Inositol 1,4,5-Trisphosphate Receptor Localization and Stability in Neonatal Cardiomyocytes Requires Interaction with Ankyrin-B
2004; Elsevier BV; Volume: 279; Issue: 13 Linguagem: Inglês
10.1074/jbc.m313979200
ISSN1083-351X
AutoresPeter J. Mohler, Jonathan Q. Davis, Lydia H. Davis, Janis A. Hoffman, Peter Michaely, Vann Bennett,
Tópico(s)Cellular transport and secretion
ResumoThe molecular mechanisms required for inositol 1,4,5-trisphosphate receptor (InsP3R) targeting to specialized endoplasmic reticulum membrane domains are unknown. We report here a direct, high affinity interaction between InsP3R and ankyrin-B and demonstrate that this association is critical for InsP3R post-translational stability and localization in cultures of neonatal cardiomyocytes. Recombinant ankyrin-B membrane-binding domain directly interacts with purified cerebellar InsP3R (Kd = 2 nm). 220-kDa ankyrin-B co-immunoprecipitates with InsP3R in tissue extracts from brain, heart, and lung. Alanine-scanning mutagenesis of the ankyrin-B ANK (ankyrin repeat) repeat β-hairpin loop tips revealed that consecutive ANK repeat β-hairpin loop tips (repeats 22-24) are required for InsP3R interaction, thus providing the first detailed evidence of how ankyrin polypeptides associate with membrane proteins. Pulse-chase biosynthesis experiments demonstrate that reduction or loss of ankyrin-B in ankyrin-B (+/-) or ankyrin-B (-/-) neonatal cardiomyocytes leads to ∼3-fold reduction in half-life of newly synthesized InsP3R. Furthermore, interactions with ankyrin-B are required for InsP3R stability as abnormal InsP3R phenotypes, including mis-localization, and reduced half-life in ankyrin-B (+/-) cardiomyocytes can be rescued by green fluorescent protein (GFP)-220-kDa ankyrin-B but not by GFP-220-kDa ankyrin-B mutants, which do not associate with InsP3R. These new results provide the first physiological evidence of a molecular partner required for early post-translational stability of InsP3R. The molecular mechanisms required for inositol 1,4,5-trisphosphate receptor (InsP3R) targeting to specialized endoplasmic reticulum membrane domains are unknown. We report here a direct, high affinity interaction between InsP3R and ankyrin-B and demonstrate that this association is critical for InsP3R post-translational stability and localization in cultures of neonatal cardiomyocytes. Recombinant ankyrin-B membrane-binding domain directly interacts with purified cerebellar InsP3R (Kd = 2 nm). 220-kDa ankyrin-B co-immunoprecipitates with InsP3R in tissue extracts from brain, heart, and lung. Alanine-scanning mutagenesis of the ankyrin-B ANK (ankyrin repeat) repeat β-hairpin loop tips revealed that consecutive ANK repeat β-hairpin loop tips (repeats 22-24) are required for InsP3R interaction, thus providing the first detailed evidence of how ankyrin polypeptides associate with membrane proteins. Pulse-chase biosynthesis experiments demonstrate that reduction or loss of ankyrin-B in ankyrin-B (+/-) or ankyrin-B (-/-) neonatal cardiomyocytes leads to ∼3-fold reduction in half-life of newly synthesized InsP3R. Furthermore, interactions with ankyrin-B are required for InsP3R stability as abnormal InsP3R phenotypes, including mis-localization, and reduced half-life in ankyrin-B (+/-) cardiomyocytes can be rescued by green fluorescent protein (GFP)-220-kDa ankyrin-B but not by GFP-220-kDa ankyrin-B mutants, which do not associate with InsP3R. These new results provide the first physiological evidence of a molecular partner required for early post-translational stability of InsP3R. The inositol 1,4,5-trisphosphate receptor (InsP3R) 1The abbreviations used are: InsP3, inositol 1,4,5-trisphosphate; Ig, immunoglobulin; ER, endoplasmic reticulum; SR, sarcoplasmic reticulum; ANK, ankyrin repeat; MBD, membrane-binding domain; SBD, spectrin-binding domain; PBS, phosphate-buffered saline; WT, wild type; GFP, green fluorescent protein. 1The abbreviations used are: InsP3, inositol 1,4,5-trisphosphate; Ig, immunoglobulin; ER, endoplasmic reticulum; SR, sarcoplasmic reticulum; ANK, ankyrin repeat; MBD, membrane-binding domain; SBD, spectrin-binding domain; PBS, phosphate-buffered saline; WT, wild type; GFP, green fluorescent protein. is a large tetrameric calcium release channel localized to intracellular endoplasmic/sarcoplasmic reticulum (ER/SR) membranes. Gating of the InsP3R is controlled by local levels of InsP3 in response to receptor-mediated hydrolysis of phosphatidylinositol lipids. Additionally, InsP3R-mediated calcium release is sensitive to increases in cytosolic calcium concentration (1Berridge M.J. Lipp P. Bootman M.D. Nat. Rev. Mol. Cell. Biol. 2000; 1: 11-21Crossref PubMed Scopus (4418) Google Scholar, 2Yang J. McBride S. Mak D.O. Vardi N. Palczewski K. Haeseleer F. Foskett J.K. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 7711-7716Crossref PubMed Scopus (168) Google Scholar). Although the physiological role for InsP3R function is not clear in all tissues (i.e. cardiac muscle), calcium release from intracellular stores via the InsP3R generates spatially and temporally complex cytosolic calcium signals for a variety of calcium-dependent cellular functions, including secretion, intercellular communication, membrane transport, and synaptic plasticity (3Berridge M.J. Bootman M.D. Lipp P. Nature. 1998; 395: 645-648Crossref PubMed Scopus (1763) Google Scholar, 4Berridge M.J. Nature. 1993; 361: 315-325Crossref PubMed Scopus (6164) Google Scholar). InsP3Rs have been localized to specialized ER membrane domains (5Sharp A.H. McPherson P.S. Dawson T.M. Aoki C. Campbell K.P. Snyder S.H. J. Neurosci. 1993; 13: 3051-3063Crossref PubMed Google Scholar, 6Kijima Y. Saito A. Jetton T.L. Magnuson M.A. Fleischer S. J. Biol. Chem. 1993; 268: 3499-3506Abstract Full Text PDF PubMed Google Scholar). However, the molecular mechanisms required for targeting of InsP3Rs to these specialized ER/SR domains are unknown. Binding to the Homer family of scaffolding proteins was thought to restrict InsP3Rs at specialized membrane sites (7Tu J.C. Xiao B. Yuan J.P. Lanahan A.A. Leoffert K. Li M. Linden D.J. Worley P.F. Neuron. 1998; 21: 717-726Abstract Full Text Full Text PDF PubMed Scopus (740) Google Scholar, 8Salanova M. Priori G. Barone V. Intravaia E. Flucher B. Ciruela F. McIlhinney R.A. Parys J.B. Mikoshiba K. Sorrentino V. Cell Calcium. 2002; 32: 193-200Crossref PubMed Scopus (50) Google Scholar). However, mice with gene disruption of Homer 1, 2, and 3 show no difference in expression or cellular localization of InsP3R subtypes (9Shin D.M. Dehoff M. Luo X. Kang S.H. Tu J. Nayak S.K. Ross E.M. Worley P.F. Muallem S. J. Cell Biol. 2003; 162: 293-303Crossref PubMed Scopus (78) Google Scholar). Additionally, a number of biochemical studies have predicted roles for proteins in InsP3R localization, including protein 4.1N (10Zhang S. Mizutani A. Hisatsune C. Higo T. Bannai H. Nakayama T. Hattori M. Mikoshiba K. J. Biol. Chem. 2003; 278: 4048-4056Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar) and the Sec 6/8 complex (11Shin D.M. Zhao X.S. Zeng W. Mozhayeva M. Muallem S. J. Cell Biol. 2000; 150: 1101-1112Crossref PubMed Scopus (77) Google Scholar). The physiological relevance of these interactions is currently unknown. To date, the only functional data regarding the localization of InsP3Rs comes from mice and rats with spontaneous gene mutations for the heavy chain of myosin Va. Both rodent models display abnormal transport of InsP3R to dendritic spines of Purkinje neurons in the cerebellum, resulting in defects in long term synaptic depression (12Takagishi Y. Oda S. Hayasaka S. Dekker-Ohno K. Shikata T. Inouye M. Yamamura H. Neurosci. Lett. 1996; 215: 169-172Crossref PubMed Scopus (140) Google Scholar, 13Miyata M. Finch E.A. Khiroug L. Hashimoto K. Hayasaka S. Oda S.I. Inouye M. Takagishi Y. Augustine G.J. Kano M. Neuron. 2000; 28: 233-244Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar). There is no evidence for a direct interaction of myosin Va with the InsP3R. Ankyrins are a family of membrane-associated proteins that interact with structurally diverse membrane proteins, including the Na/K-ATPase, Na/Ca exchanger, anion exchanger, voltage-gated sodium channels, and the L1 family of cell adhesion molecules (14Bennett V. Baines A.J. Physiol. Rev. 2001; 81: 1353-1392Crossref PubMed Scopus (788) Google Scholar). Ankyrin also has been reported to bind to InsP3Rs (15Bourguignon L.Y. Jin H. Iida N. Brandt N.R. Zhang S.H. J. Biol. Chem. 1993; 268: 7290-7297Abstract Full Text PDF PubMed Google Scholar, 16Joseph S.K. Samanta S. J. Biol. Chem. 1993; 268: 6477-6486Abstract Full Text PDF PubMed Google Scholar) as well as ryanodine receptors, another ER calcium release channel (17Bourguignon L.Y. Chu A. Jin H. Brandt N.R. J. Biol. Chem. 1995; 270: 17917-17922Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar). This laboratory identified the InsP3 receptor as a potential ankyrin-binding protein (18Davis J.Q. Bennett V. J. Cell Biol. 1991; 115: 43Google Scholar). However, this interaction was not pursued due to minimal co-localization between ankyrins and InsP3R in brain. Two additional groups also independently reported ankyrin-InsP3R interactions in multiple cell types (15Bourguignon L.Y. Jin H. Iida N. Brandt N.R. Zhang S.H. J. Biol. Chem. 1993; 268: 7290-7297Abstract Full Text PDF PubMed Google Scholar, 16Joseph S.K. Samanta S. J. Biol. Chem. 1993; 268: 6477-6486Abstract Full Text PDF PubMed Google Scholar). However, one group proposed that ankyrin associated with a site on the InsP3 receptor now known to be present on the luminal side of the ER (19Bourguignon L.Y. Jin H. J. Biol. Chem. 1995; 270: 7257-7260Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar). Because this site was not physiologically relevant, and because the proteins showed minimal co-localization, the in vivo significance of the ankyrin-InsP3R interaction was not clear. More recently, evidence for a physiological role for ankyrins in localization of InsP3Rs and ryanodine receptors has emerged from studies using ankyrin-B (+/-) and ankyrin-B (-/-) mice. Neonatal cardiomyocytes derived from ankyrin-B (+/-) and ankyrin-B (-/-) mice display abnormal localization and decreased expression of InsP3R as well as abnormal localization of ryanodine receptors (20Mohler P.J. Gramolini A.O. Bennett V. J. Biol. Chem. 2002; 277: 10599-10607Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar, 21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar, 22Tuvia S. Buhusi M. Davis L. Reedy M. Bennett V. J. Cell Biol. 1999; 147: 995-1008Crossref PubMed Scopus (109) Google Scholar). Although mis-localization of InsP3R and ryanodine receptors in ankyrin-B (-/-) cardiomyocytes is restored by transfection of 220-kDa ankyrin-B cDNA into cardiomyocytes (20Mohler P.J. Gramolini A.O. Bennett V. J. Biol. Chem. 2002; 277: 10599-10607Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar, 21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar), a direct link between ankyrin-B and InsP3R to date is largely circumstantial. We recently reported that an ankyrin-B loss-of-function mutation (E1425G) is the basis of human type 4 long QT syndrome, a cardiac arrhythmia associated with sudden cardiac death (21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar). Ankyrin-B (+/-) mice show a similar cardiac phenotype, and ventricular cardiomyocytes from these animals display elevated Ca2+ transients and catecholamine-induced extrasystoles (21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar). Levels of Na/K-ATPase, Na/Ca exchanger, as well as InsP3R levels are coordinately reduced at T-tubule/SR membranes of adult ankyrin-B (+/-) cardiomyocytes, whereas ryanodine receptors were unaffected (21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar). Significantly, the ankyrin-B E1425G loss-of-function mutation causes loss-of-activity in restoration of normal levels of Na/K-ATPase, Na/Ca exchanger, and InsP3R in neonatal ankyrin-B (+/-) cardiomyocytes (21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar). Here we report the first direct evidence for a physiological interaction between ankyrin-B and InsP3R with the demonstration that interactions between ankyrin-B are required for normal InsP3R localization and stability in mouse neonatal cardiomyocytes. Moreover, we have determined for the first time how ankyrins associate with membrane-associated proteins, mapping the InsP3R binding site on ankyrin-B to the divergent β-hairpin loop tips of consecutive ANK repeats. Ankyrin-B and InsP3R Constructs—GFP-220-kDa ankyrin-B mutant constructs were made using standard molecular techniques and QuikChange mutagenesis (Stratagene). Loop mutations were designed to replace ANK repeat beta hairpin loop tip residues with two alanines. Mutants were sequenced and expressed in HEK293 cells to confirm full-length protein expression. The membrane-binding domain (plus residues of the spectrin-binding domain) of 220-kDa ankyrin-B (residues 1-958) was inserted into pGEX-6P1 for expression as a GST fusion protein. Metabolic Labeling/Pulse-chase Experiments—Neonatal cardiomyocytes were incubated in Dulbecco's modified Eagle's medium minus methionine for 2 h followed by a 1- to 3-h pulse period in Dulbecco's modified Eagle's medium containing [35S]methionine (100 μCi/ml) at 37 °C. Cells were washed and chased in normal media plus 2 mm methionine for defined times (up to 48 h). Detergent-soluble lysates were prepared and InsP3R was immunoprecipitated using affinity-purified pan-InsP3R polyclonal Ig. Equal counts were immunoprecipitated to correct for small variations in protein concentrations. InsP3R was quantitated for each time point by phosphorimaging. InsP3R/220-kDa Ankyrin-B Membrane-binding Domain Purification—InsP3R was purified essentially by the method of (23Maeda N. Niinobe M. Mikoshiba K. EMBO J. 1990; 9: 61-67Crossref PubMed Scopus (232) Google Scholar) from bovine brain cerebellum. Protein G-conjugated Dynabeads were purchased from Dynal Biotech. Ankyrin-B membrane-binding domain (MBD) with the addition of the first 80 residues of the spectrin-binding domain (SBD) containing a monoclonal antibody epitope was expressed in bacteria and purified. Affinity-purified pan-InsP3R antibody was generated in rabbits against a bacterial-expressed cleaved fusion protein representing the C-terminal cytoplasmic domain of InsP3R. In Vitro Binding Experiments—Protein G-conjugated Dynabeads were preloaded with anti-SBD monoclonal Ig followed by ankyrin-B membrane-binding domain. Ankyrin-B beads were incubated with increasing concentrations of 125I-labeled InsP3R in binding buffer (20 mm Hepes, 50 mm NaCl, 1 mm EDTA, 1 mm NaN3, 0.2% Triton X-100, 0.5 mg/ml bovine serum albumin, pH 7.2) for 2 h at 4 °C in a final volume of 60 μl. Pellets were washed and assayed for 125I in a gamma counter. Values for nonspecific binding were determined using beads lacking the ankyrin-B membrane-binding domain and were subtracted. The inset (in Fig. 2C) displays specific and control beads run on a PAGE gel and stained for protein with Coomassie Blue. Purified cerebellar InsP3R was incubated for 2 h at 4 °C with ankyrin-B MBD at a final concentration of 100 nm for each protein in a final volume of 300 μl Binding Buffer (20 mm Hepes, 50 mm NaCl, 0.2% Triton X-100, 1 mm EDTA, 1 mm NaN3, pH 7.2). 0.5 μg of either monoclonal Ig against the ankyrin-B spectrin-binding domain epitope or control whole mouse Ig was added and incubated overnight at 4 °C. Protein G Dynabeads were added, and the mixture was incubated for an additional 2 h. The Dynabeads were collected, washed, and dissolved in 5× PAGE, and polyacrylamide gel electrophoresis was performed on samples of pellets and supernatants. Gels were either stained from protein with Coomassie Blue (see Fig. 2A) or transferred to nitrocellulose membrane for Western blot analysis using an affinity-purified anti-InsP3R antibody (Fig. 2B). In Fig. 2, A and B, the lanes represent purified InsP3R (lane 1); purified ankyrin-B MBD (lane 2); specific binding (lane 3); Dynabeads with control mouse IgG (lane 4); Dynabeads lacking ankyrin-B MBD (lane 5); ankyrin-B MBD denatured by heating at 60 °C for 10 min and then placed on ice (lane 6). Neonatal Cardiomyocytes—Hearts from P1 or P2 mice were isolated, and cardiomyocytes from wild-type, ankyrin-B (+/-), or ankyrin-B (-/-) mice were cultured and transfected as described previously (20Mohler P.J. Gramolini A.O. Bennett V. J. Biol. Chem. 2002; 277: 10599-10607Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar). Immunofluorescence—Neonatal cardiomyocytes were washed with phosphate-buffered saline (PBS, pH 7.4) and fixed in warm 4% paraformaldehyde (37 °C). Cells were blocked/permeabilized in PBS containing 0.075% Triton X-100 and 3% fish oil gelatin (Sigma) and incubated in primary antisera overnight at 4 °C. Following washes (PBS plus 0.1% Triton X-100), cells were incubated in secondary antisera (Alexa 488, 568; Molecular Probes) for 8 h at 4 °C and mounted using Vectashield (Vector) and #1 coverslips. Images were collected by using a Zeiss 510 Meta confocal microscope (100 power oil 1.45 numerical aperture (Zeiss), pinhole equals 1.0 Airy Disc) using Carl Zeiss Imaging software. Both channels were collected on PMT3. Images were imported into Adobe Photoshop for cropping and contrast adjustment. Immunoprecipitations—Rat heart, brain, and lung lysates were prepared as described (21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar). Only ∼50-60% of 220-kDa ankyrin-B and ∼40% of InsP3R was solubilized using this protocol. Detergent-soluble lysate fractions were bound to affinity-purified Ig bound to Protein A/G-Sepharose. Beads were washed to prevent nonspecific associations, and bound protein was eluted and analyzed by quantitative Western blot analysis using 125I Protein A and phosphorimaging. Statistics—When appropriate, data were analyzed using two-tailed Student's t test, and values less than p < 0.05 were considered significant. Values are expressed as the mean ± S.D. Ankyrin-B Co-immunoprecipitates with Multiple InsP3R Subtypes—We performed co-immunoprecipitations from adult rat tissues to determine whether potential ankyrin-B interactions are restricted to specific InsP3R subtypes. Affinity-purified polyclonal Ig against ankyrin-B co-immunoprecipitated a significant fraction (10-20%) of total soluble InsP3R from detergent-soluble lysates of adult rat brain, heart, and lung (Fig. 1A). Conversely, pan-InsP3R affinity-purified Ig co-immunoprecipitated a small but significant fraction (∼5%) of total soluble 220-kDa ankyrin-B from the same detergent-soluble tissue lysates (Fig. 1B). The predominant isoforms of InsP3R expressed in brain, heart, and lung are type 1, type 2, and type 3, respectively (24Inoue T. Kato K. Kohda K. Mikoshiba K. J. Neurosci. 1998; 18: 5366-5373Crossref PubMed Google Scholar, 25Bezprozvanny I. Watras J. Ehrlich B.E. Nature. 1991; 351: 751-754Crossref PubMed Scopus (1431) Google Scholar, 26Furuichi T. Simon-Chazottes D. Fujino I. Yamada N. Hasegawa M. Miyawaki A. Yoshikawa S. Guenet J.L. Mikoshiba K. Receptors Channels. 1993; 1: 11-24PubMed Google Scholar, 27Miyata M. Miyata H. Mikoshiba K. Ohama E. Acta Neuropathol. (Berl.). 1999; 98: 226-232Crossref PubMed Scopus (27) Google Scholar, 28Lipp P. Laine M. Tovey S.C. Burrell K.M. Berridge M.J. Li W. Bootman M.D. Curr. Biol. 2000; 10: 939-942Abstract Full Text Full Text PDF PubMed Scopus (200) Google Scholar, 29Perez P.J. Ramos-Franco J. Fill M. Mignery G.A. J. Biol. Chem. 1997; 272: 23961-23969Abstract Full Text Full Text PDF PubMed Scopus (139) Google Scholar, 30Sienaert I. Huyghe S. Parys J.B. Malfait M. Kunzelmann K. De Smedt H. Verleden G.M. Missiaen L. Pflugers Arch. 1998; 436: 40-48Crossref PubMed Scopus (29) Google Scholar, 31Missiaen L. Parys J.B. Sienaert I. Maes K. Kunzelmann K. Takahashi M. Tanzawa K. De Smedt H. J. Biol. Chem. 1998; 273: 8983-8986Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar). Therefore, ankyrin-B associates with multiple InsP3R isoforms in brain, lung, and heart, likely due to a binding site conserved between InsP3R subtypes. One caveat to these results is that we detected InsP3R using pan-InsP3R antibody. Therefore, it is a formal possibility that even though a single InsP3R subtype is predominant in a tissue, we have detected a large fraction of a minor InsP3R isoform. Recombinant Ankyrin-B Membrane-binding Domain Binds to Purified Cerebellar InsP3R—We next characterized the interaction between the ankyrin-B and InsP3R using purified proteins. Recombinant ankyrin-B membrane-binding domain (residues 1-958) isolated from bacteria as a glutathione S-transferase fusion protein was immobilized on magnetic Dynabeads using an affinity-purified monoclonal Ig directed against the C terminus of the expressed protein (residues 922-938). InsP3R was purified from cerebellum (type 1) and 125I-labeled for binding (see “Materials and Methods”). Immobilized ankyrin-B membrane-binding domain (MBD) associated with 125I-labeled InsP3R purified from cerebellum with an affinity of ∼2 nm (Fig. 2, C and D). Minimal 125I-labeled InsP3R binding was detected with control Ig alone (Fig. 2, A and B). Moreover, InsP3R did not associate with the Dynabeads alone (Fig. 2, A and B). Additionally, no interaction between ankyrin-B and InsP3R was observed when the ankyrin-B membrane-binding domain was heat-denatured (heated to 65 °C; Fig. 2, A and B). The low binding stoichiometry observed in this assay (maximum capacity of 0.1 mol of InsP3R per mole of ankyrin-B) may be due to steric hindrance due to the large size of InsP3R tetramers, or simultaneous binding of InsP3R to more than one immobilized ankyrin-B membrane-binding domain. Alternatively, a population of the ankyrin-B membrane-binding domain may have been inactive. However, from these data we can conclude that association of InsP3R and ankyrin-B in rat tissues (Fig. 1) is mediated by a high-affinity direct interaction between the InsP3R and the ankyrin-B membrane-binding domain (Fig. 2). InsP3R-Ankyrin-B Interaction Requires Tips of β-Hairpin Loops of ANK Repeats 22-24—We used an alanine-scanning mutagenesis approach to determine the site(s) on ankyrin-B membrane-binding domain responsible for interaction with InsP3R. We constructed full-length GFP-220-kDa ankyrin-B mutants based on the recent crystal structure of ankyrin-R membrane-binding domain as well as the known binding characteristics of other ANK repeat-containing proteins (32Michaely P. Tomchick D.R. Machius M. Anderson R.G. EMBO J. 2002; 21: 6387-6396Crossref PubMed Scopus (170) Google Scholar, 33Sedgwick S.G. Smerdon S.J. Trends Biochem. Sci. 1999; 24: 311-316Abstract Full Text Full Text PDF PubMed Scopus (661) Google Scholar). ANK repeats of the membrane-binding domain fold as two antiparallel pairs of α-helices connected by a β-hairpin loop motif (33Sedgwick S.G. Smerdon S.J. Trends Biochem. Sci. 1999; 24: 311-316Abstract Full Text Full Text PDF PubMed Scopus (661) Google Scholar). The α-helical pairs stack to form a form a superhelix (∼360° turn for 24 repeats of full-length ankyrins) around a large central cavity (see Fig. 5 (32Michaely P. Tomchick D.R. Machius M. Anderson R.G. EMBO J. 2002; 21: 6387-6396Crossref PubMed Scopus (170) Google Scholar)). We initially constructed GFP-220-kDa ankyrin-B mutants in which each of 24 ANK repeat β-hairpin loop tip residues were substituted with two alanines (Fig. 3A). The rationale for this selection was that loop tips of the ANK repeat β-hairpins are variable between repeats, face the solvent, should not be required for folding, and have been implicated in protein interactions in some ANK repeat-containing proteins (33Sedgwick S.G. Smerdon S.J. Trends Biochem. Sci. 1999; 24: 311-316Abstract Full Text Full Text PDF PubMed Scopus (661) Google Scholar). GFP-220-kDa ankyrin-B was expressed in HEK293 cells, immobilized using affinity-purified polyclonal GFP Ig, and incubated with purified cerebellar InsP3R (see “Materials and Methods”). Nonspecific binding was reduced by washes, and bound InsP3R was eluted. Bound InsP3R was analyzed by SDS-PAGE and quantitative Western blot analysis using an affinity-purified polyclonal InsP3R Ig. A second gel was used to quantify GFP-220-kDa ankyrin-B or mutant GFP-220-kDa ankyrin-B expression. InsP3R bound to mutated GFP-220-kDa ankyrin-B was corrected for relative GFP expression and compared with binding to wild-type GFP-220-kDa ankyrin-B. We observed no difference in binding of InsP3R to any of the 220-kDa ankyrin-B mutants of individual β-hairpin loops of ANK repeats 1-24 compared with wild-type GFP-220-kDa ankyrin-B (Fig. 3, B and C). We next created GFP-220-kDa ankyrin-B mutants with consecutive β-hairpin loop tip mutations (i.e. repeats 5-6, 4-6, 3-6, 2-6, and 1-6 in four sets of six ANK repeats 1-6, 7-12, 13-18, and 19-24; Fig. 4A). Mutation of six consecutive β-hairpin loops of repeats 1-6, 7-12, and 13-18 also had no effect on binding to the InsP3R (Fig. 4C). Mutation to consecutive ANK repeat β-hairpin loop tips in repeats 19-24 abolished interactions with purified cerebellar InsP3R (Fig. 4, B and C). In fact, mutation of ANK repeat 22-24 β-hairpin loop tips was sufficient to nearly eliminate InsP3R interactions (Fig. 4, B and C). These results suggest that the tips of β-hairpin loops 22-24 of ankyrin-B provide an interaction surface for direct contact with InsP3R (Fig. 5). Although our previous results (Fig. 2) would argue otherwise, a caveat to these experiments is that since GFP-ankyrin-B was immunoisolated from cell lysates, we cannot formally rule out the possibility that an intermediate protein was involved in the GFP-ankyrin-B/InsP3R interaction and therefore the interaction may not be direct. Nonetheless, these experiments provide the first detailed insight into sites on ankyrin involved in binding to a membrane protein. Loss of Ankyrin-B-InsP3R Interaction Leads to Mis-localization of InsP3R in Cardiomyocytes—Decreased expression and mis-localization of InsP3R in ankyrin-B (+/-) and ankyrin-B (-/-) neonatal cardiomyocytes can be rescued by transfection with GFP-220-kDa ankyrin-B cDNA (20Mohler P.J. Gramolini A.O. Bennett V. J. Biol. Chem. 2002; 277: 10599-10607Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar). We used the loss-of-binding mutants of 220-kDa ankyrin-B to determine whether direct interactions with ankyrin-B are required for normal localization of InsP3R. Compared with wild-type cardiomyocytes (Fig. 6A), ankyrin-B (-/-) cardiomyocytes display reduced spontaneous contraction rates, abnormal Ca2+ dynamics, and mis-sorting of InsP3R (Fig. 6B) (20Mohler P.J. Gramolini A.O. Bennett V. J. Biol. Chem. 2002; 277: 10599-10607Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar, 21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar). These abnormal phenotypes can be rescued by transfection of GFP-220-kDa ankyrin-B (Fig. 6C (20Mohler P.J. Gramolini A.O. Bennett V. J. Biol. Chem. 2002; 277: 10599-10607Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar, 21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar)). We transfected GFP-220-kDa ankyrin-B mutants, which had multiple alanine substitutions at β-hairpin loop tips (Fig. 4A) into ankyrin-B (-/-) cardiomyocytes and analyzed the expression and localization of GFP-220-kDa ankyrin-B mutants and InsP3R. All GFP-220-kDa ankyrin-B mutants were localized the same as endogenous ankyrin-B and wild-type GFP-220-kDa ankyrin-B (Fig. 6, D-F). Thirteen GFP-220-kDa ankyrin-B mutants rescued InsP3R expression and localization, but five non-binding mutants (GFP-220-kDa ankyrin-B R23-24, R22-24, R21-24, R20-24, and R19-24), representing mutations to repeats 19-24 (Fig. 4C), were unable to rescue abnormal InsP3R expression and localization (Fig. 6, D-F). Therefore, ankyrin-B association is required for normal InsP3R expression and localization in neonatal mouse cardiomyocytes. Because these experiments were performed in vivo, we cannot rule out the possibility that an intermediate protein was involved in the ankyrin-B β-hairpin loop tip interaction with InsP3R. Interaction with Ankyrin-B Increases InsP3R Stability in Neonatal Cardiomyocytes—Decreased InsP3R expression in ankyrin-B (+/-) and ankyrin-B (-/-) tissues (21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar, 22Tuvia S. Buhusi M. Davis L. Reedy M. Bennett V. J. Cell Biol. 1999; 147: 995-1008Crossref PubMed Scopus (109) Google Scholar) is the result of a post-translational event, because Northern blot analyses show no difference in InsP3R mRNA levels between wild-type and ankyrin-B (+/-) heart (21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar). We therefore measured the half-life of InsP3R in neonatal cardiomyocytes derived from wild-type, ankyrin-B (+/-), and ankyrin-B (-/-) mice to determine whether ankyrin-B is required for InsP3R stability. Primary cardiomyocytes were labeled with [35S]methionine and chased for defined time intervals (see “Materials and Methods”). Following each chase period, cells were lysed, protein was solubilized, and InsP3R was immunoprecipitated using pan-InsP3R affinity-purified Ig and analyzed by SDS-PAGE and phosphorimaging. The half-life of InsP3R is ∼11.7 h in wild-type neonatal mouse cardiomyocytes (Fig. 7, triangles). Reduction of ankyrin-B in ankyrin-B (+/-) or ankyrin-B (-/-) primary neonatal cardiomyocytes leads to a significant reduction in InsP3R half-life. Significantly, InsP3R half-life was reduced to 4.6 h in ankyrin-B (+/-) cells (Fig. 7; open squares, p < 0.05). InsP3R half-life in ankyrin-B (-/-) cells was 3.7 h (Fig. 7; closed circles, p < 0.05), similar to the half-life observed in ankyrin-B (+/-) cardiomyocytes. Ankyrin-B levels therefore regulate post-translational stability of the InsP3R in neonatal cardiomyocytes. Additionally, reduction of only ∼50% of 220-kDa ankyrin-B produces nearly the same effect on InsP3R stability as complete loss of 220-kDa ankyrin-B. We next determined whether interactions with ankyrin-B are required for InsP3R stability. Ankyrin-B (+/-) neonatal cardiomyocytes transfected with GFP-220-kDa ankyrin-B or mutated GFP-220-kDa ankyrin-B (loop tips 19-24 to alanine, does not associate with purified InsP3R; Fig. 4, B and C) were metabolically labeled and chased, and the InsP3R half-life was determined. Wild-type GFP-220-kDa ankyrin-B rescued InsP3R stability to near normal levels in ankyrin-B (+/-) cardiomyocytes (11.7 versus 11.0 h; Fig. 8, n = 3; NS). However, transfection of ankyrin-B (+/-) cells with mutant GFP-220-kDa ankyrin-B R19-24 (Fig. 4, C and D) was unable to restore InsP3R half-life to normal levels (11.7 versus 4.4 h; Fig. 8, n = 3; p < 0.05), even though the mutant GFP-220-kDa ankyrin-B was expressed at normal levels and localized as GFP-220-kDa ankyrin-B. Because these experiments were performed in vivo, we cannot rule out the possibility that an intermediate protein was involved in the ankyrin-B β-hairpin loop tip interaction with InsP3R. However, these results clearly demonstrate that ankyrin-B contributes to post-translational stability of the InsP3R through association with InsP3R. We report a direct, high affinity interaction between ankyrin-B and InsP3 receptor isoforms and demonstrate that ankyrin-B is critical for InsP3R post-translational stability and localization in primary cardiomyocytes. These new results provide the first physiological evidence of a molecular partner required for localization of InsP3R to an ER membrane domain. Furthermore, using an alanine-scanning mutagenesis strategy we determined that multiple consecutive beta-hairpin loop tips are required for InsP3R interaction, thus providing the first detailed evidence of how ankyrin polypeptides associate with membrane proteins. A puzzle remains that the majority of ankyrin-B and InsP3R do not co-localize in neonatal cardiomyocytes. The major population of ankyrin-B is over the M-line in cardiomyocytes, whereas InsP3R localize at the Z-line (Fig. 6A). One possible explanation for this discrepancy is that ankyrin-B is in large excess over the InsP3R, and a subpopulation of ankyrin-B actually is co-localized with the InsP3R. Given that InsP3R are assembled into tetramers, only one ankyrin-B per four InsP3R subunits would be sufficient for a complex, but might be difficult to detect by immunofluorescence. In support of an ankyrin-B-InsP3R complex is that ankyrin-B and InsP3R do co-immunoprecipitate from tissue extracts of heart (Fig. 1). Moreover, a subpopulation of ankyrin-B is co-localized with InsP3R at T-tubule sites in adult cardiomyocytes (21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar). The functional importance of this subpopulation of ankyrin-B is underscored by the finding that loss of the T-tubule-associated ankyrin-B in ankyrin-B (+/-) cardiomyocytes is accompanied by substantial loss of InsP3R (21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar). The binding site on the InsP3R that interacts with ankyrin-B remains to be identified. A previous study using InsP3R purified from T-lymphoma cells proposed that ankyrin associates with InsP3R at a site selected based on limited sequence similarity to an ankyrin-binding region in CD44 (19Bourguignon L.Y. Jin H. J. Biol. Chem. 1995; 270: 7257-7260Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar). However, it now is clear that ankyrin-binding proteins do not have conserved sequence motifs (14Bennett V. Baines A.J. Physiol. Rev. 2001; 81: 1353-1392Crossref PubMed Scopus (788) Google Scholar). Moreover, further characterization of InsP3R membrane topology revealed that this site is located to the lumen of the ER and therefore could not associate with the cytosolic ankyrin in vivo. The ankyrin-B-binding site on the InsP3R is likely a conserved motif on all three receptor subtypes, because all three major isoforms of InsP3R appear to associate with ankyrin-B and all are reduced in tissues of ankyrin-B (+/-) mice (data not shown). Although the functional role of InsP3R calcium-induced calcium release in excitation-contraction coupling is unclear, the interaction of ankyrin-B with InsP3R provides a potential mechanism to couple InsP3R-enriched domains of the ER to membrane proteins in the plasma membrane. Studies with anion exchanger and neurofascin have demonstrated that ankyrins are capable of binding multiple proteins at once (34Michaely P. Bennett V. J. Biol. Chem. 1995; 270: 31298-31302Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar, 35Michaely P. Bennett V. J. Biol. Chem. 1995; 270: 22050-22057Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar). The ankyrin-B-InsP3R interaction requires a relatively small region (repeats 21-24) on the ankyrin-B membrane-binding domain (Fig. 5). Therefore, a majority of the ankyrin-B membrane-binding domain is accessible to accommodate potentially multiple additional protein partners. Plasma membrane proteins that associate with ankyrin-B and are candidates to be coupled to InsP3R through ankyrin-B include the Na/K-ATPase and Na/Ca exchanger (21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar). It is of interest in this regard that ankyrin-B (+/-) cardiomyocytes exhibit coordinate loss of Na/K-ATPase, Na/Ca exchanger of T-tubules (specialized extensions of the plasma membrane) accompanied by loss of InsP3R from the sarcoplasmic reticulum (a domain of the endoplasmic reticulum (21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar)). Sarcoplasmic reticulum-derived Ca2+ transients are elevated in ankyrin-B (+/-) cardiomyocytes (21Mohler P.J. Schott J.J. Gramolini A.O. Dilly K.W. Guatimosim S. DuBell W.H. Song L.S. Haurogne K. Kyndt F. Ali M.E. Rogers T.B. Lederer W.J. Escande D. Marec H.L. Bennett V. Nature. 2003; 421: 634-639Crossref PubMed Scopus (831) Google Scholar), suggesting that Ca2+ levels in the sarcoplasmic reticulum also are elevated in the absence of ankyrin-B-coupled proteins. These data are consistent with a role of ankyrin-B-coupled InsP3R in export of Ca2+ from the SR through ankyrin-B-coupled Na/Ca exchanger and Na/K-ATPase located in the plasma membrane. Cardiomyocytes are excitable cells and experience repetitive events of extracellular Ca2+ entry through voltage-gated Ca2+ channels. In contrast, excitation-contraction coupling in both skeletal and smooth muscle does not require extracellular Ca2+ and is regulated by Ca2+ release from intracellular stores. Interestingly, smooth muscle and skeletal muscle also do not require ankyrin-B-dependent pathways for localization of InsP3R. 2P. J. Mohler, J. Q. Davis, L. H. Davis, J. A. Hoffman, P. Michaely, and V. Bennett, unpublished observations. In fact, ankyrin-B is not expressed in smooth muscle, 2P. J. Mohler, J. Q. Davis, L. H. Davis, J. A. Hoffman, P. Michaely, and V. Bennett, unpublished observations. and ankyrin-B is not localized with InsP3R at intracellular T-tubule/SR junctions in skeletal muscle. 2P. J. Mohler, J. Q. Davis, L. H. Davis, J. A. Hoffman, P. Michaely, and V. Bennett, unpublished observations. These considerations suggest ankyrin-B-coupled InsP3R is an adaptation of excitable cardiomyocytes with repetitive events of extracellular entry and export of calcium. Sarah Jones is gratefully acknowledged for assistance in maintaining the ankyrin-B mutant mouse colony.
Referência(s)