Acute Inhibition of Na/H Exchanger NHE-3 by cAMP
1999; Elsevier BV; Volume: 274; Issue: 7 Linguagem: Inglês
10.1074/jbc.274.7.3978
ISSN1083-351X
AutoresHui Zhao, Michael R. Wiederkehr, Lingzhi Fan, Roberto Collazo, Ladonna A. Crowder, Orson W. Moe,
Tópico(s)Amino Acid Enzymes and Metabolism
ResumoRegulation of the renal Na/H exchanger NHE-3 by protein kinase A (PKA) is a key intermediate step in the hormonal regulation of acid-base and salt balance. We studied the role of NHE-3 phosphorylation in this process in NHE-deficient AP-1 cells transfected with NHE-3 and in OKP cells expressing native NHE-3. A dominant-negative PKA-regulatory subunit completely abolished the effect of cAMP on NHE-3 activity demonstrating a role of PKA in the functional regulation of NHE-3 by cAMP. NHE-3 isolated from cAMP-treated cells showed lower phosphorylation by purified PKAin vitro suggesting that NHE-3 is a PKA substrate in vivo. Although changes in NHE-3 whole protein phosphorylation is difficult to detect in response to cAMP addition, the tryptic phosphopeptide map of in vivo phosphorylated NHE-3 showed a complex pattern of constitutive and cAMP-induced phosphopeptides. To test the causal relationship between phosphorylation and activity, we mutated eight serines in the cytoplasmic domain to glycine or alanine. Single or multiple mutants harboring S552A or S605G showed no PKA activation or reduced regulation by PKA activation. Ser-552 and Ser-605 were phosphorylated in vivo. However, multiple mutations of serines other than Ser-552 or Ser-605 also reduced the functional PKA regulation. We conclude that regulation of NHE-3 by PKA in vivo involves complex mechanisms, which include phosphorylation of Ser-552 and Ser-605. Regulation of the renal Na/H exchanger NHE-3 by protein kinase A (PKA) is a key intermediate step in the hormonal regulation of acid-base and salt balance. We studied the role of NHE-3 phosphorylation in this process in NHE-deficient AP-1 cells transfected with NHE-3 and in OKP cells expressing native NHE-3. A dominant-negative PKA-regulatory subunit completely abolished the effect of cAMP on NHE-3 activity demonstrating a role of PKA in the functional regulation of NHE-3 by cAMP. NHE-3 isolated from cAMP-treated cells showed lower phosphorylation by purified PKAin vitro suggesting that NHE-3 is a PKA substrate in vivo. Although changes in NHE-3 whole protein phosphorylation is difficult to detect in response to cAMP addition, the tryptic phosphopeptide map of in vivo phosphorylated NHE-3 showed a complex pattern of constitutive and cAMP-induced phosphopeptides. To test the causal relationship between phosphorylation and activity, we mutated eight serines in the cytoplasmic domain to glycine or alanine. Single or multiple mutants harboring S552A or S605G showed no PKA activation or reduced regulation by PKA activation. Ser-552 and Ser-605 were phosphorylated in vivo. However, multiple mutations of serines other than Ser-552 or Ser-605 also reduced the functional PKA regulation. We conclude that regulation of NHE-3 by PKA in vivo involves complex mechanisms, which include phosphorylation of Ser-552 and Ser-605. Mammalian plasma membrane Na/H exchangers (NHEs) 1The abbreviations used are: NHE, Na/H exchanger; PKA, protein kinase A; PAGE, polyacrylamide gel electrophoresis; CSU, catalytic subunit; RSU, regulatory subunit; TPCK, N-tyrosyl-l-phenylalanine chloromethyl ketone; ANOVA, analysis of variance; WT, wild type.1The abbreviations used are: NHE, Na/H exchanger; PKA, protein kinase A; PAGE, polyacrylamide gel electrophoresis; CSU, catalytic subunit; RSU, regulatory subunit; TPCK, N-tyrosyl-l-phenylalanine chloromethyl ketone; ANOVA, analysis of variance; WT, wild type. use downhill inward Na+ gradients to extrude H+ from cells. Six genetic NHE isoforms have been identified with specific pharmacologic characteristics and tissue distributions (reviewed in Ref. 1Wakabayashi S. Shigekawa M. Pouyssegur J. Physiol. Rev. 1997; 77: 51-74Crossref PubMed Scopus (562) Google Scholar). NHE-3 is limited to transporting epithelia such as the kidney and gastrointestinal tract (2Orlowski J. Kandasamy R.A. Shull G. J. Biol. Chem. 1992; 267: 9331-9339Abstract Full Text PDF PubMed Google Scholar, 3Tse C.-M. Brant S.R. Walker M.S. Pouyssegur J. Donowitz M. J. Biol. Chem. 1992; 267: 9340-9346Abstract Full Text PDF PubMed Google Scholar). In the kidney, NHE-3 is expressed exclusively on the apical membrane of the proximal tubule and the thick ascending limb where it mediates absorption of a significant fraction of the filtered NaCl and NaHCO3 (4Biemesderfer D. Pizzonia J.H. Exner M. Reilly R.F. Igarashi P. Aronson P.S. Am. J. Physiol. 1993; 34: F736-F742Google Scholar, 5Amemiya M. Loffing J. Lötscher M. Kaissling B. Alpern R.J. Moe O.W. Kidney Int. 1995; 48: 1206-1215Abstract Full Text PDF PubMed Scopus (348) Google Scholar, 6Biemesderfer D. Rutherford P.A. Nagy T. Pizzonia J.H. Abu-Alfa A.K. Aronson P.S. Am. J. Physiol. 1997; 273: F289-F299Crossref PubMed Google Scholar). NHE-3 plays a critical role in renal regulation of extracellular fluid volume and acid-base balance (7Moe O.W. Alpern R.J. Fleigel L. The Structure, Function and Molecular Biology of Na/H Exchangers. R. G. Landes Co., Austin, TX1998Google Scholar). Acute hormonal regulation of NHE-3 involves multiple intracellular cascades including activation of adenylyl cyclase (1Wakabayashi S. Shigekawa M. Pouyssegur J. Physiol. Rev. 1997; 77: 51-74Crossref PubMed Scopus (562) Google Scholar). The NHE-3 isoform when expressed in fibroblasts is inhibited by cAMP analogues or forskolin (8Kandasamy R.A. Yu F.H. Harris R. Boucher A. Hanrahan J.W. Orlowski J. J. Biol. Chem. 1995; 270: 29209-29216Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar, 9Azarani A. Goltzman D. Orlowski J. J. Biol. Chem. 1995; 270: 20004-20010Abstract Full Text Full Text PDF PubMed Scopus (92) Google Scholar, 10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar). Although the NHE-1 isoform is cAMP insensitive, chimeric insertion of the NHE-3 cytoplasmic domain confers cAMP sensitivity to the NHE-1 transporting domain (11Cabado A.G. Yu F.H. Kapus A. Lukacs G. Grinstein S. Orlowski J. J. Biol Chem. 1996; 271: 3590-3599Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar). Partial deletions of the cytoplasmic domain of NHE-3 reduce although more drastic truncations abolish the effect of cAMP on NHE-3 activity (10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar, 11Cabado A.G. Yu F.H. Kapus A. Lukacs G. Grinstein S. Orlowski J. J. Biol Chem. 1996; 271: 3590-3599Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar). The NHE-3 cytoplasmic domain is a substrate for protein kinase A (PKA) in vitro (10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar) and NHE-3 phosphorylation is increased by cAMP or forskolin addition in vivo (10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar, 12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar). Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) showed that Ser-605 and Ser-634 are both important for forskolin to inhibit NHE-3 activity, although only Ser-605 is phosphorylated in vivo. This study shows that PKA directly phosphorylates NHE-3 and inhibits its activity via complex mechanisms. Phosphorylation of Ser-552 and Ser-605 were increased by cAMP addition and appeared to be critical for functional inhibition of NHE-3, although other regions of the transporter are likely to be involved in PKA regulation.DISCUSSIONWe confirmed the previous finding of inhibition ofV max by cAMP in native NHE-3 in renal epithelial cells (24Pollock A.S. Warnock D.G. Strewler G.J. Am. J. Physiol. 1986; 250: F217-F225PubMed Google Scholar, 25Casavola V.C. Helmle-Kolb C. Murer H. Biochem. Biophys. Res. Commun. 1989; 165: 833-837Crossref PubMed Scopus (45) Google Scholar, 26Cano A. Preisig P.A. Alpern R.J. J. Clin. Invest. 1993; 92: 1632-1638Crossref PubMed Scopus (46) Google Scholar) and NHE-3 heterologously expressed in fibroblasts (8Kandasamy R.A. Yu F.H. Harris R. Boucher A. Hanrahan J.W. Orlowski J. J. Biol. Chem. 1995; 270: 29209-29216Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar, 9Azarani A. Goltzman D. Orlowski J. J. Biol. Chem. 1995; 270: 20004-20010Abstract Full Text Full Text PDF PubMed Scopus (92) Google Scholar, 10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar, 11Cabado A.G. Yu F.H. Kapus A. Lukacs G. Grinstein S. Orlowski J. J. Biol Chem. 1996; 271: 3590-3599Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar, 12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar). The lack of detectable NHE-3 whole protein phosphorylation is comparable with the report by Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) although we observed discernible but variable changes in NHE-3 whole protein phosphorylation with 100 μm 8-Br-cAMP. As some cation channels are directly gated by cyclic nucleotides (19Kaupp U.B. Curr. Opin. Neurobiol. 1995; 5: 434-442Crossref PubMed Scopus (130) Google Scholar), we queried if such can be a model for NHE-3 regulation by cAMP. When we inactivated the catalytic subunit of PKA, 8-Br-cAMP no longer regulated NHE-3 activity indicating that the cAMP effect is kinase-dependent. Also congruent with A kinase-mediated action is the fact that unlike the cyclic nucleotide-gated channels (27Goulding E.H. Tibbs G.R. Siegelbaum S.A. Nature. 1994; 372: 369-374Crossref PubMed Scopus (179) Google Scholar), the primary sequence of NHE-3 does not conform to the β-roll/α-helix motif derived from the catabolite gene activator protein model (28McKay D.B. Stietz T.A. Nature. 1981; 290: 744-749Crossref PubMed Scopus (460) Google Scholar). In addition, a recombinant fusion protein of maltose-binding protein/NHE-3 cytoplasmic domain failed to bind3H-cAMP (data not shown).The cytoplasmic domain of NHE-3 contains numerous PKA consensus target sites and NHE-3 is rapidly phosphorylated by PKA in vitro(10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar). Although NHE-3 is a phosphoprotein in vivo (10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar, 12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar), there is no a priori reason to assume that PKA directly phosphorylates NHE-3 in vivo. The basis of the back phosphorylation assay rests on the presumption that certain common sites are preserved for in vivo and in vitrokinase reactions and multiple kinases do not converge on the exact same serine residues. Our results suggest that PKA likely phosphorylates NHE-3 in the cell. It is possible that the regulation of NHE-3 by cAMP involves phosphorylation of proteins other than NHE-3 by PKA and that PKA may activate other kinases that phosphorylate NHE-3. Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) showed that the lack of change in total NHE-3 protein phosphorylation in vivo is because of multiple constitutively phosphorylated sites masking the change in the regulated sites. We also found that phosphopeptides B, C, and D are largely unregulated in vivo. In contrast, phosphopeptide E (Ser-605) and J (Ser-552) were increased at 1 and 10 μm 8-Br-cAMP, respectively. The dose-response profiles of phosphopeptides E and J actually correlate well with that of changes in NHE-3 activity in response to cAMP.If phosphorylation of Ser-552 and Ser-605 are important, then elimination of phosphorylation should abolish functional regulation. All the single and double point mutants showed comparable antigen expression and baseline transport activity but the quadruple mutants showed decreased and the six-point mutant showed markedly attenuated expression of both NHE-3 protein and activity. Because NHE-3 transcript levels are not affected by the mutations, one has to postulate that the multiple serine mutations significantly affected translation and/or protein half-life. This study does not address the mechanism of these changes. Ser-552 and Ser-605 were both phosphorylated in vivo and when mutated singly to nonphosphorylatable residues, functional regulation by PKA was abrogated in both mutants. Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) have reported the importance of Ser-605 and Ser-634 in the functional regulation of NHE-3 by PKA although Ser-634 appeared not to be a phosphoserine. Our phosphopeptide map and functional response of the NHE-3mut/S605A is very similar to the previous report. A major disparate finding is that we found Ser-552 to be critical for PKA regulation, whereas Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) found normal functional regulation of NHE-3mut/S552A by cAMP. In addition, Cabado and co-workers (11Cabado A.G. Yu F.H. Kapus A. Lukacs G. Grinstein S. Orlowski J. J. Biol Chem. 1996; 271: 3590-3599Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar) from the same group found that truncation of the cytoplasmic domain at amino acid 579 rendered NHE-3 nonresponsive to cAMP. The difference may reside in the fact that Ser-552, which is phosphorylated in our cells, was not phosphorylated in the previous report as is evident by the absence of phosphopeptide J (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar). It is conceivable that multiple PKA target phosphoserines may participate in mediating the functional inhibition of NHE-3 and different host cells may use different sets of serines. The basis for the differential NHE-3 phosphorylation is unclear, because both laboratories used AP-1 cells from the same source. Because AP-1 cells are mutants derived and selected from the parental Chinese hamster ovary cell, repeated passages and clonal expansion might have created different phenotypes.One possible interpretation for the lack of regulation of NHE-3mut/S552A is that some cellular cofactor(s) are missing in that particular AP-1 recipient that harbors the mutant NHE-3 and that the cAMP-insensitive phenotype stems from the cell rather than NHE-3mut/S552A. Such an explanation is unlikely for several reasons. All pooled transfectants of single or multiple mutant NHE-3s were generated about the same time from the same parental AP-1 cells. In addition, when we tried to examined NHE-3mut/S552A in as heterogenous an AP-1 cellular background as we can create, we found that the cAMP-nonresponsive phenotype prevailed.The lack of regulation of NHE-3mut/S575/S661/S690/S804 is harder to explain because none of the serines were functionally important when examined as single mutants. One can speculate on several explanations. One is that there are multiple regulatory phosphoserines with P-Ser-552 and P-Ser-605 having the most dominant effects with lesser contribution from the other phosphoserines. Although singular mutations of serines other than Ser-552 and Ser-605 do not affect the overall response, mutation of four minor phosphorylation sites may be sufficient. A second possibility is that these serines participate in specific protein-protein interactions with regulatory factors such as NHE-RF (20Weinman E.J. Steplock D. Shenolikar S. J. Clin. Invest. 1993; 92: 1781-1786Crossref PubMed Scopus (118) Google Scholar, 21Weinman E.J. Steplock D. Wang Y. Shenolikar S. J. Clin. Invest. 1995; 95: 2143-2149Crossref PubMed Scopus (310) Google Scholar, 29Weinman E.J. Dubinsky W.P. Dinh Q. Steplock D. Shenolikar S. J. Membr. Biol. 1989; 109: 233-241Crossref PubMed Scopus (28) Google Scholar, 30Morell G. Steplock D. Shenolikar S. Weinman E.J. Am. J. Physiol. 1990; 259: F867-F871PubMed Google Scholar, 31Weinman E.J. Steplock D. Bui G. Yuan N. Shenolikar S. Am. J. Physiol. 1998; 258: F1254-F1258Google Scholar), E3KARP (22Yun C.H.C. Oh S. Zizak M. Steplock D. Taso S. Tse C.M. Weinman E.J. Donowitz M. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 3010-3030Crossref PubMed Scopus (404) Google Scholar), or calmodulin (32Levine S.A. Nath S.K. Yun C.H.C. Yip J.W. Montrose M. Donowitz M. Tse C.M. J. Biol. Chem. 1995; 270: 13716-13725Crossref PubMed Scopus (103) Google Scholar) that modulate NHE-3 function, and this interaction is disrupted by the mutations. A third possibility is that the mutations distort the protein structure to such an extent that it nonspecifically alters NHE-3 function and its regulation. Crystallographic evidence indicates that one can significantly alter the protein structure of a domain of a protein by a single amino acid mutation from tyrosine to phenylalanine, which essentially only removes a single hydroxyl group (33Zhang J. Zhang F. Ebert D. Cobb M.H. Goldsmith E.J. Structure. 1995; 3: 299-307Abstract Full Text Full Text PDF PubMed Scopus (109) Google Scholar). The present paper does not distinguish the three stated possibilities. We did however observe intact regulation of all mutants including the four- to six-point mutants by acute hyperosmolality (25–40% inhibition) (data not shown). This is compatible with the ability of hyperosmolality to inhibit NHE-3 activity despite radical truncations of the cytoplasmic domain (34Nath S.K. Hang C.Y. Levine S.A. Yun C.H.C. Montrose M.H. Donowitz M. Tse C.M. Am. J. Physiol. 1996; 270: G431-G441PubMed Google Scholar).In summary, we hypothesize that the functional regulation of NHE-3 by PKA does not converge on single covalent modifications such as phosphorylation of a specific serine. We submit a model where PKA phosphorylates NHE-3 on Ser-552 and Ser-605 in addition to other serines. Phosphorylation of Ser-552 and Ser-605 each play a major although not exclusive role in the functional regulation of NHE-3, which involves specific interaction of various regions of the cytoplasmic domain of NHE-3 with other regulatory cofactors. Mammalian plasma membrane Na/H exchangers (NHEs) 1The abbreviations used are: NHE, Na/H exchanger; PKA, protein kinase A; PAGE, polyacrylamide gel electrophoresis; CSU, catalytic subunit; RSU, regulatory subunit; TPCK, N-tyrosyl-l-phenylalanine chloromethyl ketone; ANOVA, analysis of variance; WT, wild type.1The abbreviations used are: NHE, Na/H exchanger; PKA, protein kinase A; PAGE, polyacrylamide gel electrophoresis; CSU, catalytic subunit; RSU, regulatory subunit; TPCK, N-tyrosyl-l-phenylalanine chloromethyl ketone; ANOVA, analysis of variance; WT, wild type. use downhill inward Na+ gradients to extrude H+ from cells. Six genetic NHE isoforms have been identified with specific pharmacologic characteristics and tissue distributions (reviewed in Ref. 1Wakabayashi S. Shigekawa M. Pouyssegur J. Physiol. Rev. 1997; 77: 51-74Crossref PubMed Scopus (562) Google Scholar). NHE-3 is limited to transporting epithelia such as the kidney and gastrointestinal tract (2Orlowski J. Kandasamy R.A. Shull G. J. Biol. Chem. 1992; 267: 9331-9339Abstract Full Text PDF PubMed Google Scholar, 3Tse C.-M. Brant S.R. Walker M.S. Pouyssegur J. Donowitz M. J. Biol. Chem. 1992; 267: 9340-9346Abstract Full Text PDF PubMed Google Scholar). In the kidney, NHE-3 is expressed exclusively on the apical membrane of the proximal tubule and the thick ascending limb where it mediates absorption of a significant fraction of the filtered NaCl and NaHCO3 (4Biemesderfer D. Pizzonia J.H. Exner M. Reilly R.F. Igarashi P. Aronson P.S. Am. J. Physiol. 1993; 34: F736-F742Google Scholar, 5Amemiya M. Loffing J. Lötscher M. Kaissling B. Alpern R.J. Moe O.W. Kidney Int. 1995; 48: 1206-1215Abstract Full Text PDF PubMed Scopus (348) Google Scholar, 6Biemesderfer D. Rutherford P.A. Nagy T. Pizzonia J.H. Abu-Alfa A.K. Aronson P.S. Am. J. Physiol. 1997; 273: F289-F299Crossref PubMed Google Scholar). NHE-3 plays a critical role in renal regulation of extracellular fluid volume and acid-base balance (7Moe O.W. Alpern R.J. Fleigel L. The Structure, Function and Molecular Biology of Na/H Exchangers. R. G. Landes Co., Austin, TX1998Google Scholar). Acute hormonal regulation of NHE-3 involves multiple intracellular cascades including activation of adenylyl cyclase (1Wakabayashi S. Shigekawa M. Pouyssegur J. Physiol. Rev. 1997; 77: 51-74Crossref PubMed Scopus (562) Google Scholar). The NHE-3 isoform when expressed in fibroblasts is inhibited by cAMP analogues or forskolin (8Kandasamy R.A. Yu F.H. Harris R. Boucher A. Hanrahan J.W. Orlowski J. J. Biol. Chem. 1995; 270: 29209-29216Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar, 9Azarani A. Goltzman D. Orlowski J. J. Biol. Chem. 1995; 270: 20004-20010Abstract Full Text Full Text PDF PubMed Scopus (92) Google Scholar, 10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar). Although the NHE-1 isoform is cAMP insensitive, chimeric insertion of the NHE-3 cytoplasmic domain confers cAMP sensitivity to the NHE-1 transporting domain (11Cabado A.G. Yu F.H. Kapus A. Lukacs G. Grinstein S. Orlowski J. J. Biol Chem. 1996; 271: 3590-3599Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar). Partial deletions of the cytoplasmic domain of NHE-3 reduce although more drastic truncations abolish the effect of cAMP on NHE-3 activity (10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar, 11Cabado A.G. Yu F.H. Kapus A. Lukacs G. Grinstein S. Orlowski J. J. Biol Chem. 1996; 271: 3590-3599Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar). The NHE-3 cytoplasmic domain is a substrate for protein kinase A (PKA) in vitro (10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar) and NHE-3 phosphorylation is increased by cAMP or forskolin addition in vivo (10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar, 12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar). Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) showed that Ser-605 and Ser-634 are both important for forskolin to inhibit NHE-3 activity, although only Ser-605 is phosphorylated in vivo. This study shows that PKA directly phosphorylates NHE-3 and inhibits its activity via complex mechanisms. Phosphorylation of Ser-552 and Ser-605 were increased by cAMP addition and appeared to be critical for functional inhibition of NHE-3, although other regions of the transporter are likely to be involved in PKA regulation. DISCUSSIONWe confirmed the previous finding of inhibition ofV max by cAMP in native NHE-3 in renal epithelial cells (24Pollock A.S. Warnock D.G. Strewler G.J. Am. J. Physiol. 1986; 250: F217-F225PubMed Google Scholar, 25Casavola V.C. Helmle-Kolb C. Murer H. Biochem. Biophys. Res. Commun. 1989; 165: 833-837Crossref PubMed Scopus (45) Google Scholar, 26Cano A. Preisig P.A. Alpern R.J. J. Clin. Invest. 1993; 92: 1632-1638Crossref PubMed Scopus (46) Google Scholar) and NHE-3 heterologously expressed in fibroblasts (8Kandasamy R.A. Yu F.H. Harris R. Boucher A. Hanrahan J.W. Orlowski J. J. Biol. Chem. 1995; 270: 29209-29216Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar, 9Azarani A. Goltzman D. Orlowski J. J. Biol. Chem. 1995; 270: 20004-20010Abstract Full Text Full Text PDF PubMed Scopus (92) Google Scholar, 10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar, 11Cabado A.G. Yu F.H. Kapus A. Lukacs G. Grinstein S. Orlowski J. J. Biol Chem. 1996; 271: 3590-3599Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar, 12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar). The lack of detectable NHE-3 whole protein phosphorylation is comparable with the report by Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) although we observed discernible but variable changes in NHE-3 whole protein phosphorylation with 100 μm 8-Br-cAMP. As some cation channels are directly gated by cyclic nucleotides (19Kaupp U.B. Curr. Opin. Neurobiol. 1995; 5: 434-442Crossref PubMed Scopus (130) Google Scholar), we queried if such can be a model for NHE-3 regulation by cAMP. When we inactivated the catalytic subunit of PKA, 8-Br-cAMP no longer regulated NHE-3 activity indicating that the cAMP effect is kinase-dependent. Also congruent with A kinase-mediated action is the fact that unlike the cyclic nucleotide-gated channels (27Goulding E.H. Tibbs G.R. Siegelbaum S.A. Nature. 1994; 372: 369-374Crossref PubMed Scopus (179) Google Scholar), the primary sequence of NHE-3 does not conform to the β-roll/α-helix motif derived from the catabolite gene activator protein model (28McKay D.B. Stietz T.A. Nature. 1981; 290: 744-749Crossref PubMed Scopus (460) Google Scholar). In addition, a recombinant fusion protein of maltose-binding protein/NHE-3 cytoplasmic domain failed to bind3H-cAMP (data not shown).The cytoplasmic domain of NHE-3 contains numerous PKA consensus target sites and NHE-3 is rapidly phosphorylated by PKA in vitro(10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar). Although NHE-3 is a phosphoprotein in vivo (10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar, 12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar), there is no a priori reason to assume that PKA directly phosphorylates NHE-3 in vivo. The basis of the back phosphorylation assay rests on the presumption that certain common sites are preserved for in vivo and in vitrokinase reactions and multiple kinases do not converge on the exact same serine residues. Our results suggest that PKA likely phosphorylates NHE-3 in the cell. It is possible that the regulation of NHE-3 by cAMP involves phosphorylation of proteins other than NHE-3 by PKA and that PKA may activate other kinases that phosphorylate NHE-3. Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) showed that the lack of change in total NHE-3 protein phosphorylation in vivo is because of multiple constitutively phosphorylated sites masking the change in the regulated sites. We also found that phosphopeptides B, C, and D are largely unregulated in vivo. In contrast, phosphopeptide E (Ser-605) and J (Ser-552) were increased at 1 and 10 μm 8-Br-cAMP, respectively. The dose-response profiles of phosphopeptides E and J actually correlate well with that of changes in NHE-3 activity in response to cAMP.If phosphorylation of Ser-552 and Ser-605 are important, then elimination of phosphorylation should abolish functional regulation. All the single and double point mutants showed comparable antigen expression and baseline transport activity but the quadruple mutants showed decreased and the six-point mutant showed markedly attenuated expression of both NHE-3 protein and activity. Because NHE-3 transcript levels are not affected by the mutations, one has to postulate that the multiple serine mutations significantly affected translation and/or protein half-life. This study does not address the mechanism of these changes. Ser-552 and Ser-605 were both phosphorylated in vivo and when mutated singly to nonphosphorylatable residues, functional regulation by PKA was abrogated in both mutants. Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) have reported the importance of Ser-605 and Ser-634 in the functional regulation of NHE-3 by PKA although Ser-634 appeared not to be a phosphoserine. Our phosphopeptide map and functional response of the NHE-3mut/S605A is very similar to the previous report. A major disparate finding is that we found Ser-552 to be critical for PKA regulation, whereas Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) found normal functional regulation of NHE-3mut/S552A by cAMP. In addition, Cabado and co-workers (11Cabado A.G. Yu F.H. Kapus A. Lukacs G. Grinstein S. Orlowski J. J. Biol Chem. 1996; 271: 3590-3599Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar) from the same group found that truncation of the cytoplasmic domain at amino acid 579 rendered NHE-3 nonresponsive to cAMP. The difference may reside in the fact that Ser-552, which is phosphorylated in our cells, was not phosphorylated in the previous report as is evident by the absence of phosphopeptide J (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar). It is conceivable that multiple PKA target phosphoserines may participate in mediating the functional inhibition of NHE-3 and different host cells may use different sets of serines. The basis for the differential NHE-3 phosphorylation is unclear, because both laboratories used AP-1 cells from the same source. Because AP-1 cells are mutants derived and selected from the parental Chinese hamster ovary cell, repeated passages and clonal expansion might have created different phenotypes.One possible interpretation for the lack of regulation of NHE-3mut/S552A is that some cellular cofactor(s) are missing in that particular AP-1 recipient that harbors the mutant NHE-3 and that the cAMP-insensitive phenotype stems from the cell rather than NHE-3mut/S552A. Such an explanation is unlikely for several reasons. All pooled transfectants of single or multiple mutant NHE-3s were generated about the same time from the same parental AP-1 cells. In addition, when we tried to examined NHE-3mut/S552A in as heterogenous an AP-1 cellular background as we can create, we found that the cAMP-nonresponsive phenotype prevailed.The lack of regulation of NHE-3mut/S575/S661/S690/S804 is harder to explain because none of the serines were functionally important when examined as single mutants. One can speculate on several explanations. One is that there are multiple regulatory phosphoserines with P-Ser-552 and P-Ser-605 having the most dominant effects with lesser contribution from the other phosphoserines. Although singular mutations of serines other than Ser-552 and Ser-605 do not affect the overall response, mutation of four minor phosphorylation sites may be sufficient. A second possibility is that these serines participate in specific protein-protein interactions with regulatory factors such as NHE-RF (20Weinman E.J. Steplock D. Shenolikar S. J. Clin. Invest. 1993; 92: 1781-1786Crossref PubMed Scopus (118) Google Scholar, 21Weinman E.J. Steplock D. Wang Y. Shenolikar S. J. Clin. Invest. 1995; 95: 2143-2149Crossref PubMed Scopus (310) Google Scholar, 29Weinman E.J. Dubinsky W.P. Dinh Q. Steplock D. Shenolikar S. J. Membr. Biol. 1989; 109: 233-241Crossref PubMed Scopus (28) Google Scholar, 30Morell G. Steplock D. Shenolikar S. Weinman E.J. Am. J. Physiol. 1990; 259: F867-F871PubMed Google Scholar, 31Weinman E.J. Steplock D. Bui G. Yuan N. Shenolikar S. Am. J. Physiol. 1998; 258: F1254-F1258Google Scholar), E3KARP (22Yun C.H.C. Oh S. Zizak M. Steplock D. Taso S. Tse C.M. Weinman E.J. Donowitz M. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 3010-3030Crossref PubMed Scopus (404) Google Scholar), or calmodulin (32Levine S.A. Nath S.K. Yun C.H.C. Yip J.W. Montrose M. Donowitz M. Tse C.M. J. Biol. Chem. 1995; 270: 13716-13725Crossref PubMed Scopus (103) Google Scholar) that modulate NHE-3 function, and this interaction is disrupted by the mutations. A third possibility is that the mutations distort the protein structure to such an extent that it nonspecifically alters NHE-3 function and its regulation. Crystallographic evidence indicates that one can significantly alter the protein structure of a domain of a protein by a single amino acid mutation from tyrosine to phenylalanine, which essentially only removes a single hydroxyl group (33Zhang J. Zhang F. Ebert D. Cobb M.H. Goldsmith E.J. Structure. 1995; 3: 299-307Abstract Full Text Full Text PDF PubMed Scopus (109) Google Scholar). The present paper does not distinguish the three stated possibilities. We did however observe intact regulation of all mutants including the four- to six-point mutants by acute hyperosmolality (25–40% inhibition) (data not shown). This is compatible with the ability of hyperosmolality to inhibit NHE-3 activity despite radical truncations of the cytoplasmic domain (34Nath S.K. Hang C.Y. Levine S.A. Yun C.H.C. Montrose M.H. Donowitz M. Tse C.M. Am. J. Physiol. 1996; 270: G431-G441PubMed Google Scholar).In summary, we hypothesize that the functional regulation of NHE-3 by PKA does not converge on single covalent modifications such as phosphorylation of a specific serine. We submit a model where PKA phosphorylates NHE-3 on Ser-552 and Ser-605 in addition to other serines. Phosphorylation of Ser-552 and Ser-605 each play a major although not exclusive role in the functional regulation of NHE-3, which involves specific interaction of various regions of the cytoplasmic domain of NHE-3 with other regulatory cofactors. We confirmed the previous finding of inhibition ofV max by cAMP in native NHE-3 in renal epithelial cells (24Pollock A.S. Warnock D.G. Strewler G.J. Am. J. Physiol. 1986; 250: F217-F225PubMed Google Scholar, 25Casavola V.C. Helmle-Kolb C. Murer H. Biochem. Biophys. Res. Commun. 1989; 165: 833-837Crossref PubMed Scopus (45) Google Scholar, 26Cano A. Preisig P.A. Alpern R.J. J. Clin. Invest. 1993; 92: 1632-1638Crossref PubMed Scopus (46) Google Scholar) and NHE-3 heterologously expressed in fibroblasts (8Kandasamy R.A. Yu F.H. Harris R. Boucher A. Hanrahan J.W. Orlowski J. J. Biol. Chem. 1995; 270: 29209-29216Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar, 9Azarani A. Goltzman D. Orlowski J. J. Biol. Chem. 1995; 270: 20004-20010Abstract Full Text Full Text PDF PubMed Scopus (92) Google Scholar, 10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar, 11Cabado A.G. Yu F.H. Kapus A. Lukacs G. Grinstein S. Orlowski J. J. Biol Chem. 1996; 271: 3590-3599Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar, 12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar). The lack of detectable NHE-3 whole protein phosphorylation is comparable with the report by Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) although we observed discernible but variable changes in NHE-3 whole protein phosphorylation with 100 μm 8-Br-cAMP. As some cation channels are directly gated by cyclic nucleotides (19Kaupp U.B. Curr. Opin. Neurobiol. 1995; 5: 434-442Crossref PubMed Scopus (130) Google Scholar), we queried if such can be a model for NHE-3 regulation by cAMP. When we inactivated the catalytic subunit of PKA, 8-Br-cAMP no longer regulated NHE-3 activity indicating that the cAMP effect is kinase-dependent. Also congruent with A kinase-mediated action is the fact that unlike the cyclic nucleotide-gated channels (27Goulding E.H. Tibbs G.R. Siegelbaum S.A. Nature. 1994; 372: 369-374Crossref PubMed Scopus (179) Google Scholar), the primary sequence of NHE-3 does not conform to the β-roll/α-helix motif derived from the catabolite gene activator protein model (28McKay D.B. Stietz T.A. Nature. 1981; 290: 744-749Crossref PubMed Scopus (460) Google Scholar). In addition, a recombinant fusion protein of maltose-binding protein/NHE-3 cytoplasmic domain failed to bind3H-cAMP (data not shown). The cytoplasmic domain of NHE-3 contains numerous PKA consensus target sites and NHE-3 is rapidly phosphorylated by PKA in vitro(10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar). Although NHE-3 is a phosphoprotein in vivo (10Moe O.W. Amemiya M. Yamaji Y. J. Clin. Invest. 1995; 96: 2187-2194Crossref PubMed Scopus (89) Google Scholar, 12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar), there is no a priori reason to assume that PKA directly phosphorylates NHE-3 in vivo. The basis of the back phosphorylation assay rests on the presumption that certain common sites are preserved for in vivo and in vitrokinase reactions and multiple kinases do not converge on the exact same serine residues. Our results suggest that PKA likely phosphorylates NHE-3 in the cell. It is possible that the regulation of NHE-3 by cAMP involves phosphorylation of proteins other than NHE-3 by PKA and that PKA may activate other kinases that phosphorylate NHE-3. Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) showed that the lack of change in total NHE-3 protein phosphorylation in vivo is because of multiple constitutively phosphorylated sites masking the change in the regulated sites. We also found that phosphopeptides B, C, and D are largely unregulated in vivo. In contrast, phosphopeptide E (Ser-605) and J (Ser-552) were increased at 1 and 10 μm 8-Br-cAMP, respectively. The dose-response profiles of phosphopeptides E and J actually correlate well with that of changes in NHE-3 activity in response to cAMP. If phosphorylation of Ser-552 and Ser-605 are important, then elimination of phosphorylation should abolish functional regulation. All the single and double point mutants showed comparable antigen expression and baseline transport activity but the quadruple mutants showed decreased and the six-point mutant showed markedly attenuated expression of both NHE-3 protein and activity. Because NHE-3 transcript levels are not affected by the mutations, one has to postulate that the multiple serine mutations significantly affected translation and/or protein half-life. This study does not address the mechanism of these changes. Ser-552 and Ser-605 were both phosphorylated in vivo and when mutated singly to nonphosphorylatable residues, functional regulation by PKA was abrogated in both mutants. Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) have reported the importance of Ser-605 and Ser-634 in the functional regulation of NHE-3 by PKA although Ser-634 appeared not to be a phosphoserine. Our phosphopeptide map and functional response of the NHE-3mut/S605A is very similar to the previous report. A major disparate finding is that we found Ser-552 to be critical for PKA regulation, whereas Kurashima and co-workers (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar) found normal functional regulation of NHE-3mut/S552A by cAMP. In addition, Cabado and co-workers (11Cabado A.G. Yu F.H. Kapus A. Lukacs G. Grinstein S. Orlowski J. J. Biol Chem. 1996; 271: 3590-3599Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar) from the same group found that truncation of the cytoplasmic domain at amino acid 579 rendered NHE-3 nonresponsive to cAMP. The difference may reside in the fact that Ser-552, which is phosphorylated in our cells, was not phosphorylated in the previous report as is evident by the absence of phosphopeptide J (12Kurashima K. Yu F.H. Cabado A.G. Szabó E.Z. Grinstein S. Orlowski J. J. Biol. Chem. 1997; 272: 28672-28679Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar). It is conceivable that multiple PKA target phosphoserines may participate in mediating the functional inhibition of NHE-3 and different host cells may use different sets of serines. The basis for the differential NHE-3 phosphorylation is unclear, because both laboratories used AP-1 cells from the same source. Because AP-1 cells are mutants derived and selected from the parental Chinese hamster ovary cell, repeated passages and clonal expansion might have created different phenotypes. One possible interpretation for the lack of regulation of NHE-3mut/S552A is that some cellular cofactor(s) are missing in that particular AP-1 recipient that harbors the mutant NHE-3 and that the cAMP-insensitive phenotype stems from the cell rather than NHE-3mut/S552A. Such an explanation is unlikely for several reasons. All pooled transfectants of single or multiple mutant NHE-3s were generated about the same time from the same parental AP-1 cells. In addition, when we tried to examined NHE-3mut/S552A in as heterogenous an AP-1 cellular background as we can create, we found that the cAMP-nonresponsive phenotype prevailed. The lack of regulation of NHE-3mut/S575/S661/S690/S804 is harder to explain because none of the serines were functionally important when examined as single mutants. One can speculate on several explanations. One is that there are multiple regulatory phosphoserines with P-Ser-552 and P-Ser-605 having the most dominant effects with lesser contribution from the other phosphoserines. Although singular mutations of serines other than Ser-552 and Ser-605 do not affect the overall response, mutation of four minor phosphorylation sites may be sufficient. A second possibility is that these serines participate in specific protein-protein interactions with regulatory factors such as NHE-RF (20Weinman E.J. Steplock D. Shenolikar S. J. Clin. Invest. 1993; 92: 1781-1786Crossref PubMed Scopus (118) Google Scholar, 21Weinman E.J. Steplock D. Wang Y. Shenolikar S. J. Clin. Invest. 1995; 95: 2143-2149Crossref PubMed Scopus (310) Google Scholar, 29Weinman E.J. Dubinsky W.P. Dinh Q. Steplock D. Shenolikar S. J. Membr. Biol. 1989; 109: 233-241Crossref PubMed Scopus (28) Google Scholar, 30Morell G. Steplock D. Shenolikar S. Weinman E.J. Am. J. Physiol. 1990; 259: F867-F871PubMed Google Scholar, 31Weinman E.J. Steplock D. Bui G. Yuan N. Shenolikar S. Am. J. Physiol. 1998; 258: F1254-F1258Google Scholar), E3KARP (22Yun C.H.C. Oh S. Zizak M. Steplock D. Taso S. Tse C.M. Weinman E.J. Donowitz M. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 3010-3030Crossref PubMed Scopus (404) Google Scholar), or calmodulin (32Levine S.A. Nath S.K. Yun C.H.C. Yip J.W. Montrose M. Donowitz M. Tse C.M. J. Biol. Chem. 1995; 270: 13716-13725Crossref PubMed Scopus (103) Google Scholar) that modulate NHE-3 function, and this interaction is disrupted by the mutations. A third possibility is that the mutations distort the protein structure to such an extent that it nonspecifically alters NHE-3 function and its regulation. Crystallographic evidence indicates that one can significantly alter the protein structure of a domain of a protein by a single amino acid mutation from tyrosine to phenylalanine, which essentially only removes a single hydroxyl group (33Zhang J. Zhang F. Ebert D. Cobb M.H. Goldsmith E.J. Structure. 1995; 3: 299-307Abstract Full Text Full Text PDF PubMed Scopus (109) Google Scholar). The present paper does not distinguish the three stated possibilities. We did however observe intact regulation of all mutants including the four- to six-point mutants by acute hyperosmolality (25–40% inhibition) (data not shown). This is compatible with the ability of hyperosmolality to inhibit NHE-3 activity despite radical truncations of the cytoplasmic domain (34Nath S.K. Hang C.Y. Levine S.A. Yun C.H.C. Montrose M.H. Donowitz M. Tse C.M. Am. J. Physiol. 1996; 270: G431-G441PubMed Google Scholar). In summary, we hypothesize that the functional regulation of NHE-3 by PKA does not converge on single covalent modifications such as phosphorylation of a specific serine. We submit a model where PKA phosphorylates NHE-3 on Ser-552 and Ser-605 in addition to other serines. Phosphorylation of Ser-552 and Ser-605 each play a major although not exclusive role in the functional regulation of NHE-3, which involves specific interaction of various regions of the cytoplasmic domain of NHE-3 with other regulatory cofactors. We are grateful to Dr. Stan McKnight for providing reagents, to Dr. Melanie Cobb and Dr. Robert Alpern for helpful discussions, and to Dr. Michel Baum and Dr. Robert Alpern for careful reading of the manuscript.
Referência(s)