Data & Analytics

A new variant of the gamma subunit of renal Na,K-ATPase. Identification by mass spectrometry, antibody binding, and expression in cultured cells

A new variant of the gamma subunit of renal Na,K-ATPase. Identification by mass spectrometry, antibody binding, and expression in cultured cells
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
   A New Variant of the     Subunit of Renal Na,K-ATPase IDENTIFICATION BY MASS SPECTROMETRY, ANTIBODY BINDING, AND EXPRESSIONIN CULTURED CELLS* Received for publication, February 21, 2000, and in revised form, March 14, 2000Published, JBC Papers in Press, March 30, 2000, DOI 10.1074/jbc.M001411200 Bernhard Ku¨ster‡§, Alla Shainskaya ¶ , Helen X. Pu  , Rivka Goldshleger**, Rhoda Blostein  ,Matthias Mann‡, and Steven J. D. Karlish**‡‡  From the  ‡  Protein Interaction Laboratory, University of Southern Denmark, Campusvej 55, 5230 Odense M, Denmark,the    Department of Medicine, McGill University, Montreal, Quebec H3G 1A4, Canada, and the Departments of   ¶  Biological Services and **Biological Chemistry, Weizmann Institute of Science, Rehovoth 76100, Israel The     subunit is a specific regulator of Na,K-ATPaseexpressed mainly in kidney. On SDS-polyacryylamidegel electrophoresis,     runs as a doublet, but the srcinand significance of the doublet is obscure. Mass spec-trometry of the     chains of rat kidney Na,K-ATPaseshows that    a  (upper) has a mass of 7184.0    1 Da (car-bamidomethyl cysteine), corresponding closely to thatfor the published sequence without the initiator methi-onine, while    b  (lower) has a mass of 7337.9  1Da. Tryp-tic peptide mapping and sequencing by mass spectrom-etry reveals that the seven N-terminal residues of     a ,TELSANH, are replaced by Ac-MDRWYL in    b , but oth-erwise the chains are identical. Antibodies raisedagainst peptides TELSANHC and MDRWYLC recognizeeither    a  or    b  of the Na,K-ATPase, respectively.    a  or    b cDNAs have been expressed in human embryonic kid-ney and HeLa cells. The major bands expressed corre-spondto   a or   b ofrenalNa,K-ATPase.Additionalminorbands seen after transfection, namely    a   in human em-bryonic kidney and    b   in HeLa, are presumably cell-specific modifications. The present work clarifies ear-lier uncertainty regarding doublets seen in kidney andin transfected cells. In particular, the results show thatrenal Na,K-ATPase contains two variants of the     sub-unit with different sequences but otherwise are unmod-ified. We discuss the possible functional significance of the two variants. The     subunit of the Na,K-ATPase is a small, single trans-membrane protein (mass  7 kDa) expressed primarily in renaltissue,inapproximatelyequimolaramountscomparedwiththe   and    subunits (1–6). It has substantial homology to CHIF(corticosteroid-induced factor) (7), mat 8 (8), and phospholem-man (9), which have similar trans-membrane organization (Cterminus in, N terminus out) and appear to function as ionchannel regulators. A comparison of sequences shows that    subunits of different species are approximately 75% homolo-gous. If only mammalian sequences are compared, the homol-ogy increases to 93%.Several findings imply a stable association of      with    and   subunits in tissues expressing all three subunits. The expres-sion patterns of the catalytic    and     subunits are identical inrenal proximal tubules and collecting ducts (5), and the    subunit is expressed at the surface of   Xenopus  oocytes only inthe presence of     and    subunits (10). In addition, co-immuno-precipitation of the     subunit with both the    and    subunitshas been demonstrated in kidney membranes (5), and the    subunit is also co-solubilized with fragments of     and    sub-units in a complex obtained from extensively digested pig kid-ney Na,K-ATPase (11). On the other hand, Jones  et al.  (12)have reported expression of      also in the absence of the sodiumpump on the apical surface of mouse blastocysts.The functional role of the     subunit is being actively studiedboth in intact renal membranes utilizing antibodies to counter-act its effects (6, 13) and after expression in mammalian cells(13, 14), insect cells (15), or  Xenopus  oocytes (10). A number of functional effects have been described, which may reflect one ormore interactions of the     subunit with the    /    subunits of theNa,K-ATPase. Experiments with intact renal Na,K-ATPaseand mammalian cell membranes containing the expressed    subunit show that it increases the affinity for ATP either as aprimary effect or, more likely, as a consequence of a shift inconformational equilibrium in favor of E 1  form(s) (6, 13). Con-sistent with these effects, the apparent affinity for K   is de-creased under conditions of suboptimal ATP concentration (16).There is a recent report that the     subunit expressed in mam-malian kidney cells decreases the apparent cytoplasmic sodiumaffinity (14). In addition, experiments on cRNA-injected  Xeno- pus  oocytes have shown that the     subunit has an influence onthe apparent affinity of the Na,K-ATPase for extracellular K   in a complex Na  - and voltage-dependent fashion (10). In an-other report, the human     subunit was shown to induce non-selective ouabain-independent ion currents in injected  Xenopus oocytes, and  86 Rb and  22 Na influx in baculovirus-infected Sf-9cells (15).On SDS-PAGE 1 of renal membranes isolated from all speciesexamined to date, the     subunit runs as a doublet (apparentmass,   8 and   9 kDa) (5, 6), and a doublet has also been * This work was supported in part by a long term postdoctoral fel-lowship of the European Molecular Biology Organization (to B. K.), bya short term FEBS fellowship (to A. S.), by a grant from the DanishFoundation for Fundamental Research (to M. M.’s laboratory at theCenter for Experimental BioInformatics), by a grant from the Weiz-mann Institute Renal Research Fund (to S. J. D. K.), and by a grantfrom the Medical Research Council of Canada (to R. B.). The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked “ advertisement ”in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The nucleotide sequence(s) reported in this paper has been submittedto the GenBank TM /EBI Data Bank with accession number(s) AF129400. § Current address: MDS Protana A/S, 5230 Odense M, Denmark.‡‡ To whom correspondence should be addressed. Tel.: 972-8-934-2278; Fax: 972-8-934-4118; E-mail: 1 The abbreviations used are: PAGE, polyacrylamide gel electro-phoresis; MS, mass spectrometry; MALDI, matrix-assisted laser de-sorption/ionization; HEK, human embryonic kidney; PCR, polymerasechain reaction; EST, expressed sequence tag; Tricine,  N  -tris(hydroxymethyl)methylglycine. T HE  J OURNAL OF  B IOLOGICAL  C HEMISTRY   Vol. 275, No. 24, Issue of June 16, pp. 18441–18446, 2000 © 2000 by The American Society for Biochemistry and Molecular Biology, Inc.  Printed in U.S.A. This paper is available on line at  18441   a t  W ei  z m annI  n s  t  i   t   u t   e of   S  c i   en c  e on S  e p t   em b  er  8  ,2  0  0  8 www. j   b  c . or  gD  ownl   o a d  e d f  r  om  observed upon expression of      in tissue culture cells (13, 14)and in  in vitro  expression systems in the presence of pancreaticmicrosomes (5) but not in their absence (10). A post-transla-tional modification might be suggested on the basis of thefindings, while other possibilities include the presence of sep-arate isoforms or splice variants. In view of the largely tissue-specific expression of the     subunit in kidney and the newevidenceforaregulatoryrole,itbecomesimportanttoestablishthe srcin of the doublet of bands in renal membranes and theirpossible association with the diverse functional effects on renalNa,K-ATPase that have been described recently. This paperdescribes the results of experiments that define the structuraldifference between the two bands as the result of differentprimary sequences. MATERIALS AND METHODS Na,K-ATPase was purified from Milan hypertensive rat kidneys bythe method of Jorgensen (17). The specific activity was in the range16–30   mol of P i  /mg/min.The rat renal enzyme was reduced and alky-lated with iodoacetamide, and the two     subunits were separated on a10% Tricine SDS gel (18) and stained with Coomassie Brilliant Blue.The two bands were excised from the gel and destained using multiplewashings with 50% acetonitrile in 50 m M  ammonium bicarbonate. The   subunits were extracted from the gel pieces by overnight incubation ina mixture of 1:2:3 (v/v/v) formic acid, isopropyl alcohol, and water atroom temperature (19).  Mass Spectrometry  Intact Molecular Weight Measurements— Extracted     subunits werefurther purified prior to MS analysis. Briefly, the dried extract from onelane of the gel (2–3   g of      subunit) was redissolved in 1   l of 80%formic acid and immediately diluted with water to yield a final concen-tration of 5% formic acid. This preparation was passed over a microcol-umn consisting of about 300 nl of Poros R1 reversed phase material(Perseptive Biosystems, Framingham, MA) packed in an Eppendorf GelLoader pipette tip (20), and the purified protein was eluted directly intoa nano-electrospray capillary (Protana A/S, Odense, Denmark) using 1  l of 70% acetonitrile, 5% formic acid. Protein mass spectra wereacquired using a nano-electrospray ion source (Protana A/S) (21) on an API 300 triple quadrupole mass spectrometer (PE Sciex, Concord, On-tario, Canada).  In Gel Digestion, Peptide Mass Mapping, and Peptide Sequenc-ing— In gel tryptic digestion of the two separated     bands was per-formed as described (22). Briefly, protein was excised from the gel(usually one lane containing 2–3   g of      subunit), fully destained,reduced, carbamidomethylated, and digested overnight with bovinetrypsin (sequencing grade, Roche Diagnostics, Mannheim, Germany) ata concentration of 12.5 ng/   l in 50 m M  ammonium bicarbonate at 37 °C.Peptide mass mapping was performed on a Bruker Reflex III matrix-assisted laser desorption/ionization (MALDI) time-of-flight mass spec-trometer (Bruker, Bremen, Germany) equipped with a 337 nm nitrogenlaser. Matrix surfaces were made from   -cyano-4-hydroxycinnamic acidby the fast evaporation method (23, 24). About 1–2% (0.3–0.5   l) of thesupernatant of in gel trypsin digests were injected into an acidified droppreviously deposited onto the matrix surface, and the sample wasallowed to dry at ambient temperature before MS analysis. Prior topeptide sequencing by tandem mass spectrometry, the peptide mixturewas extracted from the gel by two changes of 5% formic acid, followed by100% acetonitrile. The combined extracts were dried  in vacuo . Thedried peptides were redissolved in 5% formic acid and purified using amicrocolumn consisting of approximately 300 nl of Poros R2 reversephase material (Perseptive Biosystems) (22). Peptides were eluted with60% methonol/5% formic acid directly into a nano-electrospray capillaryand sequencing was performed by nano-electrospray tandem mass spec-trometry on a prototype quadrupole time-of-flight mass spectrometer(PE Sciex). For sequencing peptides from rat    b , the in gel trypticdigestion was also performed in the presence of 50%  18 O-labeled water(25).  Expression Experiments Cloning of Rat    a  and    b  cDNAs— For transfection into HEK cells,the cDNA for    a  was obtained as described previously (13). The cDNA for    b  was obtained by PCR using CLONTECH Marathon-ready cDNA from rat kidney as template, with primers designed according to theN-terminal    b  protein sequence obtained by mass spectrometry. Theforward primer was GGGGGGGAAGCTTGCCGCCACCATGGACAGA-TGGTATCTTGGTGGCAGT containing a  Hin dIII restriction site, andthe reverse primer was GGGGAAGATCCGTCACAGCTCATCTTCATT-GACCT containing a  Bam HI restriction site. The resulting cDNA wasthen cleaved by these two endonucleases and ligated into the pREP4expression vector (CLONTECH). Positive clones were verified by nu-cleotide sequencing. pREP4, pREP-   a , and pREP4-   b  were purifiedusing the Concert   high purity plasmid purification system (Life Tech-nologies, Inc.). For transfection into HeLa cells, PCR was carried out asabove, except the following primers containing restriction sites for  Bam HI (forward primers) and  Bst  XI (reverse primer), respectively,were used. The forward primer for    a  was GGGGGGGGATCCGCCGC-CACCATGACAGAGCTGTCAGCTAAC, and for    b , GGGGGGGGATC-CGCCGCCACCATGGACAGATGGTATCTTGGTGGCAGT. The re- verse primer for both    a  and    b  was GGGGGGCCAGCACACTGGGTC- ACAGCTCATCTTCATTGACCTGCCTAT. The resulting DNAs werethen cleaved, ligated into the pIRES expression vector (CLONTECH),and then verified and purified as described above. Transfection, Tissue Culture, and Membrane Preparation— The pro-cedures for transfecting HEK cells were as described elsewhere (13),except that the LipofectAMINE Plus   reagent (Life Technologies, Inc.)was used as directed by the manufacturer. HeLa cells (50% confluent)were similarly transfected, and single clones were selected after 3weeks’ growth in Dulbecco’s modified Eagle’s medium containing 10%newborn calf serum and 400   g/ml hygromycin B. Western Blot Analysis Western blot analysis on HEK or HeLa cell membranes was carriedout as described previously (6). Briefly, 5   g of cell membranes wereanalyzed on 10% SDS-PAGE, after which the proteins were transferredto a polyvinylidene difluoride membrane (Millipore) and probed withanti-  1 (6H antibodies from Dr. Michael Caplan) and anti-    antiserum(   C33 polyclonal antiserum raised against the 10-residue C terminus of    a ), both at 1:2000 dilution. Western blots of rat kidney Na,K-ATPaseutilized either the    C33 polyclonal antiserum or specific anti-   a  andanti-   b  antisera (1:100 dilution).    a -specific or    b -specific antibodieswere raised in rabbits utilizing synthetic peptides TELSANHC (   a ) orMDRWYLC (   b ), coupled to maleimide-activated keyhole limpet hemo-cyanin as immunogens (Biological Services, Weizmann Institute, Reho- voth, Israel). RESULTS  Intact Protein Mass Determination—  As a first step towardestablishing the molecular basis for the observed difference inthe migration behavior of     a  and    b  on SDS-PAGE, the sepa-rated, intact proteins were extracted from the gel and subjectedto molecular weight determination by mass spectrometry. Theresults of the mass measurements for the intact proteins areillustrated in Fig. 1. The measured mass for the    a  subunit(7184.0    1 Da) is in excellent agreement with the mass de-duced from the published amino acid sequence of rat kidney    a without the initiator methionine (7183.1 Da, calculated forcarbamidomethyl cysteine) (see GenBank    accession no. AF129400). This result implies that the protein is not modified. F IG . 1.  Intact molecular mass determination of rat    a  and    b . Top panel , raw nano-electrospray mass spectra of rat    a  and    b  showing a series of multiply charged species of the two proteins.  Bottom panel ,deconvoluted mass spectra showing the neutral mass of the proteinstogether with oxidized species as calculated from the respective spectrain the  upper panel .  A Variant of the     Subunit of Na,K-ATPase 18442   a t  W ei  z m annI  n s  t  i   t   u t   e of   S  c i   en c  e on S  e p t   em b  er  8  ,2  0  0  8 www. j   b  c . or  gD  ownl   o a d  e d f  r  om  Satellite peaks indicative of oxidation of the protein during extraction (  16 Da) were also observed. For the    b  subunit, anumber of species were detected at 7337.9    1, 7353.8    1,7369.9    1, and 7385.9    1 Da, respectively. From the massincrements of 16 Da between the measured signals, multipleoxidation of the protein during extraction is apparent. Themeasured mass differences between    b  and the published se-quence of     a  (154, 170, 186, and 202 Da respectively) indicateeither the presence of post-translational modifications or se-quence variations. In this regard, it is interesting to note that,despite its higher molecular weight, the    b  subunit actuallymigrates faster during SDS-PAGE compared with    a .  Peptide Mapping of     a  and    b  Subunits— In order to deter-mine the region of the protein in which the observed massdiscrepancy between    a  and    b  resides, the proteins were sub- jected to in gel proteolytic cleavage and the resulting peptidemixture was mapped using MALDI MS. The mass difference of 154 Da between the    a  and    b  subunits determined from theintact mass measurements for both subunits, together with thecorresponding oxidized species, was also detected at the pep-tide level (Fig. 2). The prominent signal for the N-terminalpeptide of     a  (observed at a mass to charge ratio of   m/z  1171.61in Fig. 2  A ), was not present in the MALDI peptide mass map of    b  (Fig. 2  B ). Instead, a prominent signal at  m/z  1325.62 wasobserved together with multiple oxidized species (  16,   32,and   48 Da) of the same peptide. Hence, it can be concludedthat the modification or sequence variation maps to this pep-tide. The signals at  m/z  1719 and 1847 in the MALDI MSspectra of both gamma subunits correspond to the tryptic pep-tides GTENPFEYDYETVR and GTENPFEYDYETVRK (resi-dues 14–27/28) of the published rat     sequence. No significantdifference other than the one described was observed betweenthe peptide maps of the two subunits.  Peptide Sequencing by Tandem Mass Spectrometry— Thepeptides responsible for the mass difference between    a  and    b were subjected to sequencing by nano-electrospray tandemmass spectrometry (Fig. 3). For the peptide at  m/z  1171.61 inthe MALDI peptide mass map of     a , the sequence TELSANH-GGSAK was determined from the C-terminal Y    ion fragmention series (26) (Fig. 3  A ). This sequence corresponds exactly tothe N-terminal tryptic fragment of the sequence reported in theliterature (6, 15) without the initiator methionine. This resultconfirmed the conclusion that    a  is not modified as deducedalso from the mass of the intact    a . The corresponding N-terminal tryptic peptide of     b  ( m/z  1325.62 in Fig. 2  B ) was alsosequenced by tandem mass spectrometry. The interpretation of the spectrum shown in Fig. 3  B  resulted in the determination of the sequence Ac-MDRWY(I/L)GGSAK (note that I/L cannot bedistinguished by mass spectrometric sequencing as the tworesidues are isobaric). The sequence determination was con-firmed by performing a second experiment in which the trypticdigest was carried out in the presence of 50%  18 O-labeledwater. Partial  18 O labeling of peptides during trypsin digestionallows the determination of the direction in which the sequenceis read from the tandem mass spectrum as only those peptidefragments that carry the C terminus of the peptide are ob-served as a doublet with 2-Da spacing (see Fig. 3  B ,  inset ) (25). F IG . 2.  Peptide mass mapping of     a  and    b .  The MALDI MSspectra of     a  (  A ) and    b  (  B ) after in gel trypsin digestion reflect the massdifference of 154 Da observed for the intact proteins. The prominentpeptide at  m/z  1171.61 of     a  in  A  is replaced by a peptide ion signal at m/z  1325.62 in    b  (  B ) together with a series of oxidized species of thesame peptide (  16,  32, and  48 Da, respectively).F IG . 3.  Peptide sequencing by nano-electrospray tandem MS.  A , sequencing of the doubly protonated    a  peptide (M    2H) 2  corre-sponding to the peptide detected at  m/z  1171.61 in the MALDI peptidemass map (Fig. 2  A ) revealed the sequence TELSANHGGSAK, which isidentical to the N terminus of the published     sequence without theinitiator methionine residue.  B ,  de novo  sequencing of the N-terminalpeptide of     b  ((M  2H) 2  corresponding to the peptide detected at  m/z 1325.62 in the MALDI peptide mass map; Fig. 2  B ) resulted in thesequence Ac-MDRWYLGGSAK. Sequences were derived from the C-terminal (Y   n ) and N-terminal (a n  and b n ) fragment ion series (nomen-clature as in Ref. 26). The  inset  shows partial  18 O labeling of C-terminal(Y   ) fragment ions (doublet with 2-Da spacing) as the result of perform-ing the in gel trypsin digestion in the presence of 50%  18 O-labeledwater. N-terminal fragment ions (labeled  a n  and  b n  in the spectrum) donot contain the  18 O label, as indicated by their unaltered isotopicdistribution.  A Variant of the     Subunit of Na,K-ATPase  18443   a t  W ei  z m annI  n s  t  i   t   u t   e of   S  c i   en c  e on S  e p t   em b  er  8  ,2  0  0  8 www. j   b  c . or  gD  ownl   o a d  e d f  r  om  Comparison of the deduced sequence for    b  with that of     a shows that the 7 N-terminal amino acids of     a  TELSANH arereplaced by Ac-MDRWY(I/L) in   b . These two partial sequencesalso differ by 154 Da. The calculated intact mass of the revisedsequence of     b  (7337.4 Da,  N  -acetylated, carbamidomethyl cys-teine) is in excellent agreement with the measured value of 7337.9    1 Da. Further sequencing experiments also revealedpeptides corresponding to the oxidized species Ac-MoxDRWY(I/ L)GGSAK, Ac-M ox DRW ox  Y(I/L)GGSAK, and Ac-M ox DRW ox-ox  Y(I/L)GGSAK, thus explaining the corresponding signals ob-served in the MALDI MS peptide mass map and the spectrumof the intact protein. An EST data base search reported in Ref.27 has revealed sequences MDRWYL.., indicating that L andnot I is the correct assignment in position 6.  Sequencing by Edman Degradation— The absence of an ini-tiator methionine in    a  suggested that it would be amenable todirect Edman degradation. Sequencing indeed confirmed theN-terminal sequence TELSANHGGSA (data not shown). Nosequence for    b  was obtained by Edman degradation, consist-ent with its blocked N terminus.  Recognition of Anti-   a - and Anti-   b -specific Antibodies— Thepeptides TELSANHC and MDRWYLC, unique to the    a  and    b sequences, respectively, were synthesized and coupled to ma-leimide-activated keyhole limpet hemocyanin, which was usedto immunize rabbits. The blots of rat kidney Na,K-ATPase inFig. 4 show that anti-   a  and anti-   b  sera recognize the upper orlower bands of the     subunit, respectively, but no other bandsin their vicinity. This result is consistent with that of the MSanalysis, which detected only the two variants and no othermodified species of the     subunit in intact rat kidney enzyme.  Expression of     a  and    b  in Cultured Mammalian Cells— Weshowed previously that the full-length rat    a  cDNA encodes aprotein that appears as a doublet on Western blots of mem-branes isolated from transfected HEK cells, using antibodiesthat recognize the C terminus of the protein (13). The mobilityof the denser upper band corresponded to that of the upperband of the rat kidney     doublet; that of the lower band,designated   a  ,  appeared to be similar to that of kidney   b . Thatlatter possibility is difficult to reconcile with the present MSanalysis.Accordingly,weanalyzedtheproteinproductofcDNA encoding the lower    b  band of kidney. The cDNA for    b  wasobtained by PCR as described under “Materials and Methods,”with primers designed according to the protein sequence ob-tained by mass spectrometry (Fig. 5). The results of Westernblot analysis of rat kidney, control pREP4, and both pREP-   a -and pREP4-   b -transfected HEK cell membranes confirm ourprevious findings for    a  and indicate clearly that the cDNA for   b  encodes a protein of the same mobility as the lower bandseen in renal tissue. This result provides unequivocal evidencethat the lower    b  band detected in renal tissue is distinct fromthe lower band    a   band seen with    a -transfected cells. Bandscorresponding to    a  and    b  of kidney were also evident follow-ing cDNA transfection of HeLa cells, which in contrast to tran-siently stable HEK (see Ref. 13) are a classical stable trans-fected cell line. By comparing the results with the two differentcell lines (Fig. 5), it is evident that there are additional minorbands. Their appearance is clearly cell-specific. Thus,    a   de-scribedaboveisapparentinHEKbutnotHeLa;abandoflowermobility than that of     b , designated    b  , is apparent in HeLabut not HEK.The basis for the cell-specific appearance of     a   and    b   re-main unclear. In other experiments (data not shown), we ob-served that the faster mobility    a   species seen in    a -trans-fected HEK cells is not due to (rare) usage of the alternativetranslation initiation codon CTG (16), nor does the appearanceof a doublet reflect post-translational phosphorylation of cyto-plasmic serines since    a   appears regardless of whether (i) theCTG codon is mutated (CTG  3   CTC), or (ii) either or bothcytoplasmic serines (conserved Ser 47 ; non-conserved Ser 55 ) aremutated (Ser 3   Ala). DISCUSSION The presence of the two bands of the     subunit has long beenan intriguing finding with no obvious explanation. In view of the recent evidence for a tissue (mainly kidney)-specific role of the     subunit as a modulator of the Na,K-ATPase, it has be-come important to gain insight into the structural basis for thetwo     bands. Furthermore, it now becomes necessary to estab-lish whether one or other band is associated with particularfunctional effects already described (6, 10, 13–15) or yet to befound.The MS results (Figs. 1–3) show unequivocally that   a  and   b of the intact rat kidney Na,K-ATPase differ in their N-terminalsequences, TELSANH compared with Ac-MDRWY(I/L), but areotherwise identical. No post-translational modifications on ei-ther chain have been detected by MS except for the N-terminalacetylation of     b . Different sequences might indicate the exist-ence of separate isoforms, but, from an EST data base search,it has been inferred recently that there are two splice variants,with the same sequences as determined directly by MS(MDRWYL for    b ) (27). Given the extensive similarity betweenthe two forms, the present MS analysis is consistent with thenotion that    a  and    b  are splice variants. Of course, thestrength of MS, in addition to the direct determination of sequence, is that it demonstrates directly that both variantsare expressed as protein in the intact renal membranes, as wellas indicating the absence of post-translational modifications. A search of the EST data base indicates the presence of mRNAs F IG . 4.  Binding of     a - or    b -specific antibodies to rat kidneyNa,K-ATPase.  The blot was probed with the anti-   , anti-   a , or anti-   b antibodies as indicated and described under “Materials and Methods.”10   g (lane probed with anti-   ) or 20   g (lanes probed with anti-   a  andanti-   b ) of delipidated rat kidney Na,K-ATPase was applied to the gel.F IG . 5.  Western blot analysis of membranes derived from    a -and    b -transfected HEK and HeLa cell membranes.  Antibodiesused were monoclonal antibodies 6H for   1 and polyclonal antiserumgC33 for    .  A , immunoblot showing 5   g of membranes isolated fromHEK cells transfected with    a  (   a -TF  ),    b (   b -TF  ), the control pREP4 vector ( Con-TF  ), and 1   g of rat kidney membranes ( kidney ).  B , immu-noblot of membranes derived from transfected HeLa cells. Amountsanalyzed are the same as in  A . The lower “extra” band seen in HEK-   a -TF and the upper “extra” band seen in HeLa-   b -TF are named    a  and    b  , respectively.  A Variant of the     Subunit of Na,K-ATPase 18444   a t  W ei  z m annI  n s  t  i   t   u t   e of   S  c i   en c  e on S  e p t   em b  er  8  ,2  0  0  8 www. j   b  c . or  gD  ownl   o a d  e d f  r  om  in the tissues of interest but not necessarily that protein isexpressed in those tissues or the level of expression. This paperrepresents the first demonstration of application of MS to anal-ysis of this class of single trans-membrane segment ion trans-port regulators. Thus, the methods might be applicable forstudy of related members of this family of membrane protein.The MS sequencing results are confirmed by the immunoblotof the rat kidney enzyme using antibodies raised against thepeptides TELSANHC or MDRWYLC (Fig. 4). In particular,   a -specific and    b -specific antibodies recognize either the upperor lower band, respectively, but no bands other than    a  or    b ,consistent with absence of post-translationally modified sub-units. It was reported recently that, after overnight incubationof rat kidney Na,K-ATPase with 1  M  hydroxylamine at pH 11and 37 °C, the upper     band disappeared, leading to a sugges-tion that it contains a hydroxyester link, possibly fatty acidacylation at Ser or Thr residues (14). In our hands, alkalitreatment itself reduces the intensity of both bands,    a  morethan    b , and hydroxylamine further reduces both bands. 2  Al-kaline hydroxylamine cleaves proteins, preferentially at Asn-Gly bonds, but specificity is not absolute (28). Indeed, we haveobserved that the    subunits of rat or pig kidney Na,K-ATPase,which contain three Asn-Gly bonds, are cleaved by the alkalinehydroxylamine at many positions. Although rat     subunits donot contain Asn-Gly bonds, nonspecific hydroxylaminolysis is alikely possibility. Thus, the inference of a post-translationalmodification of the     subunit in kidney Na,K-ATPase, basedon the hydroxylamine treatment alone, appears to bequestionable.Taken together with direct sequence analysis of the twobands from kidney enzyme and the antibody binding, the ex-pression studies clarify earlier uncertainties regarding dou-blets seen in kidney and in transfected cells. The expressionexperiment in Fig. 5 clearly demonstrates that the major    a and    b  protein products of transcription/translation in bothHeLa and HEK cells have the same mobilities as the upper (   a )and lower bands (   b ), respectively, of the kidney medulla. Thisis consistent with the results of the MS and antibody binding.The additional minor bands,    a  in    a -transfected HEK and theupper    b   band in    b -transfected HeLa, represent cell-specificmodifications of     a  and    b . In cells transfected with cDNA having a single initiator methionine, a doublet of chains cannotbe due to a splice variant and must have another explanation,for example post-translational modification or alternate trans-lation initiation sites. It was reported earlier that  in vitro translation of a single rat mRNA species, in the presence of dog pancreatic microsomes, gives rise to two     subunit bands (6).Geering and co-workers (10) have shown that, in  Xenopus , thepresence of two bands of      is secondary to alternate usage of two distinct start codons in the     subunit message. As men-tioned under “Results,” the nature of     a  seen in HEK cells and   b   seen in    b -transfected HeLa is obscure. Additionally, therelevance, if any, of     a  and    b  to the function of the     subunitin kidney is unknown. In the case of     b  , its mobility corre-sponds to that of a very weak but distinct band just above    a  insome, albeit not all Western blots of renal medulla. 2 The question arises as to the stoichiometry of     a  and    b relative to the    /    subunits in renal Na,K-ATPase. In olderstudies, the overall stoichiometry of      to    /    subunits in kidneyenzyme was estimated to be approximately 1:1 (3, 4). In agree-ment with a 1:1:1 stoichiometry of    :  :    subunits, we haverepeatedly found stoichiometries of 1:1 for a fragment of the    subunit (N terminus GDVDPFYY; see Ref. 30) and other frag-ments of the    subunit isolated from extensively digested pig kidney Na,K-ATPase. 2 By scanning Coomassie-stained gels, wehave also estimated that the ratio    a :    b  in rat or pig kidneyNa,K-ATPase, as 0.8    0.06 ( n    3) and 1.6    0.06 ( n    5),respectively,  i.e.  not exactly 1:1. 2 Thus, a likely combination of subunits is  :  :   a  and  :  :   b , in proportion to the ratio of    a  and   b . A combination such as   :  :   a :   b , together with   :   withoutassociated    , is less probable since it implies an exact 1:1 ratioof     a  and    b .What is the functional significance of the two variants? Onepossibility is that the    a  and    b  differentially affect functionsalready ascribed to the   subunit. For example, one effect of the    subunit is to alter the steady-state E 1 7 E 2  equilibrium infavor of E 1  with an associated decrease in  K    ATP  (6, 13).    a  and   b  could affect this function in different ways or to a differentdegree. Second, there may be effects of the     subunit on cyto-plasmic  K   K   /   K   Na  antagonism. Sweadner and co-workers (14)have recently ascribed a 2-fold lower apparent affinity for Na  of rat kidney Na,K-ATPase compared with that in rat NRK tissue-culture cells to the presence of      in the kidney. Theseauthors transfected NRK cells with    a  cDNA and isolatedclones expressing either two bands, presumed to be eitherunmodified (lower) or post-translationally modified (upper), oronlyasingle   a band,presumablypost-translationallymodified(upper). They observed a lower  K   Na  in NRK clones expressing both bands but not in clones expressing a single    a  subunit.However, in view of the rather low level of     a  expression inthose experiments, and the present finding that    a  in intactkidney tissue is unmodified, the relevance of those results tothe role of      in kidney is unclear. Nevertheless, striking tissue-specific difference in apparent affinity of the   1 pump for Na  have been described and ascribed to differences in cytoplasmic  K   K   /   K   Na  antagonism (31, 32). The difference is particularlynotable for the   1 enzyme of kidney compared with most othertissues tested, and significant in view of the predominant ex-pression of      in the intact kidney membranes (6). Anotherpossibility is that there are variant-specific functions related totheir distinct extracellularly located N termini. For example, in  Xenopus  oocytes expression of the     subunit affects activationby extracellular potassium ions (10), which could be modulatedby the different N termini. The     subunit was shown manyyears ago to bind a photoaffinity label, nitroazidobenzoyl-oua-bain (3). Thus, it is also conceivable that the different N terminiof the    a  and    b  variants differentially modulate binding of cardiac glycosides such as the endogenous ouabain-like com-pounds thought to be involved in generation of essential hyper-tension (33, 34). It is of interest that the N-terminal sequenceof     b , Ac-MDRWYL, is found in man, rat, mouse, and as recentMS experiments show also in pig, 3 while the N-terminal se-quence of     a  is less well conserved (see Ref. 27). An implicationcould be that MDRWYL is involved in a specific interactionwith a ligand or other pump subunit related to its specificfunction. It is also possible that the different N termini of     a and    b  interact specifically with other extracellular proteinsand subserve as yet unknown functions.Evidently the functional role(s) of the     subunit, and the two variants, are open questions. Successful expression of     a  and   b ,andavailabilityof    a -specificand   b -specificantibodies,nowprovide tools for study of their individual roles. The variant-specific antibodies will also help define differences in expres-sion in specific cell types or locations in the renal tubule.  Acknowledgments— We thank Mara Ferrandi (Prassis, Milano, Italy)for a gift of purified rat renal Na,K-ATPase and Michael Caplan (YaleUniversity School of Medicine, New Haven, CT) for the 6H antibody. We 2 H. X. Pu, R. Goldshleger, R. Blostein, and S. J. D. Karlish, unpub-lished results.  3 B. Ku¨ster, A. Shainskaya, and S. J. D. Karlish, unpublished data.  A Variant of the     Subunit of Na,K-ATPase  18445   a t  W ei  z m annI  n s  t  i   t   u t   e of   S  c i   en c  e on S  e p t   em b  er  8  ,2  0  0  8 www. j   b  c . or  gD  ownl   o a d  e d f  r  om

aac device

Apr 4, 2018

sn resume

Apr 4, 2018
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks