supernatant.
  W+ Y( t7 ~+ l0 a7 k! @
As shown in Fig. 3, the NBCe1 proteinbound to Con A, glycine max agglutinin (i.e., SBA), and WGA. Con A hasan affinity for D -glucopyranose or D -mannopyranose with unmodified hydroxyl groups at the C3,C4, and C6 positions. Elution withmethyl- - D -mannopyranoside markedly reduced thebinding of NBC to Con A; the small residual signal probably reflectsincomplete washing of the beads with the hapten sugar. In otherexperiments (not shown), we saw complete dissociation of the NBCe1protein from the Con A-agarose beads. SBA preferentially binds tooligosaccharide structures with terminal GalNAc and, to a lesserextent, terminal galactose residues. Because NBCe1 did not bind tojacalin (Fig. 3 ), it is unlikely that SBA interacted with NBCe1 via aterminal nonreducing -linked galactose. Therefore, SBA probablyinteracted with a terminal GalNAc residue on NBCe1. WGA has specificityfor internal GlcNAc residues and, to a lesser extent, sialic acidresidues.. q4 w6 Y; l/ P8 L- M

Fig. 3. Lectin binding of rat kidney NBCe1. Lectins coupled toagarose beads were incubated with crude membrane proteins from ratkidney. Afterward, either nothing was added ( ), or a competingmonosaccharide was added ( ) to elute bound glycoprotein from thebeads. Immunoblotting was performed with an antibody to the COOHterminus of NBCe1. Lectins used in the experiments are concanavalin A(Con A), jacalin, lectins from Galanthus nivalis (GNA), soybean agglutinin (SBA), Lens culinaris (LCA), Tetragonobulys purpurea (LTA), wheat germ agglutinin (WGA),and Ulex europaeus (UEA-I).9 Q) H% {' u$ a9 r# H( C) h7 x
% X2 \8 y0 R# _4 d! u
In addition to jacalin, rat kidney NBCe1 does not bind to lectins fromGNA, LCA, UEA-I, or LTA. The insensitivity to GNA indicates the absenceof a terminal, nonreducing -linked mannose, which suggests that theNBCe1 oligosaccharides are neither of the high mannose nor of thehybrid type. This interpretation is consistent with the insensitivityof NBCe1 to Endo F 2 or Endo H. The absence of binding toLCA, LTA, and UEA-I indicates an absence of core and outer fucoseresidues. It is interesting to note that all three of thesefucose-dependent lectins did interact with NBCe1 isolated from ratbrain (not shown).5 @% R. x! M* P! ~0 N* n: Q" _+ b( {8 Q

Mutational analysis of consensus N-glycosylation sites. As noted before, the amino acid sequence of rat kidney NBCe1-A predictsseven consensus N-glycosylation sites (N-X-S/T): N33, N199, N208, N497,N592, N597, and N617 (Fig. 4 ). To testwhether these consensus sites are responsible for N-glycosylation, we individually mutated each asparagine (N) to glutamine (Q). We thenheterologously expressed each mutant transporter in Xenopus oocytes, which are known to functionally express wild-type rat kidneyNBCe1-A ( 34 ). After extracting crude oocyte plasmamembranes, and treating some of them with PNGase F, we analyzed themutant proteins by immunoblotting. As shown in Fig. 5, wild-type NBCe1-A expressed in oocytesand then either not treated or treated with PNGase F, had the sameapparent MWs (i.e., ~130 and ~116, respectively) as native NBCe1 inmammalian tissues (see Figs. 1 and 2 ). The same is true for each of thefirst five N-to-Q mutants. It is interesting to note that N592(predicted to be on the third extracellular loop, 28 amino acidsdownstream from the end of TM5) does not appear to be glycosylatedsignificantly in the wild-type NBCe1 protein.

Fig. 4. Consensus sites for N-glycosylation in rat NBCe1. Themembrane topology is based on the topology of AE1 ( 14 ) andNBCe2 ( 41 ). The first 3 sites (N33, N199, and N208) are inthe predicted cytoplasmic NH 2 terminus, the fourth site(N497) is in the predicted transmembrane segment 3, and thelast 3 sites (N592, N597, and N617) are in the third extracellularloop, between transmembrane segment 5 and 6.

Fig. 5. PNGase F treatment of the membrane extracts from oocytesexpressing the mutant transporters. Mutant transporters wereindividually generated by replacing asparagine (N) with glutamine (Q)at each glycosylation site and then expressed in oocytes. The crudemembranes extracted from oocytes were incubated in the absence( ) or presence ( ) of PNGase F. WT, wild type./ K3 @! b3 [& W' y9 K+ d) j

For both N597Q (which expressed only modestly in this experiment) andN617Q, the MW in the absence of PNGase F treatment was only ~122,consistent with the hypothesis that N597 and N617 contribute aboutequally to glycosylation in the wild-type protein. Treatment withPNGase F reduced the MW to ~116 for each of these two mutants.

If N597 and N617 both normally contribute to the glycosylation ofwild-type NBCe1-A, then one might predict that the double mutantN597Q/N617Q should have a MW of 116, and PNGase F should fail toproduce a MW shift. However, as shown in Fig. 6, in the absence of PNGase F,N597Q/N617Q exhibited two bands, one at a MW of ~116 (presumablyrepresenting the predicted, nonglycosylated protein) and another at~120 (presumably representing glycosylation at a residue other thanN597 or N617). Treatment with PNGase F yielded only the band at ~116.The presence of the doublet in the absence of PNGase F implies that analternative glycosylation site may be used when the two normalglycosylation sites are eliminated. To test the hypothesis that N592 isthe alternative glycosylation site, we created the triple mutantN592Q/N597Q/N617. Indeed, this triple mutant yielded a single band at~116, the expected MW of the nonglycosylated protein, both in theabsence and the presence of PNGase F. Thus, in the absence of all threeconsensus N-glycosylation sites on the third extracellular loop, Xenopus oocytes failed to glycosylate NBCe1 measurably. Weconclude that the oocyte can glycosylate N592 when bulky sugar chainsare not present at positions 597 and 617. The oocyte may alsoglycosylate N592 to a lesser extent when glycosylation occurs atposition 617 but not 597: on the original films for Fig. 5 B, bottom, as well as for a similar experiment(not shown), it is possible to see a faint doublet at ~122 kDa.

Fig. 6. PNGase F treatment of the membrane extracts from oocytesexpressing the double (N597Q/N617Q) or triple mutant(N592/N597Q/N617Q). The crude membranes extracted from oocytes wereincubated in the absence ( ) or presence ( ) of PNGase F.; l* i/ {! V3 w* b  q0 Q) M. c
$ h" O# K+ D* Z# x5 [9 B& V+ K
Functional expression of unglycosylated NBCe1. To test the functional significance of the glycosylation of rat kidneyNBCe1, we expressed the triple mutant N592Q/N597Q/N617Q in Xenopus oocytes and then used microelectrodes to monitorpH i and membrane potential ( V m ).Previous work showed that, as expressed in Xenopus oocytes,wild-type NBCe1-A is electrogenic, Na dependent, andHCO 3 − dependent ( 34 ). As shown in Fig. 7 A, for oocytes expressing thetriple mutant, introducing 1.5% CO 2 /10 mMHCO 3 − into the extracellular solution elicited a rapidfall in pH i (due to CO 2 influx), followed by anincrease that reflects the uptake of HCO 3 − or arelated ion such as CO 3 2−. The introduction ofCO 2 /HCO 3 − also caused an immediatehyperpolarization (Fig. 7 B ), indicative of the electrogenic influx of Na with two or more HCO 3 − ions(or an equivalent species, such as CO 3 2− ). Thenegative shift in V m was followed by a decay,the time course of which approximately followed the time course of thepH i recovery.
! g. f- J3 W2 ]: x  l# F
Fig. 7. Microelectrode measurements of the intracellular pH(pH i ) and membrane potential ( V m ) inan oocyte expressing the triple mutant (N592/N597Q/N617Q). The oocytewas initially bathed with CO 2 /HCO 3 − -freeND-96 solution. During the indicated period, the oocyte was superfusedwith 1.5% CO 2 /10 mM HCO 3 −. The mean rateof pH i change (dpH i /d t ) was13.8 ± 1.7 × 10 5 pH U/s during thepH i recovery from the CO 2 -induced acid load and 25.3 ± 7.3 × 10 5 pH U/s during theacidification elicited by removing Na ( n = 6). These values are not significantly different from those in oocytesexpressing wild-type NBCe1. For these, the meandpH i /d t was 14.8 ± 1.8 × 10 5 pH U/s during the pH i recovery( n = 5; P = 0.35 in unpaired, 2-tailed t -test with comparable value for the triple mutant) and 27.6 ± 4.9 × 10 5 pH U/s in the absence ofNa ( n = 5; P = 0.41). Inoocytes expressing the mutant transporter, applyingCO 2 /HCO 3 − caused a mean hyperpolarizationof 63.4 ± 2.5 mV, whereas removing Na caused a meandepolarization of 53.4 ± 1.8 mV ( n = 6). Inoocytes expressing wild-type NBCe1 ( n = 5), applyingCO 2 /HCO 3 − caused a mean hyperpolarizationof 45.2 ± 7.7 mV ( n = 5; P = 0.02), whereas removing Na caused a mean depolarizationof 31.8 ± 5.1 mV ( n = 5; P = 0.001).

Removing extracellular Na reversibly converted thepH i recovery to a very slow acidification, reflectingreversal of the mutant transporter. Simultaneously, Na removal caused a rapid depolarization, consistent with the electrogenic efflux of Na plus two or more HCO 3 − ions(or their equivalent). A statistical analysis (Fig. 7 legend) showsthat the rates of pH i change, both during thepH i recovery from the CO 2 -induced acid and thepH i decline elicited by Na removal, were notsignificantly different between oocytes expressing the triplemutants and those expressing wild-type NBCe1-A. The magnitudes of the V m changes elicited by applyingCO 2 /HCO 3 − or removing Na weresignificantly greater for oocytes expressing the triple mutant.However, this latter observation could reflect differences in severalother parameters (e.g., membrane resistance) not related to thetransport rate. We conclude that, similar to wild type, thetriple mutant is electrogenic and Na dependent andactivated by applying CO 2 /HCO 3 −. Theseresults indicate that glycosylation is not required for the basicfunction of the protein. We obtained similar results (not shown) inmutants lacking one or both of the two natural glycosylation sites(i.e., N597Q, N617Q, and N597Q/N617Q).

DISCUSSION" b5 V/ @! f3 J# O. J  p

Carbohydrate structure of rat kidney NBCe1. In this study, we provide the first direct evidence for theglycosylation of any of the Na -coupledHCO 3 − transporters. Specifically, the MW shiftproduced by PNGase F indicates that the electrogenic Na -HCO 3 − cotransporter from rat kidney NBCe1 is a glycoprotein in which the carbohydrate moiety is linked tothe protein by an N-glycosidic bond. The insensitivity of rat kidneyNBCe1 to either Endo F 2 or Endo H suggests thatoligosaccharide structure of rat kidney NBCe1 is neither of thebiantennary nor of the high mannose/hybrid type. Thus the carbohydrateis probably of the tri- or tetra-antennary/complex type. Thelectin-binding assays further suggest that the carbohydrate componentsof rat kidney NBCe1 probably do not include a terminal nonreducing -linked galactose (insensitivity to jacalin); a terminal nonreducing -linked mannose (insensitivity to GNA); or either a core or outerfucose (insensitivity to LCA, LTA, and UEA-I). On the other hand, the carbohydrate component may include a terminal GalNAc (sensitivity toSBA), either an internal GlcNAc or a terminal sialic acid (WGA sensitivity), and a D -glucopyranose or D -mannopyranose with unmodified C3, C4, and C6 hydroxylgroups (Con A sensitivity).

The MW of the glycosylated renal NBCe1 is ~130 in rabbit, rat, andbovine. In rat, NBCe1 bands of similar MW appear in other tissues,including brain ( 6 ) and parotid gland ( 36 ).On the other hand, the MW of NBCe1 is ~160 in rat epididymis( 22 ), and our unpublished data show that NBCe1 is mostlypresent in an unglycosylated form in rat stomach. Therefore, thecarbohydrate component of NBCe1 is organ specific.  \3 Z% f* h5 v
7 E) {1 U  k: z, c/ p: E
Our data indicate that the N-glycosylation of NBCe1 is common amongvarious species. A sequence comparison reveals that three consensusglycosylation sites on the third extracellular loop of NBCe1 (i.e.,N592, N597, and N617) are conserved among bovine, rabbit, human, mouse,and salamander. Salamander NBCe1 also has a fourth N-glycosylation onthis loop. In addition, salamander NBCe1 has an additional carbohydratelinkage, probably an O-glycosylation because PNGase F treatment reducedthe apparent MW of salamander NBCe1 to ~130, rather than thepredicted MW of 116. It is interesting to note that the thirdextracellular loop of salamander NBCe1 has a cluster of threonines thatare flanked by strongly hydrophobic residues, reminiscent of amucin-type O-glycosylation site ( 13 ).$ }8 C; Y$ l7 U5 [5 J
# B" e" k3 w- F/ ]% J, O" R
Glycosylation sites. Our data indicate that the N-linked glycosylation in NBCe1 normallyoccurs at N597 and N617 and thus prove that both of these residues,which are located between putative TM5 and TM6, face the extracellular fluid.* R* }- x6 z- x1 f& W6 ~5 s1 U
5 c" C; Z$ t8 Y2 u$ G1 x
Glycosylation at N597 and N617 probably contributes about equally tothe total glycosylation of the wild-type protein because 1 )disrupting either N597 or N617 reduces the apparent MW of the mutantprotein from ~130 to ~122, about one-half of the way to the MW ofthe unglycosylated protein; and 2 ) treatment of either mutant with PNGase F reduces the MW to ~116, the other one-half ofthe way. The N597Q/N617Q double mutant exhibits a pair of bands (MW of116 and 120 in Fig. 6 ) that presumably represent the protein with orwithout glycosylation of N592. However, in the absence of PNGase Ftreatment, the N617Q single mutant exhibited only a single band (MW of122), suggesting that glycosylation at N597 prevents glycosylation at N592.
* h/ b3 {! R! S  W+ ^
By steric hindrance, the bulky carbohydrate moiety at N597 maycontribute to the absence of glycosylation at N592 in wild-type NBCe1.However, steric hindrance is not the entire explanation because, asnoted above, glycosylation at N592 is optional even for single anddouble mutants in which we disrupt glycosylation at N597. Onepossibility is that N592, which is a 28-amino acid residue from theputative end of TM5, is too close to the membrane for efficientglycosylation. However, the N-glycosylation site on the fourthextracellular loop of AE1 is only 15 residues from the putative end ofTM5. Alternatively, folding of the extracellular loop near N592 maymake this asparagine an intrinsically poor substrate for theoligosaccharyl transferase in the endoplasmic reticulum.) D6 x9 j1 b& o6 E1 G+ m
3 X; h: c6 J" b9 n9 k
Retention of function in unglycosylated NBCe1. The Xenopus oocyte expression system has been very usefulfor analyzing the effect of glycosylation on the function of membrane proteins ( 25 ). In most of the membrane proteins studied todate, such as the glutamate transporter GLT1 ( 40 ), the -aminobutyric acid transporter GAT1 ( 26 ), and thenicotinic acetylcholine receptor AChR ( 15 ), the naturallyoccurring N-glycosylation is essential for functional expression of theglycoprotein. Nonetheless, in other cases, such as theNa -dicarboxylate cotransporter NaDC-1 ( 29 ),the Na-Pi cotransporter NaPi-2 ( 20 ), the water channelaquaporin-1 ( 43 ), and the Na channel ENaC( 9 ), the N-glycosylation is not essential for proteinfunction. We found that mutations at one of the two natural sites(N597Q or N617Q), at both of the two natural sites (N597Q/N617Q), or atthe two natural sites plus the one alternative site (N592Q/N597Q/N617Q) yield transporters, as expressed in Xenopus oocytes,functionally comparable to the wild-type NBCe1 in terms ofelectrogenicity as well as Na and HCO 3 − dependence.4 u& p1 D* B- E( L* d
. v' |. I& d5 ?
The unglycosylated AE1 mutant is also functional in the Xenopus expression system ( 30 ). However, theoocytes expressing the unglycosylated AE1 mutant exhibit a lowerexchanger activity than those expressing wild-type AE1, possibly due toincorrect protein folding in the endoplasmic reticulum. It is alsoreported that removing oligosaccharides from AE2 increases thesusceptibility to proteolytic degradation ( 44 ). We cannotrule out the possibility that lack of a natural glycosylation patternin our NBCe1 mutants might have reduced their folding efficiency and/orstability. However, if so, these deleterious effects must have beenovercome by a very high level of expression.! W3 b+ r+ q3 i" M. V
9 z* T+ v) ?& x
ACKNOWLEDGEMENTS
( c# C: x% C4 f5 }; u
We thank D. Wong for computer support.
      【参考文献】- R" s7 x- |. Z2 p- R
1. Abuladze, N,Lee I,Newman D,Hwang J,Boorer K,Pushkin A,andKurtz I. Molecular cloning, chromosomal localization, tissue distribution, and functional expression of the human pancreatic sodium bicarbonate cotransporter. J Biol Chem 273:17689-17695,1998 .
, C- E# k! @" O" s4 x' J# E( z$ R- \
) @1 k* U7 s7 v7 _8 X

2. Abuladze, N,Song M,Pushkin A,Newman D,Lee I,Nicholas S,andKurtz I. Structural organization of the human NBC1 gene: kNBC1 is transcribed from an alternative promoter in intron 3. Gene 251:109-122,2000  .

3. Alper, SL,Kopito RR,Libresco SM,andLodish HF. Cloning and characterization of a murine band 3-related cDNA from kidney and from a lymphoid cell line. J Biol Chem 263:17092-17099,1988 .
# o* ^* i7 N: s) X- L
- Y7 `9 |  a' I/ @/ J' K6 S$ ~2 y
( r! Q6 \+ c* m6 X
4. Alpern, RJ,Warnock DG,andRector FC. Renal acidification mechanism.In: The Kidney, edited by Brenner BM.. Philadelphia, PA: Saunders, 1986, p. 206-249.; s, }% Z; O+ d

3 @$ e0 Y4 t) t& }9 a9 Q* g
& j' N; p" o# W
5. Bevensee, MO,Apkon M,andBoron WF. Intracellular pH regulation in cultured astrocytes from rat hippocampus. II. Electrogenic Na/HCO 3 cotransport. J Gen Physiol 110:467-483,1997 .

5 g/ G) o8 R7 H( _) o, j* c- p
6. Bevensee, MO,Schmitt BM,Choi I,Romero MF,andBoron WF. An electrogenic Na/HCO 3 cotransporter (NBC) with a novel COOH terminus, cloned from rat brain. Am J Physiol Cell Physiol 278:C1200-C1211,2000 .

4 C6 a1 o; {) u3 g* }1 z3 L

7. Boron, WF,andBoulpaep EL. Intracellular pH regulation in the renal proximal tubule of the salamander: basolateral HCO 3 − transport. J Gen Physiol 81:53-94,1983 .; f$ T/ T, T6 |; E
* p. g$ G5 C3 c8 U' s+ M7 }) h

. ~! o7 I( ?: l- v
8. Burnham, CE,Amlal H,Wang Z,Shull GE,andSoleimani M. Cloning and functional expression of a human kidney Na :HCO 3 − cotransporter. J Biol Chem 272:19111-19114,1997 .

% d* b& Q% n2 c- w) D' O6 l0 y2 t9 z

9. Canessa, CM,Merillat AM,andRossier BC. Membrane topology of the epithelial sodium channel in intact cells. Am J Physiol Cell Physiol 267:C1682-C1690,1994 .

/ G! K" S1 R) ^
1 `& C+ l4 G+ I9 V4 l( \1 b
10. Choi, I,Romero MF,Khandoudi N,Bril A,andBoron WF. Cloning and characterization of a human electrogenic Na -HCO 3 − cotransporter isoform (hhNBC). Am J Physiol Cell Physiol 276:C576-C584,1999 .

11. Deitmer, JW,andSchlue WR. The regulation of intracellular pH by identified glial cells and neurones in the central nervous system of the leech. J Physiol 388:261-283,1987 ./ P5 E! T' w5 H8 L1 u4 `# ]
& R( ~" d) g! F

12. Demuth, DR,Showe LC,Ballantine M,Palumbo A,Frzser PJ,Cioe L,Rovera G,andCurtis PJ. Cloning and structural characterization of human non-erythroid band 3-like protein. EMBO J 5:1205-1214,1986  .
0 B: x% O8 [0 r3 d4 @; H" ^6 P9 J
9 q) E, _# j; |1 w) c9 k) ]2 r

13. Elhammer, AP,Kezdy FJ,andKurosaka A. The acceptor specificity of UDP-GalNAc:polypeptide N- acetylgalactosaminyltransferases. Glycoconj J 16:171-180,1999  .
& v& c' ]& _( S- J

% I: x: |( [- j3 c( r' d/ s# t
14. Fujinaga, J,Tang XB,andCasey JR. Topology of the membrane domain of human erythrocyte anion exchange protein, AE1. J Biol Chem 274:6626-6633,1999 ." T, _" e3 Y- z# K

  \' z& \3 D% d: I& `7 i
. i+ F! C! f* ~% J
15. Gehle, VM,Walcott EC,Nishizaki T,andSumikawa K. N-glycosylation at the conserved sites ensures the expression of properly folded functional ACh receptors. Brain Res Mol Brain Res 45:219-229,1997 .
8 l# R% X) D- M2 {9 f; S3 }
& P" d( R8 c( I4 `9 R3 `9 F: b

16. Grichtchenko, II,andChesler M. Depolarization-induced acid secretion in gliotic hippocampal slices. Neuroscience 62:1057-1070,1994  .
: O$ L: r, q/ K- z9 N

8 Y# M! X3 d, |% Y
17. Grichtchenko, II,Choi I,Zhong X,Bray-Ward P,Russell JM,andBoron WF. Cloning, characterization, and chromosomal mapping of a human electroneutral Na -driven Cl-HCO 3 exchanger. J Biol Chem 276:8358-8363,2001 .  U9 t2 b7 E) u6 `4 o; [
- P8 ?, C. i1 ~' R* T) f

18. Grichtchenko, II,Romero MF,andBoron WF. Extracellular HCO 3 − dependence of electrogenic Na/HCO 3 cotransporters cloned from salamander and rat kidney. J Gen Physiol 115:533-545,2000 .4 K; X6 M0 q) ~+ _
- t- S# H1 `6 v0 L
0 ^5 W& C2 A. b. }% S' O& D0 x

19. Groves, JD,andTanner MJ. Role of N-glycosylation in the expression of human band 3-mediated anion transport. Mol Membr Biol 11:31-38,1994  .

$ i: F' I: p' i9 X3 \6 `
9 g' Q; J5 p" e, |7 F- [
20. Hayes, G,Busch A,Lotscher M,Waldegger S,Lang F,Verrey F,Biber J,andMurer H. Role of N-linked glycosylation in rat renal Na/Pi-cotransport. J Biol Chem 269:24143-24149,1994 ., x8 b  Q3 w! P+ U0 M' k! k9 d  u

. Q6 m4 ]3 h2 U& F9 ]! t0 D

21. Isiguro, H,Steward MC,Wilson RW,andCase RM. Bicarbonate secretion in interlobular ducts from guinea pig pancreas. J Physiol 495:179-191,1996  .
4 B7 s5 l% h3 M# `

22. Jensen, L,Schmitt BM,Brown D,Berger UV,Hediger MA,Boron WF,andBreton S. Localization of sodium bicarbonate co-transporter (NBC) protein and mRNA in rat epididymis. Biol Reprod 60:573-579,1999 .

' ^1 w" i4 g. y% Y; s
+ R/ N6 N% x+ I
23. Kopito, RR,andLodish HF. Primary structure and transmembrane orientation of the murine anion exchange protein. Nature 316:234-238,1985 .

24. Landolt-Marticorena, C,Charuk JH,andReithmeier RA. Two glycoprotein populations of band 3 dimers are present in human erythrocytes. Mol Membr Biol 15:153-158,1998  .

% z3 N9 q, C( z1 F

25. Li, P,Gao XG,Arellano RO,andRenugopalakrishnan V. Glycosylated and phosphorylated proteins-expression in yeast and oocytes of Xenopus : prospects and challenges-relevance to expression of thermostable proteins. Protein Expr Purif 22:369-380,2001  .# i$ I1 Z' i) N+ Y
8 e  \, S3 w3 Y5 \) ]
9 X  A7 j4 C2 J, M7 l; d) c' n

26. Liu, Y,Eckstein-Ludwig U,Fei J,andSchwarz W. Effect of mutation of glycosylation sites on the Na dependence of steady-state and transient currents generated by the neuronal GABA transporter. Biochim Biophys Acta 1415:246-254,1998 .% t/ z. G& Y% Y; G& v

) j3 J: f5 O5 T( e) |6 M! u
3 I6 T/ Z/ O6 Y- V/ H
27. Marino, CR,Jeans V,Boron WF,andSchmitt BM. Expression and distribution of the Na -HCO 3 − cotransporter in human pancreas. Am J Physiol Gastrointest Liver Physiol 277:G487-G494,1999 .
  q$ O. w& K$ k  n
$ `% p, O& @3 |' w( a  Z

28. Maunsbach, AB,Vorum H,Kwon TH,Nielsen S,Simonsen B,Choi I,Schmitt BM,Boron WF,andAalkjaer C. Immunoelectron microscopic localization of the electrogenic Na/HCO 3 cotransporter in rat and ambystoma kidney. J Am Soc Nephrol 11:2179-2189,2000 ." {( J8 N# W$ @; j6 t8 G
4 y& Q! |& j8 d5 I* x
3 o( g8 x* }9 e0 ]$ G7 R7 J

29. Pajor, AM,andSun N. Characterization of the rabbit renal Na -dicarboxylate cotransporter using antifusion protein antibodies. Am J Physiol Cell Physiol 271:C1808-C1816,1996 ./ l' k( D5 d9 S% C) z6 S

30. Pineda, J,andAghajanian GK. Carbon dioxide regulates the tonic activity of locus coeruleus neurons by modulating a proton- and polyamine-sensitive inward rectifier potassium current. Neuroscience 77:723-743,1997  .0 ]" f5 D2 d# N; j% j. N

$ m1 T3 I" n. i2 }1 U! |
1 ?% ~  o9 ?) x6 S
31. Popov, M,andReithmeier RA. Calnexin interaction with N-glycosylation mutants of a polytopic membrane glycoprotein, the human erythrocyte anion exchanger 1 (band 3). J Biol Chem 274:17635-17642,1999 .
4 j8 A& _0 s/ o% o5 ^2 v

0 Z0 }$ B* V6 w; M& k' N( {$ g
32. Popov, M,Tam LY,Li J,andReithmeier RA. Mapping the ends of transmembrane segments in a polytopic membrane protein. Scanning N-glycosylation mutagenesis of extracytosolic loops in the anion exchanger, band 3. J Biol Chem 272:18325-18332,1997 .
3 p, }, R! z& Q0 l' ~

33. Pusch, M,Steinmeyer K,Koch MC,andJentsch TJ. Mutations in dominant human myotonia congenita drastically alter the voltage dependence of the CIC-1 chloride channel. Neuron 15:1455-1463,1995  .# k. l9 C$ S" X7 R, i  Y

34. Romero, MF,Fong P,Berger UV,Hediger MA,andBoron WF. Cloning and functional expression of rNBC, an electrogenic Na -HCO 3 − cotransporter from rat kidney. Am J Physiol Renal Physiol 274:F425-F432,1998 .# d' c! r' f. @9 |# Q  F: ~3 F" P
& y* C/ b7 @& B3 d& J" g5 z
+ q, S  @, S) ]) ]! h+ Y5 b

35. Romero, MF,Hediger MA,Boulpaep EL,andBoron WF. Expression cloning and characterization of a renal electrogenic Na /HCO 3 − cotransporter. Nature 387:409-413,1997 .
; N( v- c, ^* W- G7 E4 u

36. Roussa, E,Romero MF,Schmitt BM,Boron WF,Alper SL,andThevenod F. Immunolocalization of anion exchanger AE2 and Na -HCO 3 − cotransporter in rat parotid and submandibular glands. Am J Physiol Gastrointest Liver Physiol 277:G1288-G1296,1999 .

' r4 L+ H3 F/ H, a/ e3 J0 p
) x$ h6 }& p. \1 C# `3 E& Q1 d
37. Schmitt, BM,Berger UV,Douglas R,Bevensee MO,Hediger MA,Haddad GG,andBoron WF. Na/HCO 3 cotransporters in rat brain: expression in glia, neurons and choroid plexus. J Neurosci 20:6839-6848,2000 .

! h% O) j  \8 L# z: X$ w

38. Schmitt, BM,Biemesderfer D,Romero MF,Boulpaep EL,andBoron WF. Immunolocalization of the electrogenic Na /HCO 3 − cotransporter in mammalian and amphibian kidney. Am J Physiol Renal Physiol 276:F27-F36,1999 .9 W1 W0 \& a  N) z1 p# Y

3 g9 L- H+ S! k& b+ r: @
39. Sciortino, CM,andRomero MF. Cation and voltage dependence of rat kidney electrogenic Na -HCO 3 − cotransporter, rkNBC, expressed in oocytes. Am J Physiol Renal Physiol 277:F611-F623,1999 .+ s$ b6 \" `' N+ _; N6 p

6 l$ r" e; V$ m- Z" a
40. Trotti, D,Aoki M,Pasinelli P,Berger UV,Danbolt NC,Brown RH, Jr,andHediger MA. Amyotrophic lateral sclerosis-linked glutamate transporter mutant has impaired glutamate clearance capacity. J Biol Chem 276:576-582,2001 .
: C! J, M: v* V9 I8 B  c6 N/ K! o1 s- D
0 N& R% t" |. ~( K6 c/ i8 D6 p7 b

41. Virkki, LV,Wilson DA,Vaughan-Jones RD,andBoron WF. Functional characterization of human NBC4 as an electrogenic Na -HCO 3 cotransporter (NBCe2). Am J Physiol Cell Physiol 282:C1278-C1289,2002 .1 P; ]0 Z) b. u9 z4 g" h) [
$ T$ X" R3 ]/ r3 D/ Q( N
. C+ m* e& P' i4 ?; {0 _4 N

42. Vorum, H,Kwon TH,Fulton C,Simonsen B,Choi I,Boron W,Maunsbach AB,Nielsen S,andAalkjaer C. Immunolocalization of electroneutral Na-HCO 3 − cotransporter in rat kidney. Am J Physiol Renal Physiol 279:F901-F909,2000 .. W) i+ x0 }' b2 X1 S' A6 C

0 v5 A* B$ a! v
43. Zhang, R,Van Hoek AN,Biwersi J,andVerkman AS. A point mutation at cysteine 189 blocks the water permeability of rat kidney water channel CHIP28k. Biochemistry 32:2938-2941,1993 .. l& _# u' c, b0 Q6 b* M; a

$ @* P& o/ ?& U; N8 [4 c+ D
44. Zolotarev, AS,Townsend RR,Stuart-Tilley A,andAlper SL. HCO 3 − -dependent conformational change in gastric parietal cell AE2, a glycoprotein naturally lacking sialic acid. Am J Physiol Gastrointest Liver Physiol 271:G311-G321,1996 .

收藏0 分享0 顶0 踩0