Plasma Membrane Ca2+-ATPase Isoforms 2b and 4b Interact Promiscuously and Selectively with Members of the Membrane-associated Guanylate Kinase Family of PDZ (PSD95/Dlg/ZO-1) Domain-containing Proteins
2001; Elsevier BV; Volume: 276; Issue: 24 Linguagem: Inglês
10.1074/jbc.m101448200
ISSN1083-351X
AutoresSteven J. DeMarco, Emanuel E. Strehler,
Tópico(s)Vitamin K Research Studies
ResumoSpatial and temporal regulation of intracellular Ca2+ signaling depends on localized Ca2+ microdomains containing the requisite molecular components for Ca2+ influx, efflux, and signal transmission. Plasma membrane Ca2+-ATPase (PMCA) isoforms of the "b" splice type contain predicted PDZ (PSD95/Dlg/ZO-1) interaction domains. The COOH-terminal tail of PMCA2b isolated the membrane-associated guanylate kinase (MAGUK) protein SAP97/hDlg as a binding partner in a yeast two-hybrid screen. The related MAGUKs SAP90/PSD95, PSD93/chapsyn-110, SAP97, and SAP102 all bound to the COOH-terminal tail of PMCA4b, whereas only the first three bound to the tail of PMCA2b. Coimmunoprecipitations confirmed the interaction selectivity between PMCA4b and SAP102 as opposed to the promiscuity of PMCA2b and 4b in interacting with other SAPs. Confocal immunofluorescence microscopy revealed the exclusive presence and colocalization of PMCA4b and SAP97 in the basolateral membrane of polarized Madin-Darby canine kidney epithelial cells. In hippocampal neurons, PMCA2b was abundant throughout the somatodendritic compartment and often extended into the neck and head of individual spines where it colocalized with SAP90/PSD95. These data show that PMCA "b" splice forms interact promiscuously but also with specificity with different members of the PSD95 family of SAPs. PMCA-SAP interactions may play a role in the recruitment and maintenance of the PMCA at specific membrane domains involved in local Ca2+ regulation. Spatial and temporal regulation of intracellular Ca2+ signaling depends on localized Ca2+ microdomains containing the requisite molecular components for Ca2+ influx, efflux, and signal transmission. Plasma membrane Ca2+-ATPase (PMCA) isoforms of the "b" splice type contain predicted PDZ (PSD95/Dlg/ZO-1) interaction domains. The COOH-terminal tail of PMCA2b isolated the membrane-associated guanylate kinase (MAGUK) protein SAP97/hDlg as a binding partner in a yeast two-hybrid screen. The related MAGUKs SAP90/PSD95, PSD93/chapsyn-110, SAP97, and SAP102 all bound to the COOH-terminal tail of PMCA4b, whereas only the first three bound to the tail of PMCA2b. Coimmunoprecipitations confirmed the interaction selectivity between PMCA4b and SAP102 as opposed to the promiscuity of PMCA2b and 4b in interacting with other SAPs. Confocal immunofluorescence microscopy revealed the exclusive presence and colocalization of PMCA4b and SAP97 in the basolateral membrane of polarized Madin-Darby canine kidney epithelial cells. In hippocampal neurons, PMCA2b was abundant throughout the somatodendritic compartment and often extended into the neck and head of individual spines where it colocalized with SAP90/PSD95. These data show that PMCA "b" splice forms interact promiscuously but also with specificity with different members of the PSD95 family of SAPs. PMCA-SAP interactions may play a role in the recruitment and maintenance of the PMCA at specific membrane domains involved in local Ca2+ regulation. intracellular free calcium concentration plasma membrane Ca2+-ATPase 95-kDa protein of the postsynaptic density PSD95/Dlg/ZO-1 membrane-associated guanylate kinase synapse-associated protein glutathioneS-transferase Madin-Darby canine kidney Dulbecco's phosphate-buffered saline Calcium ion (Ca2+) homeostasis is crucial for cell function and survival (1Berridge M.J. Bootman M.D. Lipp P. Nature. 1998; 395: 645-648Crossref PubMed Scopus (1807) Google Scholar). A finely controlled system of Ca2+ transporters, channels, and Ca2+-binding proteins allows for transient increases in the intracellular free calcium concentration ([Ca2+] i ),1while over the long term maintaining a low resting [Ca2+] i (2Brini M. Carafoli E. Cell Mol. Life Sci. 2000; 57: 354-370Crossref PubMed Scopus (155) Google Scholar). The exquisite specificity of Ca2+ signaling mandates that both the entry and the removal of Ca2+ are under precise temporal and spatial control (3Berridge M.J. Dupont G. Curr. Opin. Cell Biol. 1994; 6: 267-274Crossref PubMed Scopus (199) Google Scholar, 4Thomas A.P. Bird G.S.J. Hajnoczky G. Robb-Gaspers L.D. Putney Jr., J.W. FASEB J. 1996; 10: 1505-1517Crossref PubMed Scopus (424) Google Scholar). Accordingly, the molecular machinery involved in local Ca2+ signaling must be assembled, maintained, and regulated with the requisite spatial and temporal resolution. Mechanisms that increase local [Ca2+] i have been studied extensively over the last few years; consequently, significant progress has been made in understanding the regulation and targeting of calcium channels (5Blackstone C. Sheng M. Cell Calcium. 1999; 26: 181-192Crossref PubMed Scopus (33) Google Scholar). By contrast, much less is known about the spatial organization of Ca2+ extrusion mechanisms, specifically that provided by plasma membrane Ca2+-ATPases (PMCAs). These primary ion pumps are essential for the long term maintenance of low intracellular Ca2+ but more recently have also been implicated in dynamic events such as the regulation of Ca2+spikes and local Ca2+ signaling (for review, see Refs.6Monteith G.R. Roufogalis B.D. Cell Calcium. 1995; 18: 459-470Crossref PubMed Scopus (120) Google Scholar, 7Penniston J.T. Enyedi A. J. Membr. Biol. 1998; 165: 101-109Crossref PubMed Scopus (159) Google Scholar, 8Garcia M.L. Strehler E.E. Front. Biosci. 1999; 4: 869-882Crossref PubMed Google Scholar). A multigene family of four non-allelic members encodes four conserved mammalian PMCA isoforms (designated PMCA1–4), with additional diversity generated by alternative mRNA splicing affecting the protein at two major locations (9Strehler E.E. Zacharias D.A. Physiol. Rev. 2001; 81: 21-50Crossref PubMed Scopus (486) Google Scholar). The four PMCA gene products do not differ greatly in their overall tertiary structure, which includes 10 predicted transmembrane spans, intracellular NH2 and COOH termini, and a large cytosolic catalytic loop between membrane spans 4 and 5 (10Carafoli E. FASEB J. 1994; 8: 993-1002Crossref PubMed Scopus (367) Google Scholar). However, there are substantial differences in their regulation by kinases, proteases, and the Ca2+-binding protein calmodulin. Moreover, the major PMCA variants "a" and "b" generated by alternative splicing in the COOH-terminal coding region differ markedly in their regulatory properties, most notably in calmodulin sensitivity (7Penniston J.T. Enyedi A. J. Membr. Biol. 1998; 165: 101-109Crossref PubMed Scopus (159) Google Scholar). The alternative splice affects the pump protein after the last transmembrane domain and creates different COOH-terminal amino acid sequences for the "a" and "b" variants because of a change in translational reading frame. The last few residues of all "b" variants are highly conserved, and the final four residues of PMCA4b match the minimal consensus sequence (E-T/S-X-V*, where the asterisk indicates the COOH-terminal residue) of protein ligands for type I PDZ (PSD95/Dlg/ZO-1) domains (11Kim E. DeMarco S.J. Marfatia S.M. Chishti A.H. Sheng M. Strehler E.E. J. Biol. Chem. 1998; 273: 1591-1595Abstract Full Text Full Text PDF PubMed Scopus (145) Google Scholar). PDZ domains are present in a large variety of different proteins and constitute a ubiquitous protein-protein interaction motif (12Ponting C.P. Protein Sci. 1997; 6: 464-468Crossref PubMed Scopus (196) Google Scholar, 13Fanning A.S. Anderson J.M. Pawson A.J. Protein Modules in Signal Transduction. 228. Springer Verlag, Berlin1998: 209-233Google Scholar). Indeed, we reported previously that PMCA4b was able to interact with high affinity with the PDZ domains of several members of the MAGUK (membrane-associated guanylate kinase) protein family, and that this interaction was dependent on the presence of the T-S-V* COOH-terminal sequence (11Kim E. DeMarco S.J. Marfatia S.M. Chishti A.H. Sheng M. Strehler E.E. J. Biol. Chem. 1998; 273: 1591-1595Abstract Full Text Full Text PDF PubMed Scopus (145) Google Scholar). MAGUKs are multimodular PDZ domain-containing proteins implicated in the formation and maintenance of specialized cell-cell junctions and in signaling processes at the cell membrane (14Kim S.K. Curr. Opin. Cell Biol. 1995; 7: 641-649Crossref PubMed Scopus (99) Google Scholar, 15Anderson J.M. Curr. Biol. 1996; 6: 382-384Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar). A subfamily of the MAGUKs are PSD95 (95-kDa protein of the postsynaptic density)-like molecules, also referred to as SAPs (synapse-associated proteins). SAPs are found throughout the brain, and some are also present in other tissues such as gut and kidney epithelium (14Kim S.K. Curr. Opin. Cell Biol. 1995; 7: 641-649Crossref PubMed Scopus (99) Google Scholar, 15Anderson J.M. Curr. Biol. 1996; 6: 382-384Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar, 16Gomperts S.N. Cell. 1996; 84: 659-662Abstract Full Text Full Text PDF PubMed Scopus (224) Google Scholar, 17Sheng M. Neuron. 1996; 17: 575-578Abstract Full Text Full Text PDF PubMed Scopus (296) Google Scholar, 18Garner C.C. Kindler S. Trends Cell Biol. 1996; 6: 429-433Abstract Full Text PDF PubMed Scopus (63) Google Scholar, 19Kornau H.-C. Seeburg P.H. Kennedy M.B. Curr. Opin. Neurobiol. 1997; 7: 368-373Crossref PubMed Scopus (313) Google Scholar). Their precise function is not well understood, but they are thought to act as scaffolding and/or clustering proteins for various membrane receptors and transporters. In addition, they recruit regulatory and signaling proteins to the membrane and may thus be required for the formation of multiprotein signaling complexes at specific membrane domains (for review, see Refs. 16Gomperts S.N. Cell. 1996; 84: 659-662Abstract Full Text Full Text PDF PubMed Scopus (224) Google Scholar, 17Sheng M. Neuron. 1996; 17: 575-578Abstract Full Text Full Text PDF PubMed Scopus (296) Google Scholar, 18Garner C.C. Kindler S. Trends Cell Biol. 1996; 6: 429-433Abstract Full Text PDF PubMed Scopus (63) Google Scholar, 19Kornau H.-C. Seeburg P.H. Kennedy M.B. Curr. Opin. Neurobiol. 1997; 7: 368-373Crossref PubMed Scopus (313) Google Scholar, 20Kim S.K. Curr. Opin. Cell Biol. 1997; 9: 853-859Crossref PubMed Scopus (106) Google Scholar, 21Garner C.C. Nash J. Huganir R.L. Trends Cell Biol. 2000; 10: 274-280Abstract Full Text Full Text PDF PubMed Scopus (482) Google Scholar). Several ion channels and receptor proteins have been identified as binding targets of the SAPs, includingN-methyl-d-aspartate receptor subunits and K+ channels (16Gomperts S.N. Cell. 1996; 84: 659-662Abstract Full Text Full Text PDF PubMed Scopus (224) Google Scholar, 17Sheng M. Neuron. 1996; 17: 575-578Abstract Full Text Full Text PDF PubMed Scopus (296) Google Scholar, 22Kim E. Niethammer M. Rothschild A. Jan Y.N. Sheng M. Nature. 1995; 378: 85-88Crossref PubMed Scopus (904) Google Scholar, 23Kornau H.C. Schenker L.T. Kennedy M.B. Seeburg P.H. Science. 1995; 269: 1737-1740Crossref PubMed Scopus (1645) Google Scholar). Here, we used the PMCA2b COOH-terminal sequence as bait in an unbiased yeast two-hybrid screen of a human brain cDNA library and isolated a clone coding for SAP97 as one of the specific interactors. To address the promiscuity versus specificity in the interaction of PMCA2b and PMCA4b with different SAPs, we used pull-down and coimmunoprecipitation assays to analyze their interaction with the four SAPs SAP90/PSD95, SAP93/chapsyn-110, SAP97/hDlg, and SAP102. The results show that PMCA2b and 4b interact promiscuously with all SAPs except for SAP102 which binds to PMCA4b but not to PMCA2b. Using confocal immunofluorescence microscopy, we demonstrate colocalization of PMCA4b and SAP97 at the basolateral membrane of MDCK epithelial cells, and of PMCA2b and SAP90/PSD95 at most, but not all, synaptic spines in hippocampal neurons. A direct, PDZ domain-mediated interaction with SAPs may be an attractive means to localize PMCA isoforms to specific membrane domains and to recruit them into multiprotein Ca2+ signaling complexes. Plasmids encoding fusion proteins of the PMCA carboxyl-terminal sequences were made by standard molecular biology techniques using polymerase chain reaction or suitable restriction enzyme digests. Codons for the final 72 (PMCA2b) or 71 (PMCA4b) amino acids were cloned as EcoRI-BamHI fragments into pAS2–1 (CLONTECH, Palo Alto, CA) to produce the DNA binding fusions (pDB-CT2b and pDB-CT4b) for yeast two-hybrid screens and assays. The same codons were inserted into the pGEX-2TK (Amersham Pharmacia Biotech) vector to produce GST fusions (GST-2b and GST-4b). Plasmid pMM2-PMCA4b for expression of full-length human PMCA4b in mammalian cells has been described previously (24Adamo H.P. Verma A.K. Sanders M.A. Heim R. Salisbury J.L. Wieben E.D. Penniston J.T. Biochem. J. 1992; 285: 791-797Crossref PubMed Scopus (57) Google Scholar). A vector expressing full-length human PMCA2b was assembled from overlapping partial PMCA2 cDNAs (25Heim R. Hug M. Iwata T. Strehler E.E. Carafoli E. Eur. J. Biochem. 1992; 205: 333-340Crossref PubMed Scopus (55) Google Scholar) using a combination of restriction digests and polymerase chain reaction. The final PMCA2b cDNA fragment was cloned as anSalI-MluI fragment into the modified pMM2 expression vector (24Adamo H.P. Verma A.K. Sanders M.A. Heim R. Salisbury J.L. Wieben E.D. Penniston J.T. Biochem. J. 1992; 285: 791-797Crossref PubMed Scopus (57) Google Scholar) to create plasmid pMM2-PMCA2b. The integrity of all final constructs was confimed by DNA sequencing in the Mayo Clinic Molecular Biology Core Facility. Mammalian expression constructs for SAP93/chapsyn-110 and SAP102 were kind gifts from Morgan Sheng (Harvard Medical School, Boston) and Craig Garner (University of Alabama, Birmingham), respectively. Yeast two-hybrid screening was performed according to the instructions for the Matchmaker Two-Hybrid System 2 (CLONTECH). Yeast of strain CG1945 were cotransformed using the lithium acetate method (26Gietz R.D. Schiestl R.H. Methods Mol. Cell Biol. 1995; 5: 255-269Google Scholar) with bait plasmid pDB-CT2b and a human brain cDNA library (CLONTECH) made in the vector pACT2. Approximately 5 × 106 independent cDNA clones were screened and assayed for HIS3 and β-galactosidase expression. Selection of initial positives was done after 7 days of growth on SD/-Trp/-Leu/-His agar plates containing 5 mm3-aminotriazole (Sigma). Positive clones were propagated on SD/-Trp/-Leu medium. β-Galactosidase assays were performed after freeze-thaw of yeast colonies on filter lifts of 3-day-old streak plates. The bait plasmid was dropped out by 1 μg/ml cycloheximide counterselection, and growth on -Leu medium. After bait dropout, plasmid inserts from positive yeast clones were amplified directly after picking 3-day-old yeast colonies into TE (10 mmTris-HCl, pH 7.5, 1 mm EDTA) containing 0.25 unit/μl lyticase (Sigma). After 30 min at 37 °C, the yeast were subjected to one round of freeze-thaw followed by alkaline lysis and phenol/chloroform extraction. The aqueous fraction was precipitated with 1/10 volume of 5 m sodium acetate, pH 5.3, and 0.7 volume isopropyl alcohol; the nucleic acid pellet was washed in 70% ethanol and dissolved in 10 mm Tris, pH 8.0. 1% of the dissolved DNA was used as template for polymerase chain reaction with pACT2-specific primers. Library plasmids were prepared from counterselected yeast containing only "prey" plasmids by using the above alkaline lysis method (27Wach A. Pick H. Philippsen P. Johnston J.R. Molecular Genetics of Yeast: A Practical Approach. Oxford University Press, New York1994: 1-16Google Scholar) and transformed directly intoEscherichia coli strain DH5α by electroporation. GST and GST fusion proteins were expressed as described (28Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Seidman J.G. Smith J.A. Struhl K. Current Protocols in Molecular Biology. John Wiley and Sons, New York1998Google Scholar) in E. coliBL21(DE3) upon induction with 0.7 mmisopropyl-1-thio-β-d-galactopyranoside for 4 h. Cells were pelleted, resuspended in TBS (50 mm Tris-HCl, pH 7.4, 150 mm NaCl) plus protease inhibitors (0.2 mm phenylmethylsulfonyl fluoride, 1 μg/ml pepstatin A, 2 μg/ml leupeptin, 0.2 μg/ml aprotinin, 10 mm EDTA), and 30 mm β-mercaptoethanol, and lysed by the addition of sarkosyl (Curtis-Matheson Scientific, Houston, TX) to a final concentration of 1.5%. After 15 min on ice, the lysate was cleared by centrifugation at 10,000 × g and supplemented by the addition of Triton X-100 to 2%. This lysate was then bound to glutathione-Sepharose (Sigma) and washed with TBST (TBS + 0.1% Tween), and TBS. The quantity of bound fusion protein was estimated by Coomassie Blue staining of SDS-polyacrylamide gels of known amounts of fusion protein-containing glutathione-Sepharose beads. All fusion proteins were adjusted to ∼0.5 mg/ml for pull-down assays. Brains were removed from male Harlan Sprague-Dawley rats (∼250 g) and were immediately homogenized using a glass-Teflon homogenizer in cold 10 mm Tris-HCl, pH 7.4, containing 0.3 m sucrose, 20 mm EDTA, 10 mm EGTA, 75 mm NaCl (10 ml/brain), and a mixture of protease inhibitors. After ∼20 strokes, the homogenate was centrifuged at 3,700 × g for 2 min. The supernatant was subjected to high speed centrifugation at 150,000 ×g to pellet membranes. Membranes were then solubilized in SDS buffer (50 mm Tris-HCl, pH 7.4, 5 mm EDTA, 5 mm EGTA, 50 mm NaCl, and 1% SDS), for 20 min at 55 °C. SDS in the membrane extract was neutralized in 4 volumes of cold 1% Triton X-100, 5 mm EDTA, 5 mm EGTA, and the extract was chilled on ice for 10 min before centrifuging at 20,000 × g for 30 min. 5 μg of GST alone or of GST fusion proteins on agarose beads was rocked overnight with 750 μl of cleared brain extract. The beads were pelleted and washed three times in TBS + 1% Triton X-100. Bound proteins were eluted in Laemmli buffer (28Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Seidman J.G. Smith J.A. Struhl K. Current Protocols in Molecular Biology. John Wiley and Sons, New York1998Google Scholar) and separated on 10% polyacrylamide gels followed by transfer to nitrocellulose after standard Western blotting procedures (28Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Seidman J.G. Smith J.A. Struhl K. Current Protocols in Molecular Biology. John Wiley and Sons, New York1998Google Scholar). Nitrocellulose membranes were blocked in TBST + 5% milk before immunoblotting with appropriate primary and secondary antibodies. All secondary antibodies on immunoblots were detected using RenaissanceTM chemiluminescent reagent (NEN Life Science Products). COS-1 cells were grown to ∼80% confluence on six-well plates (Costar) in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, nonessential amino acids, 1 mm sodium pyruvate, glutamine, and antibiotic/antimycotic mixture (all cell culture reagents were purchased from Life Technologies, Inc.). Cells were transfected with 2 μg of total DNA using LipofectAMINE according to the manufacturer's instructions (Life Technologies, Inc.). After ∼48 h, the cells were rinsed with cold D-PBS (Ca2+- and Mg2+-free) and lysed in a buffer containing 50 mm HEPES at pH 7.4, 150 mm NaCl, 1% Nonidet P-40, 0.5% sodium deoxycholate, and protease inhibitors. After 10 min on ice the cells were scraped from the plates using a rubber policeman and were centrifuged at 13,000 × g for 10 min at 4 °C. Half of the lysate (500 μl) was used for each immunoprecipitation, to which 2–3 μl of antibody (anti-PMCA monoclonal 5F10) was added. After 2 h of rocking at 4 °C, 50 μl of protein A/G-agarose was added to each mixture, and rocking continued overnight at 4 °C. Protein A/G-agarose was pelleted at 4,000 × g for 30 s and quickly washed three times in cold TBST. Bound proteins were eluted in Laemmli buffer (28Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Seidman J.G. Smith J.A. Struhl K. Current Protocols in Molecular Biology. John Wiley and Sons, New York1998Google Scholar). All of the bound protein and 5% of the starting lysate were separated on 10% polyacrylamide gels followed by transfer to nitrocellulose for Western blotting as described above. The following antibodies were obtained from the indicated source and used at the indicated dilution for immunoblotting. Anti-SAP90 was from Transduction Laboratories (Lexington, KY) and used at 1:400 dilution. Anti-SAP93 was obtained from David Bredt (University of California, San Francisco) and used at 1:1,000 dilution. Anti-SAP97 and anti-SAP102 were obtained from Craig Garner (University of Alabama, Birmingham), and both were used at dilutions of 1:2,000. Anti-PMCA2 (affinity-purified and concentrated NR2) and anti-PMCA4 (JA9) were obtained from John Penniston and Adelaida Filoteo (Mayo Clinic) and used at dilutions of 1:5,000 and 1:600, respectively. Secondary goat anti-mouse or goat anti-rabbit antibodies were purchased from Sigma and used at 1:5,000 dilution. Type I MDCK epithelial cells (ATCC CCL-34, Manassas, VA) were grown to confluence on glass coverslips in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum and supplemented with 1% antibiotic-antimycotic (Life Technologies, Inc.). The cells were then fixed for 5 min at room temperature in 4% paraformaldehyde (Tousimis, Rockville, MD) diluted in D-PBS + Ca2+/Mg2+ (D-PBS+CM). After five brief washes in D-PBS+CM, coverslips were further fixed and permeablized in prechilled methanol for 5 min at −20 °C. The cells were blocked in D-PBS+CM containing 5% normal goat serum and 1% bovine serum albumin and were then incubated for 1 h at room temperature with monoclonal pan-anti-PMCA antibody 5F10 (a gift from John Penniston and Adelaida Filoteo, Mayo Clinic) and polyclonal anti-SAP97 diluted 1:600 and 1:200, respectively, in blocking buffer. After washing 3 × 5 min in D-PBS+CM, the cells were incubated for 1 h at room temperature with anti-mouse Alexa 488 and anti-rabbit Alexa 594 (both from Molecular Probes, Eugene, OR), each diluted 1:600 in blocking buffer. Primary rat hippocampal neurons were prepared from 17/18-day-old embryos essentially as described (29Fitzsimonds R.M. Song H.-J. Poo M.-M. Nature. 1997; 388: 439-448Crossref PubMed Scopus (145) Google Scholar) and used at postnatal day 21–30. Cells grown on coverslips were fixed in 4% paraformaldehyde, 4% sucrose, D-PBS+CM for 20 min, washed, and incubated in blocking buffer as above, and stained with anti-PMCA2 antibody NR2 (diluted 1:800 in blocking buffer) and anti-SAP90 (1:250 dilution in blocking buffer) each for 1 h at room temperature. After washing three times for 5 min, the cells were incubated for 1 h at room temperature with anti-mouse Alexa 488 and anti-rabbit Alexa 594 secondary antibodies diluted 1:800 in blocking buffer. After secondary antibody incubation, the cells were washed five times for 10 min with D-PBS+CM. After the final washing, coverslips were mounted in Prolong mounting media (Molecular Probes). Confocal micrographs were taken on a Zeiss LSM 510 using an Apochromat 63× (MDCK), or Apochromat 100× (neurons) objective, and captured using LSM 510 software (Zeiss). Images were imported and edited using Adobe Photoshop 5.0. A sequence alignment of the extreme COOH termini of the PMCA "b" forms suggested that PMCA1b, 2b, and 3b would bind the same PDZ domains (Fig. 1 A), whereas PMCA4b would interact with perhaps a different set of PDZ domains. For this reason, we sought to identify proteins that would interact with the COOH-terminal sequence of PMCA1b/2b/3b, using PMCA2b as the representative isoform in an unbiased screen. A protein consisting of the Gal4 DNA binding domain fused to the final 72 amino acids of hPMCA2b (Fig. 1 B) was used as bait in a yeast two-hybrid screen of a human brain cDNA library. Screening 5 × 106 independent clones yielded ∼30 HIS3- and β-galactosidase-positive clones that were analyzed further. Sequencing of the cDNA inserts revealed that five clones specified proteins containing at least one PDZ domain. Of these, clone J5 was one of the strongest interactors (as determined by β-galacosidase assay; data not shown) and encoded a nearly full-length sequence (amino acids 242–926) of human SAP97/hDlg (Fig. 1 C). We speculated that some selectivity might exist in the interaction of these PMCAs with different members of the SAP90/PSD95 family of MAGUKs. To test this, we used GST fusion proteins containing the COOH termini of PMCA2b (CT2b) and PMCA4b (CT4b) in pull-down assays from rat brain extracts. Rat brain contains all of the SAP proteins (SAP90, SAP93, SAP97, SAP102) as well as all isoforms and most splice forms of the PMCAs (9Strehler E.E. Zacharias D.A. Physiol. Rev. 2001; 81: 21-50Crossref PubMed Scopus (486) Google Scholar, 18Garner C.C. Kindler S. Trends Cell Biol. 1996; 6: 429-433Abstract Full Text PDF PubMed Scopus (63) Google Scholar). As shown in Fig.2, there is clearly promiscuity between PMCA2b and 4b in their ability to bind different SAPs. However, there is also a distinct degree of binding specificity of these two PMCAs for some SAPs as illustrated most clearly for SAP102. Although the data in Fig. 2 indicate that SAPs generally interact with higher affinity with PMCA4b than with PMCA2b, the results also show that with the exception of SAP102, all are capable of binding PMCA2b in vitro. To determine if the promiscuity and selectivity of the interaction of PMCA2b and 4b with different members of the SAP family of MAGUKs are maintained when the full-length calcium pumps are coexpressed with these SAPs in an in vivo environment, we cotransfected COS-1 cells with plasmids encoding different combinations of the pumps and either SAP93 or SAP102. As predicted from the pull-down assays using the recombinant COOH-terminal tails of the PMCAs, antibodies against the PMCA were able to coimmunoprecipitate SAP93 together with either PMCA2b or PMCA4b (Fig. 3, left panel). By contrast, SAP102 only coimmunoprecipitated well with PMCA4b but not with PMCA2b (Fig. 3, right panel), confirming the selectivity of the interaction between PMCA4b and SAP102. Using the pull-down and coimmunoprecipitation data as a predictive guide, we sought mammalian tissues and cell types that coexpress SAP and PMCA family members to determine their cellular localization in vivo. In intestine and kidney epithelia, the PMCA has previously been shown to be predominantly localized at the basolateral membrane (30Borke J.L. Minami J. Verma A. Penniston J.T. Kumar R. J. Clin. Invest. 1987; 80: 1225-1231Crossref PubMed Scopus (102) Google Scholar, 31Borke J.L. Caride A. Verma A.K. Penniston J.T. Kumar R. Eur. J. Physiol. 1990; 417: 120-122Crossref PubMed Scopus (44) Google Scholar). We therefore used cultured MDCK cells as a representative epithelial cell type showing distinct apical and basolateral membrane domains. However, the isoform composition and subcellular distribution of PMCAs in MDCK cells had not yet been well documented. Immunoblotting of membrane proteins from MDCK cells indicated that PMCA4 is a major isoform in these cells. Using membrane protein from PMCA4b-overexpressing COS-1 cells as a sizing standard, we determined that the PMCA4 variant in MDCK cells is of the "b" splice form (Fig.4 A). This was confirmed independently by sequencing reverse transcription-polymerase chain reaction products from MDCK mRNA. 2S. Kip and E. E. Strehler, unpublished data. The anti-SAP97 antibody (a gift from C. C. Garner) recognized a single major protein of ∼140 kDa in Western blots of MDCK membranes (Fig.4 A). This corresponds to the size of SAP97 reported by other authors (32Müller B.M. Kistner U. Veh R.W. Cases-Langhoff C. Becker B. Gundelfinger E.D. Garner C.C. J. Neurosci. 1995; 15: 2354-2366Crossref PubMed Google Scholar), demonstrating that MDCK cells express endogenous SAP97. Using these antibodies, we detected strong basolateral membrane staining for both PMCA4b and SAP97 in polarized MDCK cells (Fig.4 B) by confocal microscopy. Sections through the cells in the x-z plane confirmed that the staining was exclusively basolateral for both proteins, with essentially complete overlap of fluorescence (Fig. 4 B). The same results were obtained in the human MCF7 cell line from breast epithelium (data not shown). These results demonstrate that in two different epithelial cell lines, PMCA4b and SAP97 coexist at the basolateral membranes. We next examined neurons from rat hippocampus because initial immunofluorescence suggested that PMCA2 was a major Ca2+ pump isoform in these cells. This is in agreement with previously published data on the PMCA isoform distribution in rodent and human hippocampus (33Stahl W.L. Eakin T.J. Owens J.W.M. Breininger J.F. Filuk P.E. Anderson W.R. Mol. Brain Res. 1992; 16: 223-231Crossref PubMed Scopus (98) Google Scholar, 34Zacharias D.A. Dalrymple S.J. Strehler E.E. Mol. Brain Res. 1995; 28: 263-272Crossref PubMed Scopus (44) Google Scholar, 35Stauffer T.P. Guerini D. Celio M.R. Carafoli E. Brain Res. 1997; 748: 21-29Crossref PubMed Scopus (76) Google Scholar). An immunoblot from primary hippocampal neurons indicated that PMCA2 was indeed expressed in these cells. Based on the apparent size of the PMCA2 isoform we concluded that these cells contain the "b" splice form (Fig.5 A). Additional immunoblots did not detect any PMCA2a isoform or PMCA4 (not shown); hence, any immunostaining on these cells with the anti-PMCA2 antibody will represent expression of PMCA2b. PMCA2b staining was apparent throughout the plasma membrane of the pyramidal neurons of the hippocampus (Fig.5 B). The entire boundary of the soma can be seen as well as extensive dendritic staining. Closer examination of the dendrites reveals dendritic spines that are rich in PMCA2b fluorescence (Fig.5 C). This is reminiscent of earlier results obtained with Purkinje cells of rodent cerebellum, in which PMCA2 is enriched in the dendritic spines (
Referência(s)