Differential Regulation of Renal Na,K-ATPase by Splice Variants of the γ Subunit
2002; Elsevier BV; Volume: 277; Issue: 12 Linguagem: Inglês
10.1074/jbc.m111552200
ISSN1083-351X
AutoresElena Arystarkhova, Claudia Donnet, Natalya K. Asinovski, Kathleen J. Sweadner,
Tópico(s)Hydrogen Storage and Materials
ResumoSodium and potassium-exchanging adenosine triphosphatase (Na,K-ATPase) in the kidney is associated with the γ subunit (γ, FXYD2), a single-span membrane protein that modulates ATPase properties. Rat and human γ occur in two splice variants, γa and γb, with different N termini. Here we investigated their structural heterogeneity and functional effects on Na,K-ATPase properties. Both forms were post-translationally modified duringin vitro translation with microsomes, indicating that there are four possible forms of γ. Site-directed mutagenesis revealed Thr2 and Ser5 as potential sites for post-translational modification. Similar modification can occur in cells, with consequences for Na,K-ATPase properties. We showed previously that stable transfection of γa into NRK-52E cells resulted in reduction of apparent affinities for Na+ and K+. Individual clones differed in γ post-translational modification, however, and the effect on Na+ affinity was absent in clones with full modification. Here, transfection of γb also resulted in clones with or without post-translational modification. Both groups showed a reduction in Na+ affinity, but modification was required for the effect on K+ affinity. There were minor increases in ATP affinity. The physiological importance of the reduction in Na+ affinity was shown by the slower growth of γa, γb, and γb′ transfectants in culture. The differential influence of the four structural variants of γ on affinities of the Na,K-ATPase for Na+ and K+, together with our previous finding of different distributions of γa and γb along the rat nephron, suggests a highly specific mode of regulation of sodium pump properties in kidney. Sodium and potassium-exchanging adenosine triphosphatase (Na,K-ATPase) in the kidney is associated with the γ subunit (γ, FXYD2), a single-span membrane protein that modulates ATPase properties. Rat and human γ occur in two splice variants, γa and γb, with different N termini. Here we investigated their structural heterogeneity and functional effects on Na,K-ATPase properties. Both forms were post-translationally modified duringin vitro translation with microsomes, indicating that there are four possible forms of γ. Site-directed mutagenesis revealed Thr2 and Ser5 as potential sites for post-translational modification. Similar modification can occur in cells, with consequences for Na,K-ATPase properties. We showed previously that stable transfection of γa into NRK-52E cells resulted in reduction of apparent affinities for Na+ and K+. Individual clones differed in γ post-translational modification, however, and the effect on Na+ affinity was absent in clones with full modification. Here, transfection of γb also resulted in clones with or without post-translational modification. Both groups showed a reduction in Na+ affinity, but modification was required for the effect on K+ affinity. There were minor increases in ATP affinity. The physiological importance of the reduction in Na+ affinity was shown by the slower growth of γa, γb, and γb′ transfectants in culture. The differential influence of the four structural variants of γ on affinities of the Na,K-ATPase for Na+ and K+, together with our previous finding of different distributions of γa and γb along the rat nephron, suggests a highly specific mode of regulation of sodium pump properties in kidney. K-ATPase, sodium and potassium-exchanging adenosine triphosphatase human embryonic kidney N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine gene nomenclature based on a conserved motif, Phe-X-Tyr-Asp, pronounced like "fix-it." The Na,K-ATPase,1 or sodium pump, is the principal enzyme in animal cells that maintains ionic gradients of Na+ and K+ at the expense of ATP hydrolysis. In kidney, the Na+ gradient provides the driving force not only for the reabsorption of sodium, but also for secondary transepithelial transport of various essential solutes and water. Controlling sodium pump activity is thus an essential element of renal regulation. The Na,K-ATPase has two obligatory subunits, the catalytic α subunit and a glycoprotein, β, which are encoded by multigene families (1Lingrel J.B. Orlowski J. Shull M.M. Price E.M. Prog. Nucleic Acid Res. Mol. Biol. 1990; 38: 37-89Crossref PubMed Scopus (369) Google Scholar). Expression of α and β isoforms is developmentally regulated, and appears to be species- and tissue-specific (2Sweadner K.J. Biochim. Biophys. Acta. 1989; 988: 185-220Crossref PubMed Scopus (904) Google Scholar, 3Blanco G. Mercer R.W. Am. J. Physiol. 1998; 275: F633-F650PubMed Google Scholar). In several experimental expression systems, the exchange of either α or β subunit isoforms affected enzymatic properties of the complex, most notably affinities for K+ and Na+ (3Blanco G. Mercer R.W. Am. J. Physiol. 1998; 275: F633-F650PubMed Google Scholar). Functional differences in intrinsic properties of Na,K-ATPase have been reported in the kidney; the affinity for Na+ varies significantly along the rat and rabbit nephron (4Feraille E. Carranza M.L. Rousselot M. Favre H. Am. J. Physiol. 1994; 267: F55-F62Crossref PubMed Google Scholar, 5Feraille E. Carranza M.L. Buffin-Meyer B. Rousselot M. Doucet A. Favre H. Am. J. Physiol. 1995; 268: C1277-C1283Crossref PubMed Google Scholar, 6Buffin-Meyer B. Marsy S. Barlet-Bas C. Cheval L. Younes-Ibrahim M. Rajerison R. Doucet A. J. Physiol. 1996; 490: 623-632Crossref PubMed Scopus (20) Google Scholar, 7Barlet Bas C. Cheval L. Khadouri C. Marsy S. Doucet A. Am. J. Physiol. 1990; 259: F246-F250PubMed Google Scholar). This led to an effort in many laboratories to determine whether α and β isoforms could account for the differences, but no segment-specific localization of isoforms other than α1 and β1 was found for either mRNA or protein (8Farman N. Corthesy-Theulaz I. Bonvalet J.P. Rossier B.C. Am. J. Physiol. 1991; : C468-C474Crossref PubMed Google Scholar, 9Welling P.A. Caplan M. Sutters M. Giebisch G. J. Biol. Chem. 1993; 268: 23469-23476Abstract Full Text PDF PubMed Google Scholar, 10Tumlin J.A. Hoban C.A. Medford R.M. Sands J.M. Am. J. Physiol. 1994; 266: F240-F245PubMed Google Scholar, 11Lucking K. Nielsen J.M. Pedersen P.A. Jorgensen P.L. Am. J. Physiol. 1996; 271: F253-F260PubMed Google Scholar, 12Hayward A.L. Hinojos C.A. Nurowska B. Hewetson A. Sabatini S. Oefner P.J. Doris P.A. J. Hypertens. 1999; 17: 1081-1087Crossref PubMed Scopus (21) Google Scholar). Two regulatory proteins that modulate intrinsic properties of the Na,K-ATPase have now been identified in kidney: the Na,K-ATPase γ subunit (13Forbush III, B. Kaplan J.H. Hoffman J.F. Biochemistry. 1978; 17: 3667-3676Crossref PubMed Scopus (231) Google Scholar, 14Mercer R.W. Biemesderfer D. Bliss Jr., D.P. Collins J.H. Forbush III., B. J. Cell Biol. 1993; 121: 579-586Crossref PubMed Scopus (185) Google Scholar), and CHIF ("channel-inducing factor") (15Attali B. Latter H. Rachamim N. Garty H. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 6092-6096Crossref PubMed Scopus (177) Google Scholar, 16Shi H. Levy-Holzman R. Cluzeaud F. Farman N. Garty H. Am. J. Physiol. 2001; 280: F505-F512Crossref PubMed Google Scholar). Based on sequence homology, both proteins belong to the FXYD gene family, which unites small (5–15 kDa) single-span membrane proteins including phospholemman (17Palmer C.J. Scott D. Jones L.R. J. Biol. Chem. 1991; 266: 11126-11130Abstract Full Text PDF PubMed Google Scholar), MAT-8 ("mammary tumor antigen-8") (18Morrison B.W. Moorman J.R. Kowdley G.C. Kobayashi Y.M. Jones L.R. Leder P. J. Biol. Chem. 1995; 270: 2176-2182Abstract Full Text Full Text PDF PubMed Scopus (140) Google Scholar), RIC ("related to ion channel") (19Fu X. Kamps M.P. Mol. Cell. Biol. 1997; 17: 1503-1512Crossref PubMed Scopus (94) Google Scholar), as well as two new gene products, called FXYD6 and FXYD7 (20Sweadner K.J. Rael E. Genomics. 2000; 68: 41-56Crossref PubMed Scopus (359) Google Scholar). Although Na,K-ATPase is widely distributed in kidney, expression of γ and CHIF is segment-specific. γ co-localizes with Na,K-ATPase in proximal tubules, distal convoluted tubules, and medullary thick ascending limb (21Pu H.X. Cluzeaud F. Goldshleger R. Karlish S.J.D. Farman N. Blostein R. J. Biol. Chem. 2001; 276: 20370-20378Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar, 22Wetzel R.K. Sweadner K.J. Am. J. Physiol. 2001; 281: F531-F545Crossref PubMed Google Scholar, 23Arystarkhova E. Wetzel R.K. Sweadner K.J. Am. J. Physiol. 2002; 282: F393-F407Crossref PubMed Scopus (83) Google Scholar), whereas CHIF co-localizes with the pump in collecting duct (16Shi H. Levy-Holzman R. Cluzeaud F. Farman N. Garty H. Am. J. Physiol. 2001; 280: F505-F512Crossref PubMed Google Scholar). The functional significance of γ (FXYD2) for the Na,K-ATPase complex was a mystery for a long time. The key properties of the Na,K-ATPase (ouabain binding, enzymatic activity, and cation transport) can be obtained without it (24Freytag J.W. FEBS Lett. 1983; 159: 280-284Crossref PubMed Scopus (10) Google Scholar, 25DeTomaso A.W. Xie Z.J. Liu G. Mercer R.W. J. Biol. Chem. 1993; 268: 1470-1478Abstract Full Text PDF PubMed Google Scholar, 26Scheiner-Bobis G. Farley R.A. Biochim. Biophys. Acta. 1994; 1193: 226-234Crossref PubMed Scopus (46) Google Scholar). It is also not required for αβ assembly and transport of functional units to the plasma membrane (27Geering K. Beggah A. Good P. Girardet S. Roy S. Schaer D. Jaunin P. J. Cell Biol. 1996; 133: 1193-1204Crossref PubMed Scopus (125) Google Scholar). However, incorporation of a photoaffinity-labeled derivative of ouabain into γ as well as α suggested that it may comprise part of the ouabain-binding site (13Forbush III, B. Kaplan J.H. Hoffman J.F. Biochemistry. 1978; 17: 3667-3676Crossref PubMed Scopus (231) Google Scholar). The first clear effect of γ on Na,K-ATPase transport properties was detected by electrophysiological measurements in Xenopusoocytes. It influenced the apparent affinity of the Na,K-ATPase pump current for extracellular K+ (28Beguin P. Wang X. Firsov D. Puoti A. Claeys D. Horisberger J.D. Geering K. EMBO J. 1997; 16: 4250-4260Crossref PubMed Scopus (215) Google Scholar). An antibody against γ inhibited Na,K-ATPase in vitro and decreased affinity for ATP (29Therien A.G. Goldshleger R. Karlish S.J.D. Blostein R. J. Biol. Chem. 1997; 272: 32628-32634Abstract Full Text Full Text PDF PubMed Scopus (129) Google Scholar). The functional role of γ was further assessed with γ-transfected mammalian cells. In HEK cells, γ increased affinity for ATP (30Therien A.G. Karlish S.J.D. Blostein R. J. Biol. Chem. 1999; 274: 12252-12256Abstract Full Text Full Text PDF PubMed Scopus (119) Google Scholar). NRK-52E cells express the same α1β1 combination as kidney but no γ, and apparent affinities for Na+ and K+ were higher than for α1β1γ from renal medulla (31Arystarkhova E. Wetzel R.K. Asinovski N.K. Sweadner K.J. J. Biol. Chem. 1999; 274: 33183-33185Abstract Full Text Full Text PDF PubMed Scopus (185) Google Scholar). Stable transfection of γa (the only known splice variant at the time) resulted in a reduction of affinities for Na+ and K+. This was observed with partially purified enzyme, indicating a stable functional alteration of enzyme properties. Interestingly, the modulation of Na+ affinity was abolished by a post-translational modification of γ that occurred in cell culture and that shifted its gel mobility (31Arystarkhova E. Wetzel R.K. Asinovski N.K. Sweadner K.J. J. Biol. Chem. 1999; 274: 33183-33185Abstract Full Text Full Text PDF PubMed Scopus (185) Google Scholar). A comparable decrease in apparent Na+ affinity has recently been ascribed to an increase in K+ antagonism of cytoplasmic Na+activation in γ-transfected HeLa cells (21Pu H.X. Cluzeaud F. Goldshleger R. Karlish S.J.D. Farman N. Blostein R. J. Biol. Chem. 2001; 276: 20370-20378Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar). Similarly, γ was shown to influence Na,K-ATPase pump current by lowering the apparent affinity for intracellular Na+ (32Beguin P. Crambert G. Guennoun S. Garty H. Horisberger J.-D. Geering K. EMBO J. 2001; 20: 3993-4002Crossref PubMed Scopus (134) Google Scholar). The injection of CHIF cRNA intoXenopus oocytes, in contrast, resulted in an increase in apparent affinity for Na+ (32Beguin P. Crambert G. Guennoun S. Garty H. Horisberger J.-D. Geering K. EMBO J. 2001; 20: 3993-4002Crossref PubMed Scopus (134) Google Scholar). Taken together with the almost complementary distributions of γ and CHIF along the rat nephron (16Shi H. Levy-Holzman R. Cluzeaud F. Farman N. Garty H. Am. J. Physiol. 2001; 280: F505-F512Crossref PubMed Google Scholar, 22Wetzel R.K. Sweadner K.J. Am. J. Physiol. 2001; 281: F531-F545Crossref PubMed Google Scholar), the functional data correlate very well with the differences in Na+ affinities in successive segments of the nephron (4Feraille E. Carranza M.L. Rousselot M. Favre H. Am. J. Physiol. 1994; 267: F55-F62Crossref PubMed Google Scholar, 6Buffin-Meyer B. Marsy S. Barlet-Bas C. Cheval L. Younes-Ibrahim M. Rajerison R. Doucet A. J. Physiol. 1996; 490: 623-632Crossref PubMed Scopus (20) Google Scholar, 7Barlet Bas C. Cheval L. Khadouri C. Marsy S. Doucet A. Am. J. Physiol. 1990; 259: F246-F250PubMed Google Scholar, 33Feraille E. Rousselot M. Rajerison R. Favre H. J. Physiol. 1995; 488: 171-180Crossref PubMed Scopus (53) Google Scholar). In rat kidney, γ occurs as at least two splice variants, γa and γb, with different N termini (20Sweadner K.J. Rael E. Genomics. 2000; 68: 41-56Crossref PubMed Scopus (359) Google Scholar, 34Sweadner K.J. Rael E. Wetzel R.K. Arystarkhova E. Taniguchi K. Kaya S. Na/K-ATPase and Related ATPases. Elsevier, Amsterdam2000: 543-546Google Scholar, 35Küster B. Shainskaya A. Pu H.X. Goldshleger R. Blostein R. Karlish S.J.D. J. Biol. Chem. 2000; 275: 18441-18446Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). The splice forms colocalized in the inner stripe of the outer medulla in the thick ascending limb of the loop of Henle (21Pu H.X. Cluzeaud F. Goldshleger R. Karlish S.J.D. Farman N. Blostein R. J. Biol. Chem. 2001; 276: 20370-20378Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar, 23Arystarkhova E. Wetzel R.K. Sweadner K.J. Am. J. Physiol. 2002; 282: F393-F407Crossref PubMed Scopus (83) Google Scholar), and co-immunoprecipitation with each other and α showed that they are associated in macromolecular complexes (23Arystarkhova E. Wetzel R.K. Sweadner K.J. Am. J. Physiol. 2002; 282: F393-F407Crossref PubMed Scopus (83) Google Scholar). Our observations indicate that their distribution in renal cortex and outer stripe of the outer medulla, however, is strongly biased to different segments. γa predominated in proximal tubules, whereas γb was detected only in distal and connecting tubules and the thick ascending limb in the outer stripe (23Arystarkhova E. Wetzel R.K. Sweadner K.J. Am. J. Physiol. 2002; 282: F393-F407Crossref PubMed Scopus (83) Google Scholar). Very recently, a third splice variant of γ has been described in mouse (36Jones D.H. Golding M.C. Barr K.J. Fong G.-H. Kidder G.M. Physiol. Genom. 2001; 6: 129-135Crossref PubMed Scopus (23) Google Scholar), although not in human (37Sweadner K.J. Wetzel R.K. Arystarkhova E. Biochem. Biophys. Res. Commun. 2000; 279: 196-201Crossref PubMed Scopus (42) Google Scholar). All of the splice variants in these characterized genes have different promoters, which provides a basis for differential regulation of gene expression. Here we have investigated the functional properties of γ subunit splice variants. Others have reported that γa and γb have similar effects on intrinsic properties of the pump. Both splice variants decreased apparent affinity for Na+ without significant alteration of K+ affinity in HeLa transfected cells (21Pu H.X. Cluzeaud F. Goldshleger R. Karlish S.J.D. Farman N. Blostein R. J. Biol. Chem. 2001; 276: 20370-20378Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar) and in Xenopus oocytes (32Beguin P. Crambert G. Guennoun S. Garty H. Horisberger J.-D. Geering K. EMBO J. 2001; 20: 3993-4002Crossref PubMed Scopus (134) Google Scholar), and increasedK m for ATP in HeLa transfectants (21Pu H.X. Cluzeaud F. Goldshleger R. Karlish S.J.D. Farman N. Blostein R. J. Biol. Chem. 2001; 276: 20370-20378Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar). We find, however, that post-translational modifications that occur in some but not all NRK-52E clones stably expressing γa or γb affect the final enzyme properties; consequently, γa can have modulating effects on Na,K-ATPase activity different from that of γb. Taken together with the differential distribution of γa and γb in kidney, the regulatory effects of γ on intrinsic properties of the Na,K-ATPase appear to be subject to multiple layers of control. Preliminary reports have been presented (38Arystarkhova E. Wetzel R.K. Rael E. Asinovski N.K. Sweadner K. Biophys. J. 2000; 78: 279AGoogle Scholar, 39Arystarkhova E. Asinovski N.K. Sweadner K.J. J. Am. Soc. Nephrol. 2000; 11: 24AGoogle Scholar). Isolation of membranes and purification of Na,K-ATPase from rat kidney outer medulla and whole mouse kidneys was performed by the Jørgensen procedure (40Jorgensen P.L. Methods Enzymol. 1974; 32: 277-290Crossref PubMed Scopus (262) Google Scholar). Briefly, microsomal fractions were collected by differential centrifugation, followed by SDS treatment and equilibrium centrifugation on sucrose density gradients. cDNA for the γb splice variant was obtained by reverse transcriptase-PCR from total rat kidney RNA (CLONTECH). The primers were based on nucleotide sequences for rat γb in the dbEST data base (GenBank™) plus EcoRI and BamHI restriction sites for unidirectional cloning: forward degenerate primer, 5′-GCGAATTCCACCATGGAYAGGTGGTACYTG-3′, where Y = (C/T); reverse primer, 5′-CGCGGATCCCAGCTCATCTTCATTGAC-3′. Gel-purified DNA was ligated into pIRES vector (CLONTECH) containing an internal ribosome binding site and the neomycin resistance gene. Several clones containing the full-length cDNA of γb were verified by nucleotide sequencing (GenBank™ AF233060). Point mutations were introduced into the γa cDNA in the pIRES vector by PCR and confirmed by DNA sequencing. The full-length cDNAs of rat γa (wild type and mutated) and γb were ligated into pT7 vector (Novagen) and translated in vitro using the TNT-T7 system (Promega) with or without addition of canine pancreatic microsomes according to the manufacturer's procedure. The reaction was for 90 min at 30 °C, followed by centrifugation in an Airfuge (Beckman) (10 min, 4 °C). The samples were washed with 100 mm NaCl and 25 mm HEPES, pH 7.4, and the final pellets were resuspended in 315 mm sucrose, 1 mm EDTA, and 20 mm Tris-Cl, pH 7.5. Portions were dissolved in SDS sample buffer, run on an SDS-Tricine gel (41Schagger H. von Jagow G. Anal. Biochem. 1987; 166: 368-379Crossref PubMed Scopus (10802) Google Scholar), and transferred to nitrocellulose. Detection of the newly synthesized proteins was with the RCT-G1 antibodies against the C terminus of γ (31Arystarkhova E. Wetzel R.K. Asinovski N.K. Sweadner K.J. J. Biol. Chem. 1999; 274: 33183-33185Abstract Full Text Full Text PDF PubMed Scopus (185) Google Scholar). Luciferase-T7 DNA was used as a positive nonradioactive control; samples without DNA added served as negative controls for transcription/translation background. Blots were scanned with a laser densitometer. NRK-52E is a rat renal cell line ("normal rat kidney") with polarized epithelial morphology. Transfection of NRK-52E cells and selection of γb-containing stable clones were performed as before for γa (31Arystarkhova E. Wetzel R.K. Asinovski N.K. Sweadner K.J. J. Biol. Chem. 1999; 274: 33183-33185Abstract Full Text Full Text PDF PubMed Scopus (185) Google Scholar), and isolated clones were propagated in the presence of G418 antibiotic selection. Purification of the Na,K-ATPase from mock- or γ-transfected cells was as described (31Arystarkhova E. Wetzel R.K. Asinovski N.K. Sweadner K.J. J. Biol. Chem. 1999; 274: 33183-33185Abstract Full Text Full Text PDF PubMed Scopus (185) Google Scholar). Briefly, cells were grown in 75-cm2 flasks until confluent in Dulbecco's modified Eagle's medium with 10% fetal bovine serum, washed with Dulbecco's PBS, and frozen. Crude membranes from the scraped cells were obtained by homogenization and differential centrifugation. Final purification of Na,K-ATPase was with SDS extraction, which leaves the protein in the lipid bilayer but removes many contaminating proteins. Electrophoresis was in SDS-Tricine gels (41Schagger H. von Jagow G. Anal. Biochem. 1987; 166: 368-379Crossref PubMed Scopus (10802) Google Scholar). Proteins were transferred to nitrocellulose, and the blots were incubated with specific antibodies. Detection was with chemiluminescence (Pierce). The RCT-G1 polyclonal antibody raised against the C-terminal peptide of γ (31Arystarkhova E. Wetzel R.K. Asinovski N.K. Sweadner K.J. J. Biol. Chem. 1999; 274: 33183-33185Abstract Full Text Full Text PDF PubMed Scopus (185) Google Scholar) was employed to detect both splice variants of the protein. The RNGB antibody (rat N terminus of γb) (23Arystarkhova E. Wetzel R.K. Sweadner K.J. Am. J. Physiol. 2002; 282: F393-F407Crossref PubMed Scopus (83) Google Scholar) was used for γb detection. Monoclonal antibody McK1 was used to stain α1 (42Arystarkhova E. Sweadner K.J. J. Biol. Chem. 1996; 271: 23407-23417Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar). Na,K-ATPase activity was measured as a function of Na+ concentration in media containing 3 mm Tris-ATP, 3 mm MgCl2, 30 mm histidine, pH 7.4, and in the presence of various concentrations of K+ (5–100 mm). Na,K-ATPase activity was also measured as a function of K+concentration (0–20 mm) with fixed [Na+] at 140 mm. ATP activation curves were obtained with 140 mm Na+, 20 mm K+, and 4 mm Mg2+ in the reaction medium. All the reactions were performed at 37 °C for 30–45 min with and without 3 mm ouabain, and ouabain-sensitive Pi release was measured colorimetrically by using either Fiske-Subbarow or Baginski methods, or by release of 32P from [γ-32P]ATP. Data were analyzed by nonlinear regression using Sigma Plot Graph System (Jandel Scientific). Na+ and K+ activation curves were fitted according to the Hill model for ligand binding. K m values for ATP stimulation were derived from the Michaelis-Menten equation, also by nonlinear regression. Student's statistical test was used to assess differences. An exception to the above methods applies to data replotted from Ref.43Sweadner K.J. J. Biol. Chem. 1985; 260: 11508-11513Abstract Full Text PDF PubMed Google Scholar. There, activity was assayed continuously in a spectrophotometric coupled assay performed in a thermostatted cuvette. Assays were performed with different starting K+ concentrations and nominally zero Na+, and aliquots of NaCl were added successively, allowing determination of a new slope with each addition. Although it is well accepted that there are two γ splice variants in the rat (20Sweadner K.J. Rael E. Genomics. 2000; 68: 41-56Crossref PubMed Scopus (359) Google Scholar, 21Pu H.X. Cluzeaud F. Goldshleger R. Karlish S.J.D. Farman N. Blostein R. J. Biol. Chem. 2001; 276: 20370-20378Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar, 23Arystarkhova E. Wetzel R.K. Sweadner K.J. Am. J. Physiol. 2002; 282: F393-F407Crossref PubMed Scopus (83) Google Scholar, 32Beguin P. Crambert G. Guennoun S. Garty H. Horisberger J.-D. Geering K. EMBO J. 2001; 20: 3993-4002Crossref PubMed Scopus (134) Google Scholar, 34Sweadner K.J. Rael E. Wetzel R.K. Arystarkhova E. Taniguchi K. Kaya S. Na/K-ATPase and Related ATPases. Elsevier, Amsterdam2000: 543-546Google Scholar, 35Küster B. Shainskaya A. Pu H.X. Goldshleger R. Blostein R. Karlish S.J.D. J. Biol. Chem. 2000; 275: 18441-18446Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar), the nature and role of structural modifications of these variants have been controversial. Here we begin by examining the electrophoretic mobilities of γ forms found in kidney, and then use in vitro translation of γa and γb to detect post-translational modification of both splice variants. Fig. 1 A demonstrates γ band resolution in gels of typical preparations of Na,K-ATPase from rat and mouse kidney. The blot was stained with the RCT-G1 antibody that recognizes the C terminus of γ, a site that is highly conserved between species and identical in the splice variants. γ was seen as a clear doublet in both species, although the resolution was significantly better for mouse. 2A third, larger splice variant, γc, has been reported for mouse (36Jones D.H. Golding M.C. Barr K.J. Fong G.-H. Kidder G.M. Physiol. Genom. 2001; 6: 129-135Crossref PubMed Scopus (23) Google Scholar), but if it was present in adult mouse kidney, it was at levels that were not detected here. A similar slower migration of γa was observed in Na,K-ATPase preparations from sheep kidney (not shown). Based on gel mobility, we estimated the differences in apparent molecular masses for splice variants as 1.5 and 2.5 kDa for rat and mouse, respectively, with γa appearing larger. For the rat, the calculated molecular masses based on amino acid sequence are almost identical, however: 7245.7 for γa and 7234.8 for γb, assuming no modification or oxidation. Determination of the molecular masses by mass spectrometry revealed 7184.0 ± 1 for rat γa (with carbamidomethyl cysteine and without initiator methionine) and 7337.9 ± 1 for rat γb (with carbamidomethyl cysteine and with acetylated methionine) (35Küster B. Shainskaya A. Pu H.X. Goldshleger R. Blostein R. Karlish S.J.D. J. Biol. Chem. 2000; 275: 18441-18446Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar), indicating that the larger species actually migrates faster. The sequence of mouse γa is four amino acids longer (Fig. 1B), and the calculated difference between molecular masses of unmodified mouse γa and γb is about 300 (7515.9 for γaversus 7204.7 for γb). For both species, the calculated masses are close enough that the proteins could co-migrate on SDS gels, but splice variants of γ are generally seen as a doublet (23Arystarkhova E. Wetzel R.K. Sweadner K.J. Am. J. Physiol. 2002; 282: F393-F407Crossref PubMed Scopus (83) Google Scholar, 32Beguin P. Crambert G. Guennoun S. Garty H. Horisberger J.-D. Geering K. EMBO J. 2001; 20: 3993-4002Crossref PubMed Scopus (134) Google Scholar,35Küster B. Shainskaya A. Pu H.X. Goldshleger R. Blostein R. Karlish S.J.D. J. Biol. Chem. 2000; 275: 18441-18446Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). Thus, the apparent difference in electrophoretic mobility must be influenced by something else, such as net charge or a labile post-translational modification. We investigated the electrophoretic mobilities by analyzing proteins translated in vitro from synthetic mRNA. First we examined proteins made in reticulocyte lysates without addition of any microsomes. Newly synthesized proteins were analyzed on SDS-Tricine gels and stained with the RCT-G1 antibody, the blot was scanned with a laser densitometer, and the scans were superimposed. As shown in Fig.2 A, the splice variants of rat γ migrated at different rates, γa slower than γb, as shown with tissue-derived samples (23Arystarkhova E. Wetzel R.K. Sweadner K.J. Am. J. Physiol. 2002; 282: F393-F407Crossref PubMed Scopus (83) Google Scholar, 35Küster B. Shainskaya A. Pu H.X. Goldshleger R. Blostein R. Karlish S.J.D. J. Biol. Chem. 2000; 275: 18441-18446Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). Addition of endoplasmic reticulum to the reaction mixture permits ribosome attachment and cotranslational membrane protein insertion. When canine pancreatic microsomes were added to the reaction mixture, the electrophoretic mobility of both proteins was decreased, implying that both splice variants can be post-translationally modified (Fig. 2,B and C). The shift observed for γa was more pronounced than that for γb, indicating either different types of modification, or modification of more sites in γa. Either way, the difference in the shift between the splice variants is conjectural evidence that the location of the modification is in the spliced N-terminal segment, and that it occurs in the lumen of rough microsomes. Beguin et al. (32Beguin P. Crambert G. Guennoun S. Garty H. Horisberger J.-D. Geering K. EMBO J. 2001; 20: 3993-4002Crossref PubMed Scopus (134) Google Scholar) reported similar shifts in electrophoretic mobility for both splice variants of γ synthesized in a reticulocyte system supplemented with microsomes. Thus γ may exist (at least in vitro) in four different structural forms: γa, γa′, γb, and γb′, where the mark represents post-translational modification(s). To identify possible sites of modification by acylation or esterification, we mutated the oxygen-containing residues of the N-terminal segment of γ. Thr2 and Ser5residues are conserved in γa from all mammalian species characterized so far (20Sweadner K.J. Rael E. Genomics. 2000; 68: 41-56Crossref PubMed Scopus (359) Google Scholar). Site-directed mutagenesis to alanine was followed byin vitro translation and electrophoresis. As shown in Fig.3, both residues appeared to be important for post-translational modification. Mutations of either Thr2 or Ser5 almost completely blocked the shift in mobility of γa seen with pancreatic microsomes. However, when Ser11, a potential site for modification in the sequence shared by γa and γb, was replaced by Ala, migration of the mutated protein was reduced to an intermediate position (Fig.3 C). In the absence of microsomes, all mutants migrated at the same position as their unmutated form (data not shown). The data are consistent with a possible coopera
Referência(s)