A Cypher/ZASP Mutation Associated with Dilated Cardiomyopathy Alters the Binding Affinity to Protein Kinase C
2004; Elsevier BV; Volume: 279; Issue: 8 Linguagem: Inglês
10.1074/jbc.m311849200
ISSN1083-351X
AutoresTakuro Arimura, Takeharu Hayashi, Hajime Terada, Su-Yeoun Lee, Qiang Zhou, Megumi Takahashi, Kazuo Ueda, Tatsuhito Nouchi, Shigeru Hohda, Makoto Shibutani, Masao Hirose, Ju Chen, Jeong-Euy Park, Michio Yasunami, Hideharu Hayashi, Akinori Kimura,
Tópico(s)Viral Infections and Immunology Research
ResumoDilated cardiomyopathy is characterized by ventricular dilation with systolic dysfunction of cardiac muscle. Recent genetic studies have revealed that mutations in genes for cytoskeleton proteins distributed in the Z-disc and/or intercalated discs of the cardiac muscle are major predictors of cardiomyopathy. However, as mutations in these genes can account for only a part of the patient population, there should be another disease-causing gene(s) for cardiomyopathy. Cypher/ZASP appears to be an ideal candidate for the cardiomyopathy causative gene, because Cypher/ZASP encodes a Z-disc associated protein, and recent studies have demonstrated that Cypher/ZASP knock-out mice develop cardiomyopathy. In this study, we searched for sequence variations in Cypher/ZASP in 96 unrelated Japanese patients with dilated cardiomyopathy. A D626N mutation located within the third LIM domain was identified in a familial case but not found in the unrelated controls. A family study of the patient showed that all affected siblings tested had the same mutation. Clinical information of the affected family members suggested that the mutation was associated with late onset cardiomyopathy. To reveal the biochemical changes due to the mutation, we performed a yeast two-hybrid assay and a pull-down assay. It was demonstrated by both assays that the D626N mutation of Cypher/ZASP increased the affinity of the LIM domain for protein kinase C, suggesting a novel biochemical mechanism of the pathogenesis of dilated cardiomyopathy. Dilated cardiomyopathy is characterized by ventricular dilation with systolic dysfunction of cardiac muscle. Recent genetic studies have revealed that mutations in genes for cytoskeleton proteins distributed in the Z-disc and/or intercalated discs of the cardiac muscle are major predictors of cardiomyopathy. However, as mutations in these genes can account for only a part of the patient population, there should be another disease-causing gene(s) for cardiomyopathy. Cypher/ZASP appears to be an ideal candidate for the cardiomyopathy causative gene, because Cypher/ZASP encodes a Z-disc associated protein, and recent studies have demonstrated that Cypher/ZASP knock-out mice develop cardiomyopathy. In this study, we searched for sequence variations in Cypher/ZASP in 96 unrelated Japanese patients with dilated cardiomyopathy. A D626N mutation located within the third LIM domain was identified in a familial case but not found in the unrelated controls. A family study of the patient showed that all affected siblings tested had the same mutation. Clinical information of the affected family members suggested that the mutation was associated with late onset cardiomyopathy. To reveal the biochemical changes due to the mutation, we performed a yeast two-hybrid assay and a pull-down assay. It was demonstrated by both assays that the D626N mutation of Cypher/ZASP increased the affinity of the LIM domain for protein kinase C, suggesting a novel biochemical mechanism of the pathogenesis of dilated cardiomyopathy. Dilated cardiomyopathy (DCM) 1The abbreviations used are: DCM, dilated cardiomyopathy; MLP, muscle LIM protein; PKC, protein kinase C; RT, reverse transcription; SSCP, single-stranded conformation polymorphism; Y2H, yeast two-hybrid; β-gal, β-galactosidase; ENH, Enigma homologue protein; CA, cerebellar ataxia; HA, hemagglutinin A; RACK, receptor for activated C kinase; RICK, receptor for inactivated C-kinase. is characterized by ventricular dilation accompanied by systolic dysfunction in the absence of known causes that affect the cardiac function. Family studies of DCM patients have demonstrated that 20-35% of DCM is caused by inherited gene mutations (familial DCM) (1Grunig E. Tasman J.A. Kucherer H. Franz W. Kubler W. Katus H.A. J. Am. Coll. Cardiol. 1998; 31: 186-194Crossref PubMed Scopus (358) Google Scholar). Although familial DCM can be transmitted as autosomal recessive, X-linked, or through mitochondrial traits, it is evident that autosomal dominant inheritance occurs the most (2Seidman J.G. Seidman C. Cell. 2001; 104: 557-567Abstract Full Text Full Text PDF PubMed Scopus (899) Google Scholar, 3Schonberger J. Seidman C.E. Am. J. Hum. Genet. 2001; 69: 249-260Abstract Full Text Full Text PDF PubMed Scopus (118) Google Scholar). Missense mutations in genes for cardiac α-actin, desmin, lamin A/C, δ-sarcoglycan, β-cardiac myosin heavy chain, cardiac troponin T, α-tropomyosin, metavinculin, titin, muscle LIM protein (MLP), Tcap/telethonin, and phospholamban are all associated with autosomal dominant DCM (4Olson T.M. Michels V.V. Thibodeau S.N. Tai Y.S. Keating M.T. Science. 1998; 280: 750-752Crossref PubMed Scopus (599) Google Scholar, 5Li D. Tapscoft T. Gonzalez O. Burch P.E. Quinones M.A. Zoghbi W.A. Hill R. Bachinski L.L. Mann D.L. Roberts R. Circulation. 1999; 100: 461-464Crossref PubMed Scopus (388) Google Scholar, 6Fatkin D. MacRae C. Sasaki T. Wolff M.R. Porcu M. Frenneaux M. Atherton J. Vidaillet Jr., H.J. Spudich S. De Girolami U. Seidman J.G. Seidman C. Muntoni F. Muehle G. Johnson W. McDonough B. N. Engl. J. Med. 1999; 341: 1715-1724Crossref PubMed Scopus (1088) Google Scholar, 7Tsubata S. Bowles K.R. Vatta M. Zintz C. Titus J. Muhonen L. Bowles N.E. Towbin J.A. J. Clin. Investig. 2000; 106: 655-662Crossref PubMed Scopus (311) Google Scholar, 8Kamisago M. Sharma S.D. DePalma S.R. Solomon S. Sharma P. McDonough B. Smoot L. Mullen M.P. Woolf P.K. Wigle E.D. Seidman J.G. Seidman C.E. N. Engl. J. Med. 2000; 343: 1688-1696Crossref PubMed Scopus (589) Google Scholar, 9Olson T.M. Kishimoto N.Y. Whitby F.G. Michels V.V. J. Mol. Cell. Cardiol. 2001; 33: 723-732Abstract Full Text PDF PubMed Scopus (229) Google Scholar, 10Olson T.M. Illenberger S. Kishimoto N.Y. Huttelmaier S. Keating M.T. Jockusch B.M. Circulation. 2002; 105: 431-437Crossref PubMed Scopus (225) Google Scholar, 11Gerull B. Gramlich M. Atherton J. McNabb M. Trombitas K. Klaassen S.S. Seidman J.G. Seidman C. Granzier H. Labeit S. Frenneaux M. Thierfelder L. Nat. Genet. 2002; 30: 201-204Crossref PubMed Scopus (460) Google Scholar, 12Itoh-Satoh M. Hayashi T. Nishi H. Koga Y. Arimura T. Koyanagi T. Takahashi M. Hohda S. Ueda K. Nouchi T. Hiroe M. Marumo F. Imaizumi T. Yasunami M. Kimura A. Biochem. Biophys. Res. Commun. 2002; 291: 385-393Crossref PubMed Scopus (212) Google Scholar, 13Knoll R. Hoshijima M. Hoffman H.M. Person V. Lorenzen-Schmidt I. Bang M.L. Hayashi T. Shiga N. Yasukawa H. Schaper W. McKenna W. Yokoyama M. Schork N.J. Omens J.H. McCulloch A.D. Kimura A. Gregorio C.C. Poller W. Schaper J. Schultheiss H.P. Chien K.R. Cell. 2002; 111: 943-955Abstract Full Text Full Text PDF PubMed Scopus (636) Google Scholar, 14Schmitt J.P. Kamisago M. Asahi M. Li G.H. Ahmad F. Mende U. Kranias E.G. MacLennan D.H. Seidman J.G. Seidman C.E. Science. 2003; 299: 1410-1413Crossref PubMed Scopus (501) Google Scholar, 15Haghighi K. Kolokathis F. Pater L. Lynch R.A. Asahi M. Gramolini A.O. Fan G.C. Tsiapras D. Hahn H.S. Adamopoulos S. Liggett S.B. Dorn III, G.W. MacLennan D.H. Kremastinos D.T. Kranias E.G. J. Clin. Investig. 2003; 111: 869-876Crossref PubMed Scopus (370) Google Scholar). However, mutations in the known causative genes can be found only in a part of the patient population. In addition, linkage studies in multiplex families have suggested that there are several other disease loci of which the gene responsible for causing DCM remains to be identified, including one mapped on 10q21-q23 (16Bowles K.R. Gajarski R. Porter P. Goytia V. Bachinski L. Roberts R. Pignatelli R. Towbin J.A. J. Clin. Investig. 1996; 98: 1355-1360Crossref PubMed Scopus (132) Google Scholar). Abnormalities in Z-disc-related cytoskeletal proteins are associated with DCM also in animal models. For example, mice lacking MLP display many characteristic features of human DCM (17Arber S. Hunter J.J. Ross Jr., J. Hongo M. Sansig G. Borg J. Perriard J.C. Chien K.R. Caroni P. Cell. 1997; 88: 393-403Abstract Full Text Full Text PDF PubMed Scopus (708) Google Scholar). Similarly, loss of δ-sarcoglycan results in DCM in a hamster model (18Nigro V. Okazaki Y. Belsito A. Piluso G. Matsuda Y. Politano L. Nigro G. Ventura C. Abbondanza C. Molinari A.M. Acampora D. Nishimura M. Hayashizaki Y. Puca G.A. Hum. Mol. Genet. 1997; 6: 601-607Crossref PubMed Scopus (249) Google Scholar), and a mouse knock-out for the actinin-associated LIM protein develops DCM (19Pashmforoush M. Pomies P. Peterson K.L. Kubalak S. Ross Jr., J. Hefti A. Aebi U. Beckerle M.C. Chien K.R. Nat. Med. 2001; 7: 591-597Crossref PubMed Scopus (164) Google Scholar). These observations suggest that abnormalities in cytoskeletal proteins expressed in cardiac muscle cause DCM (20Towbin J.A. Bowles N.E. Curr. Cardiol. Rep. 2000; 2: 475-480Crossref PubMed Scopus (36) Google Scholar). Myocardial function is associated with the regulation or activation of cell signal kinases such as protein kinase C (PKC). For example, modification of myocardial proteins by PKC plays a key role in the regulation of contractility and the growth of cardiomyocytes (21Steinberg S.F. Goldberg M. Rybin V.O. J. Mol. Cell. Cardiol. 1995; 27: 141-153Abstract Full Text PDF PubMed Scopus (144) Google Scholar), whereas alterations in the expression, activity, or localization of PKC are associated with cardiac hypertrophy and failure (22Bowling N. Walsh R.A. Song G. Estridge T. Sandusky G.E. Fouts R.L. Mintze K. Pickard T. Roden R. Bristow M.R. Sabbah H.N. Mizrahi J.L. Gromo G. King G.L. Vlahos C.J. Circulation. 1999; 99: 384-391Crossref PubMed Scopus (388) Google Scholar, 23Jalili T. Takeishi Y. Song G. Ball N.A. Howles G. Walsh R.A. Am. J. Physiol. 1999; 277: H2298-H2304PubMed Google Scholar). Recently, a gene for the PDZ and LIM domain-containing cytoskeletal protein, Cypher/ZASP, was identified in mouse (Cypher) (24Zhou Q. Ruiz-Lozano P. Martone M.E. Chen J. J. Biol. Chem. 1999; 274: 19807-19813Abstract Full Text Full Text PDF PubMed Scopus (205) Google Scholar) and human (ZASP) (25Faulkner G. Pallavicini A. Formentin E. Comelli A. Ievolella C. Trevisan S. Bortoletto G. Scannapieco P. Salamon M. Mouly V. Valle G. Lanfranchi G. J. Cell Biol. 1999; 146: 465-475Crossref PubMed Scopus (188) Google Scholar). Cypher is also called Oracle (26Passier R. Richardson J.A. Olson E.N. Mech. Dev. 2000; 92: 277-284Crossref PubMed Scopus (65) Google Scholar). Cypher/ZASP is expressed in the striated muscles and encodes several isoforms (24Zhou Q. Ruiz-Lozano P. Martone M.E. Chen J. J. Biol. Chem. 1999; 274: 19807-19813Abstract Full Text Full Text PDF PubMed Scopus (205) Google Scholar, 25Faulkner G. Pallavicini A. Formentin E. Comelli A. Ievolella C. Trevisan S. Bortoletto G. Scannapieco P. Salamon M. Mouly V. Valle G. Lanfranchi G. J. Cell Biol. 1999; 146: 465-475Crossref PubMed Scopus (188) Google Scholar, 26Passier R. Richardson J.A. Olson E.N. Mech. Dev. 2000; 92: 277-284Crossref PubMed Scopus (65) Google Scholar, 27Huang C. Zhou Q. Liang P. Hollander M.S. Sheikh F. Li X. Greaser M. Shelton G.D. Evans S. Chen J. J. Biol. Chem. 2003; 278: 7360-7365Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar). The PDZ domain at the N terminus is required for binding to α-actinin in the Z-disc (25Faulkner G. Pallavicini A. Formentin E. Comelli A. Ievolella C. Trevisan S. Bortoletto G. Scannapieco P. Salamon M. Mouly V. Valle G. Lanfranchi G. J. Cell Biol. 1999; 146: 465-475Crossref PubMed Scopus (188) Google Scholar), and the LIM domains at the C terminus are involved in the binding to PKCs (24Zhou Q. Ruiz-Lozano P. Martone M.E. Chen J. J. Biol. Chem. 1999; 274: 19807-19813Abstract Full Text Full Text PDF PubMed Scopus (205) Google Scholar). In addition, Cypher/ZASP knock-out mice display disorganized and fragmented Z-discs in both skeletal and cardiac muscles, leading to a severe form of congenital myopathy and cardiomyopathy (27Huang C. Zhou Q. Liang P. Hollander M.S. Sheikh F. Li X. Greaser M. Shelton G.D. Evans S. Chen J. J. Biol. Chem. 2003; 278: 7360-7365Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar, 28Zhou Q. Chu P.H. Huang C. Cheng C.F. Martone M.E. Knoll G. Shelton G.D. Evans S. Chen J. J. Cell Biol. 2001; 155: 605-612Crossref PubMed Scopus (238) Google Scholar). Moreover, human Cypher/ZASP is mapped on chromosome 10q22.3-q23.2 (University of California at Santa Cruz Human Genome Browser, July 2003; chromosome 10, 88093025-88159329) (25Faulkner G. Pallavicini A. Formentin E. Comelli A. Ievolella C. Trevisan S. Bortoletto G. Scannapieco P. Salamon M. Mouly V. Valle G. Lanfranchi G. J. Cell Biol. 1999; 146: 465-475Crossref PubMed Scopus (188) Google Scholar), overlapping with a DCM locus (16Bowles K.R. Gajarski R. Porter P. Goytia V. Bachinski L. Roberts R. Pignatelli R. Towbin J.A. J. Clin. Investig. 1996; 98: 1355-1360Crossref PubMed Scopus (132) Google Scholar). We report here a Cypher/ZASP gene mutation, associated with late-onset familial DCM, that increases the LIM-PKC interaction. This is the first report suggesting that an abnormality in the anchoring-protein of PKC may play an important role in the pathogenesis of DCM. Reverse Transcription (RT)-PCR Analysis of Cypher/ZASP—Expression of Cypher/ZASP was investigated by RT-PCR analyses using mRNAs from various human tissues. The combination of primers used in the RT-PCR are as follows: F1N (5′-GTGCCCTCTCACTCAACCCTCT-3′) in exon 1 and R4 (5′-GAGTCCAGCCTGTTGGGGTGG-3′) in exon 9; F2 (5′-CAGGAACGCTTCAACCCCAGTG-3′) in exon 6 and R8 (5′-GGGCTGTAGGAAGAGGCCTGG-3′) in exon 11; 365PF (5′-TTCTCCTCACTCGCCGAGG-3′) in exon 4 and R2 (5′-GCATGAATTCTGTCCCCGTCATCT-3′) in exon 7; and LIMF (5′-GAATTCAGCAGCCGGACTCCACTCT-3′) in exon 12 and R3N (5′-CCTCCCGAATCCTTTGTTGCCAG-3′) in exon 16. Subjects—Patients and family members were evaluated as described previously (12Itoh-Satoh M. Hayashi T. Nishi H. Koga Y. Arimura T. Koyanagi T. Takahashi M. Hohda S. Ueda K. Nouchi T. Hiroe M. Marumo F. Imaizumi T. Yasunami M. Kimura A. Biochem. Biophys. Res. Commun. 2002; 291: 385-393Crossref PubMed Scopus (212) Google Scholar). Thirty-four proband patients with familial DCM and 62 sporadic DCM cases were the subjects. These patients had been examined before and had no mutation in the genes for dystrophin, α-cardiac actin, desmin, lamin A/C, β-cardiac myosin heavy chain, cardiac troponin T, α-tropomyosin, vinculin, MLP, Tcap/telethonin, and titin. When a sequence variation in Cypher/ZASP was found, family relatives and 400 unrelated healthy controls were examined. Blood samples were collected after obtaining informed consent from the subjects. All patients and controls were Japanese. The protocol for research on human materials was approved by the Ethics Reviewing Committee of the Medical Research Institute, Tokyo Medical and Dental University. Genomic DNA Extraction and PCR-SSCP Analysis—DNA was extracted from peripheral blood leukocytes from each subject. Extracted DNA was subjected to PCR by using primer pairs specific to the analyzed regions. Sequences of primers are available upon request. The PCR products from the subjects were searched for sequence variations by the PCR-SSCP method (12Itoh-Satoh M. Hayashi T. Nishi H. Koga Y. Arimura T. Koyanagi T. Takahashi M. Hohda S. Ueda K. Nouchi T. Hiroe M. Marumo F. Imaizumi T. Yasunami M. Kimura A. Biochem. Biophys. Res. Commun. 2002; 291: 385-393Crossref PubMed Scopus (212) Google Scholar). When an abnormal SSCP pattern was obtained, the PCR fragment was sequenced on both strands. To confirm the D626N mutation in exon 15, TspEI was used for detection of the mutation, and the digestion products were electrophoresed in a 2% agarose gel. Yeast Two-hybrid (Y2H) Assay—The cDNA fragments of human Cypher/ZASP and PKC were obtained by RT-PCR from human adult heart cDNA. A cDNA fragment covering the third LIM domain of the wild type Cypher/ZASP (Cypher-C) was amplified using primers LIMcF (5′-GAATTCAGCACCAAGTGCCATGGCTGCGAT-3′; underlined sequences are added for cloning) and LIMR (5′-GGATCCCTACAAGTTGATGGTGTGTGC-3′), whereas an equivalent mutant cDNA fragment (Cypher-M) having a G to A substitution on the sense strand (boldfaced letters in LIMcF and LIMmF) was amplified using primers LIMmF (5′-GAATTCAGCACCAAGTGCCATGGCTGCAAT-3′) and LIMR. The cDNA fragments containing a conserved region of PKC-α (PKCA-P), PKC-ϵ (PKCE-P), and PKC-ζ (PKCZ-P) were obtained using the following primers: PKC-AF (5′-GAATTCATGGCTGACGTTTTCCCGG-3′) and PKC-AR (5′-GGATCCGCAGAGGCTGGGGACATTG-3′) (containing two C1 domains; residues 1-160); PKC-EF (5′-GAATTCATGGTAGTGTTCAATGGCCTTC-3′) and PKC-ER (5′-GGATCCACAGTTGGGAGCCACGTTG-3′) (C2 domain and two C1 domains; residues 1-297); and PKC-ZF (5′-GAATTCATGCCCAGCAGGACCGACC-3′) and PKC-ZR (5′-GGATCCGCAGGTCAGCGGGACGAGG-3′) (C1 domain; residues 1-180), respectively. The amplicons were cloned into pCR2.1 for sequence confirmation and excised by digestion with EcoRI and BamHI. The excised DNA fragments were then cloned into the pGBKT7 vector as a bait (Cypher-CB and Cypher-MB) or into the pGADT7 vector as a prey (Cypher-CP, Cypher-MP, PKCA-P, PKCE-P, and PKCZ-P) (MATCHMAKER two-hybrid system 3; Clontech). These constructs were sequenced again to ensure that no errors were introduced and used to transform Saccharomyces cerevisiae. The transformations were tested by the filter assay for β-galactosidase (β-gal) activity according to the manufacturer's instructions (Clontech). The β-gal activities were also measured with o-nitrophenyl-β-d-galactopyranoside (Sigma) and expressed arbitrarily as means ± S.E., as reported previously (12Itoh-Satoh M. Hayashi T. Nishi H. Koga Y. Arimura T. Koyanagi T. Takahashi M. Hohda S. Ueda K. Nouchi T. Hiroe M. Marumo F. Imaizumi T. Yasunami M. Kimura A. Biochem. Biophys. Res. Commun. 2002; 291: 385-393Crossref PubMed Scopus (212) Google Scholar). Statistical differences in the β-gal activities were examined by Student's t test for paired value, and p values <0.05 were considered to be significant. Plasmid Construction and Immunoprecipitation from 293 Cells— Expression vectors for PKC-α, -β, and -ϵ and wild type Cypher1 have been described previously (24Zhou Q. Ruiz-Lozano P. Martone M.E. Chen J. J. Biol. Chem. 1999; 274: 19807-19813Abstract Full Text Full Text PDF PubMed Scopus (205) Google Scholar). The mutant Cypher1 was constructed by PCR-based mutagenesis and sequenced to ensure that no PCR errors were introduced. Plasmids containing wild type, mutant Cypher, and distinct PKCs (5 μg each) were transfected into 293 cells, which were plated in 10-cm dishes at 80% confluence via Superfect (Qiagen Inc.). Cells were harvested, and 100 μg of total protein from PKC-transfected cells was mixed with 100 μg of total protein from either wild type or mutant Cypher-transfected cells to perform immunoprecipitation assay as described (24Zhou Q. Ruiz-Lozano P. Martone M.E. Chen J. J. Biol. Chem. 1999; 274: 19807-19813Abstract Full Text Full Text PDF PubMed Scopus (205) Google Scholar). Expression of Cypher/ZASP in Human Tissues—Several different cDNA isoforms of Cypher/ZASP were reported in human and mouse striated muscle (24Zhou Q. Ruiz-Lozano P. Martone M.E. Chen J. J. Biol. Chem. 1999; 274: 19807-19813Abstract Full Text Full Text PDF PubMed Scopus (205) Google Scholar, 25Faulkner G. Pallavicini A. Formentin E. Comelli A. Ievolella C. Trevisan S. Bortoletto G. Scannapieco P. Salamon M. Mouly V. Valle G. Lanfranchi G. J. Cell Biol. 1999; 146: 465-475Crossref PubMed Scopus (188) Google Scholar, 26Passier R. Richardson J.A. Olson E.N. Mech. Dev. 2000; 92: 277-284Crossref PubMed Scopus (65) Google Scholar). These isoforms are generated by alternative splicing of a single gene (Fig. 1A, Ensembl gene identification number ENSG00000122367). To investigate the alternative splice pattern in various human tissues, we performed RT-PCR analysis. As shown in Fig. 1B, human Cypher/ZASP expressed several transcripts. There were two different PCR products from exon 1 to 9; a short form (Cypher 2s; nomenclature is according to Ref. 27Huang C. Zhou Q. Liang P. Hollander M.S. Sheikh F. Li X. Greaser M. Shelton G.D. Evans S. Chen J. J. Biol. Chem. 2003; 278: 7360-7365Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar) was preferentially expressed in the skeletal muscle, whereas a long form (Cypher 2c) was abundant in the fetal heart. On the other hand, transcripts covering exons 4-7 were expressed preferentially in the fetal heart. Two different transcripts spanning exons 6-11 were generated; the longer form (corresponding to Cypher 1s) was expressed in the heart. Several other combinations of primers were used in addition to investigate the alternative splicing (for example, exons 4-11; data not shown). The RT-PCR products were sequenced to confirm that there were at least six different splice variants in human (Fig. 1C) as was reported in mouse (27Huang C. Zhou Q. Liang P. Hollander M.S. Sheikh F. Li X. Greaser M. Shelton G.D. Evans S. Chen J. J. Biol. Chem. 2003; 278: 7360-7365Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar). On the other hand, ZASP variant 2, which was reported to have a deletion of 31 base pairs (25Faulkner G. Pallavicini A. Formentin E. Comelli A. Ievolella C. Trevisan S. Bortoletto G. Scannapieco P. Salamon M. Mouly V. Valle G. Lanfranchi G. J. Cell Biol. 1999; 146: 465-475Crossref PubMed Scopus (188) Google Scholar), presumably as a result of a splice to a minor acceptor site in exon 11, could not be identified in this study with any combinations of primers (data not shown). Identification of a Missense Mutation in the Cypher/ZASP Gene in a DCM Family—Sequence variations in the Cypher/ZASP gene were searched in 96 patients with DCM, and five different variations were identified (Fig. 1A). These include a T to C transition in an intron (-13 in intron 6) and four variations in the exons (GTC to ATC at codon 55 in exon 2, GTC to ATC at codon 588 in exon 14, GAT to AAT at codon 626 in exon15, and CAT to CAC at codon 644 in exon 15; codon numbers are from Cypher/ZASP 1c in Fig. 1C). Among the variations leading to amino acid replacement, V55I and V588I were polymorphisms because they were also found in unrelated healthy controls. In contrast, the D626N variation identified in a proband patient of familial DCM (designated II-6; Fig. 2, A and C) was not found in 400 unrelated healthy controls. A family study showed that the D626N mutation was present in all affected members tested (Fig. 2, B and D). The mutation was located at the fifth position next to a constant cystein (29Sanchez-Garcia I. Rabbitts T.H. Trends Genet. 1994; 10: 315-320Abstract Full Text PDF PubMed Scopus (322) Google Scholar) in the third LIM domain, and this position was occupied exclusively by acidic residues in Cypher/ZASP and other PDZ-LIM proteins and in Enigma and the Enigma homologue protein (ENH) from various species (Fig. 2E). The patients in this family developed DCM after age 50 (in the early (II-1 and II-9) or late (II-5 and II-6) fifties in male cases and at age 69 in a female case (II-2)), suggesting that the mutation was associated with late-onset DCM (Fig. 2A). Electrocardiogram findings of the affected individuals demonstrated no primary conduction defect. It was interesting to note that a sister (II-4) had the mutation but did not suffer from DCM, although she had been affected with cerebellar ataxia (CA). The CA was initially considered to be as a clinical consequence of her carrying the mutation; however, this possibility was ruled out because another brother (II-3) who also suffered from CA did not have the mutation. No sign of skeletal muscle involvement was noted in the DCM patients, although a muscle biopsy could not be performed because consent was not obtained. Altered Interaction of LIM-LIM and LIM-PKC Binding Due to the D626N Mutation—Because the D626N mutation changes the acidic residue conserved in the PDZ-LIM protein family, we tested the functional alteration using Y2H assays. Wild type or mutant bait plasmid containing the third LIM domain of Cypher/ZASP (Cypher-CB or Cypher-MB, respectively) was co-transformed with a prey plasmid containing wild type Cypher/ZASP (Cypher-CP), mutant Cypher/ZASP (Cypher-MP), PKC-α (PKCA-P), PKC-ϵ (PKCE-P), or PKC-ζ (PKCZ-P) (Fig. 3A). In the test for LIM-LIM homodimeric interaction, the β-gal activity of colonies containing Cypher-CB and Cypher-CP was 1.447 ± 0.094, whereas that of Cypher-MB and Cypher CP was significantly low (0.714 ± 0.075, p < 0.001) (Fig. 3B). Similarly, β-gal activity in colonies of Cypher-CB and Cypher-MP was low (0.510 ± 0.024, p < 0.001) (data not shown). On the other hand, the Y2H assays showed that the β-gal activity in the colonies of Cypher-MB and PKCA-P was significantly higher than that of Cypher-CB and PKCA-P (1.072 ± 0.108 versus 0.747 ± 0.094, p < 0.05). In addition, the β-gal activity obtained from the mutant LIM and PKC-ϵ interaction was significantly higher than that of the normal LIM and PKC-ϵ interaction (0.872 ± 0.054 versus 0.562 ± 0.036, p < 0.001). Similarly, the β-gal activity for the mutant LIM-PKC-ζ interaction was significantly higher than that for the normal LIM-PKC-ζ interaction (0.554 ± 0.026 versus 0.325 ± 0.010, p < 0.001) (Fig. 3B). We investigated, through an independent approach, whether the mutation would affect the LIM-PKC interaction. In the pull-down experiments, a mutation equivalent to human D626N was introduced into mouse Cypher/ZASP (24Zhou Q. Ruiz-Lozano P. Martone M.E. Chen J. J. Biol. Chem. 1999; 274: 19807-19813Abstract Full Text Full Text PDF PubMed Scopus (205) Google Scholar). Western blot analysis of immunoprecipitates of wild type or mutant Cypher with HA-tagged PKCs (PKC-α, PKC-β, and PKC-ϵ) revealed that, despite equal expression of genes, mutant Cypher had a higher affinity to the PKCs than the normal Cypher (1.96 ± 0.03-fold, p < 0.01 for PKC-α; 1.38 ± 0.12-fold, p < 0.02 for PKC-β; and 1.50 ± 0.29-fold, p < 0.04 for PKC-ϵ) (Fig. 3, C-E). In this study, a Cypher/ZASP mutation in the third LIM domain was found in a DCM family, and the mutation altered the function of the LIM domain (i.e. a decrease in dimeric binding while the binding to PKCs was increased). All affected members had the mutation, but one female carrying the mutation did not suffer from DCM (II-4, 65 years old). She might develop DCM later in life, because the DCM phenotype was late-onset and the eldest sister (II-2) developed the disease at the age of 69. Another reason why II-4 may not develop DCM might also be because she could not exercise for several years due to CA, since DCM due to the Z-disc abnormality can be exacerbated by cardiac stress, as demonstrated in MLP-deficient mice (13Knoll R. Hoshijima M. Hoffman H.M. Person V. Lorenzen-Schmidt I. Bang M.L. Hayashi T. Shiga N. Yasukawa H. Schaper W. McKenna W. Yokoyama M. Schork N.J. Omens J.H. McCulloch A.D. Kimura A. Gregorio C.C. Poller W. Schaper J. Schultheiss H.P. Chien K.R. Cell. 2002; 111: 943-955Abstract Full Text Full Text PDF PubMed Scopus (636) Google Scholar). The Cypher/ZASP isoforms have a PDZ domain and three LIM domains. These structures are highly homologous to other PDZ-LIM proteins such as Enigma and ENH (30Wu R.Y. Gill G.N. J. Biol. Chem. 1994; 269: 25085-25090Abstract Full Text PDF PubMed Google Scholar, 31Kuroda S. Tokunaga C. Kiyohara Y. Higuchi O. Konishi H. Mizuno K. Gill G.N. Kikkawa U. J. Biol. Chem. 1996; 271: 31029-31032Abstract Full Text Full Text PDF PubMed Scopus (217) Google Scholar). The ability of Cypher/ZASP isoforms to bind to α-actinin through the PDZ domain and various PKC subtypes via the LIM domains in the Z-discs of the cardiac muscle imply that Cypher/ZASP might play a role in stretch response. The LIM domain consists of 50-60 amino acid residues and participates in protein-protein interactions (29Sanchez-Garcia I. Rabbitts T.H. Trends Genet. 1994; 10: 315-320Abstract Full Text PDF PubMed Scopus (322) Google Scholar). The mutant LIM showed increased binding affinity to PKCs in both Y2H assays and pull-down assays, although the extent of the increased affinity was different between these assays. This could be because one human LIM domain was used in the Y2H assays, and three mouse LIM domains were used for the pull-down assays. Nevertheless, our results demonstrated that the interaction of the Cypher/ZASP LIM domain to PKC, especially the ϵ subtype, was augmented by the mutation. PKCs localize in nucleus, perinucleus, cytosol, and Z-discs in the cardiomyocytes (32Mochily-Rosen D. Gordon A.S. FASEB J. 1998; 12: 35-42Crossref PubMed Google Scholar, 33Mackay K. Mochly-Rosen D. J. Mol. Cell. Cardiol. 2001; 33: 1301-1307Abstract Full Text PDF PubMed Scopus (129) Google Scholar). Upon activation by lipid-derived second messengers, PKCs are known to translocate from one cell component to another via the function of anchoring proteins, and the proteins anchoring the inactivated PKCs are referred to as RICKs (receptors for inactivated protein kinase C), whereas those anchoring the activated PKCs are referred to as RACKs (receptors for activated protein kinase C), and each PKC isozyme associates with specific anchoring proteins to mediate isozyme-specific PKC functions (34Robia S.L. Ghanta J. Robu V.G. Walker J.W. Biophys. J. 2001; 80: 2140-2151Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar). We have shown here, as reported previously (24Zhou Q. Ruiz-Lozano P. Martone M.E. Chen J. J. Biol. Chem. 1999; 274: 19807-19813Abstract Full Text Full Text PDF PubMed Scopus (205) Google Scholar), that Cypher/ZASP interacts with PKCs. Although Cypher/ZASP is not a bona fide RICK or RACK, because it can bind multiple PKC isozymes it will be of interest to determine whether Cypher/ZASP has RICK- or RACK-like functions. On the other hand, it has been shown that the expression of PKC-ϵ and its translocation from a cytosol fraction to a membrane fraction is increased during the transition from compensated hypertrophy to congestive heart failure (23Jalili T. Takeishi Y. Song G. Ball N.A. Howles G. Walsh R.A. Am. J. Physiol. 1999; 277: H2298-H2304PubMed Google Scholar). Several studies have demonstrated that the N-terminal part of PKC-ϵ interacts selectively with a specific isotype of RACKs, RACK2 (32Mochily-Rosen D. Gordon A.S. FASEB J. 1998; 12: 35-42Crossref PubMed Google Scholar, 33Mackay K. Mochly-Rosen D. J. Mol. Cell. Cardiol. 2001; 33: 1301-1307Abstract Full Text PDF PubMed Scopus (129) Google Scholar), and it was reported that PKC-ϵ·RACK2 interactions played a key role in cardioprotection against ischemic stress (35Mochly-Rosen D. Wu G. Hahn H. Osinska H. Liron T. Lorenz J.N. Yatani A. Robbins J. Dorn II, G.W. Circ. Res. 2000; 86: 1173-1179Crossref PubMed Scopus (179) Google Scholar, 36Pass J.M. Zheng Y. Wead W.B. Zhang J. Li R.C. Bolli R. Ping P. Am. J. Physiol. Heart Circ. Physiol. 2001; 280: H946-H995Crossref PubMed Google Scholar). In contrast, disruption of PKC-ϵ·RACK2 interaction has been shown to inhibit cardiac cell contraction (37Johnson J.A. Gray M.O. Chen C.H. Mochly-Rosen D. J. Biol. Chem. 1996; 271: 24962-24966Abstract Full Text Full Text PDF PubMed Scopus (343) Google Scholar) and accelerate cell death (38Liu G.S. Cohen M.V. Mochly-Rosen D. Downey J.M. J. Mol. Cell. Cardiol. 1999; 31: 1937-1948Abstract Full Text PDF PubMed Scopus (221) Google Scholar). It is likely that the translocation of PKC-ϵ in the heart may be involved in myocardial remodeling and ultimately lead to heart failure. Because the N-terminal part of PKC-ϵ interacts with both Cypher/ZASP and RACK2, an interesting hypothesis may be that the augmentation of binding affinity between Cypher/ZASP and PKC-ϵ by the Cypher/ZASP mutation might, in turn, reduce the amount of a PKC-ϵ·RACK2 complex and cause early progression to heart failure. This hypothesis should be tested in future studies. In summary, we identified a Cypher/ZASP mutation associated with DCM that increased the LIM-PKCs interaction. These observations suggest an association between DCM and the inherited abnormality involved in signal transduction, which is consistent with the findings of mutations in phospholamban and MLP. Although the molecular mechanisms of DCM due to the Cypher/ZASP mutation remain to be elucidated, and other DCM patients should be screened for Cypher/ZASP mutation to formally confirm the causality of the mutation in DCM, our observations imply that the cardiac dysfunction might be associated not only with the alteration in each sarcomeric interaction but also with the altered recruitment of molecules participating in intracellular signaling. We are grateful to Drs. H. Nishi, T. Koyanagi, Y. Koga, H. Toshima, T. Imaizumi, R. Nagai, Y. Yazaki, T. Izumi, A. Matsumori, and S. Sasayama for participating in the clinical evaluation and blood sampling of patients with DCM. We thank Dr. S. Kuroda and Dr. F. Sheikh for providing us with the PKC constructs used in the pull-down experiments and for critical reading of the manuscript, respectively. We also greatly appreciate Dr. C. Prioleau for helpful advice and critical reading of the manuscript.
Referência(s)