A Novel Family of Calmodulin-binding Transcription Activators in Multicellular Organisms
2002; Elsevier BV; Volume: 277; Issue: 24 Linguagem: Inglês
10.1074/jbc.m200268200
ISSN1083-351X
AutoresNicolas Bouché, Ariel Scharlat, Wayne A. Snedden, David Bouchez, Hillel Fromm,
Tópico(s)Fungal and yeast genetics research
ResumoScreening of cDNA expression libraries derived from plants exposed to stress, with35S-labeled recombinant calmodulin as a probe, revealed a new family of proteins containing a transcription activation domain and two types of DNA-binding domains designated the CG-1 domain and the transcription factor immunoglobulin domain, ankyrin repeats, and a varying number of IQ calmodulin-binding motifs. Based on domain organization and amino acid sequence comparisons, similar proteins, with the same domain organization, were identified in the genomes of other multicellular organisms including human, Drosophila, and Caenorhabditis, whereas none were found in the complete genomes of single cell eukaryotes and prokaryotes. This family of proteins was designated calmodulin-binding transcription activators (CAMTAs). Arabidopsis thalianacontains six CAMTA genes (AtCAMTA1–AtCAMTA6). The transcription activation domain of AtCAMTA1 was mapped by testing a series of protein fusions with the DNA-binding domain of the bacterial LexA transcription factor and two reporter genes fused to LexA recognition sequences in yeast cells. Two human proteins designated HsCAMTA1 and HsCAMTA2 were also shown to activate transcription in yeast using the same reporter system. Subcellular fractionation ofArabidopsis tissues revealed the presence of CAMTAs predominantly in the nucleus. Calmodulin binding assays identified a region of 25 amino acids capable of binding calmodulin with high affinity (Kd = 1.2 nm) in the presence of calcium. We suggest that CAMTAs comprise a conserved family of transcription factors in a wide range of multicellular eukaryotes, which possibly respond to calcium signaling by direct binding of calmodulin. Screening of cDNA expression libraries derived from plants exposed to stress, with35S-labeled recombinant calmodulin as a probe, revealed a new family of proteins containing a transcription activation domain and two types of DNA-binding domains designated the CG-1 domain and the transcription factor immunoglobulin domain, ankyrin repeats, and a varying number of IQ calmodulin-binding motifs. Based on domain organization and amino acid sequence comparisons, similar proteins, with the same domain organization, were identified in the genomes of other multicellular organisms including human, Drosophila, and Caenorhabditis, whereas none were found in the complete genomes of single cell eukaryotes and prokaryotes. This family of proteins was designated calmodulin-binding transcription activators (CAMTAs). Arabidopsis thalianacontains six CAMTA genes (AtCAMTA1–AtCAMTA6). The transcription activation domain of AtCAMTA1 was mapped by testing a series of protein fusions with the DNA-binding domain of the bacterial LexA transcription factor and two reporter genes fused to LexA recognition sequences in yeast cells. Two human proteins designated HsCAMTA1 and HsCAMTA2 were also shown to activate transcription in yeast using the same reporter system. Subcellular fractionation ofArabidopsis tissues revealed the presence of CAMTAs predominantly in the nucleus. Calmodulin binding assays identified a region of 25 amino acids capable of binding calmodulin with high affinity (Kd = 1.2 nm) in the presence of calcium. We suggest that CAMTAs comprise a conserved family of transcription factors in a wide range of multicellular eukaryotes, which possibly respond to calcium signaling by direct binding of calmodulin. calmodulin-binding transcription activator ankyrin transcription factor immunoglobulin-like domain nuclear localization signal expressed sequence tag phenylmethylsulfonyl fluoride glutathioneS-transferase o-nitrophenyl-β-d-galactoside complete minimal 1,4-piperazinediethanesulfonic acid Despite the completion or near completion of the genome sequence of several prokaryotes and eukaryotes including human, fly, nematode, and higher plants, the function of a large proportion of the genes remains unknown. Transcription factors play a crucial role in regulating every aspect of the organism's life cycle and are fit to respond to signals originating from within and without the organism. Not surprisingly, a high proportion of eukaryote genomes encode transcription factors, estimated to be ∼2,000 in humans (roughly 5% of the genome) where the gene expression machinery seems to be particularly complex (1Tupler R. Perini G. Green M.R. Nature. 2001; 409: 832-833Crossref PubMed Scopus (305) Google Scholar). In Arabidopsis, a remarkable estimate of 3,000 genes (11.8% of the genome) were suggested to be involved in different aspects of transcription regulation (2The Arabidopsis Genome InitiativeNature. 2000; 408: 796-815Crossref PubMed Scopus (7199) Google Scholar). These include many new factors whose roles in gene expression are unknown. In mammalian cells, Ca2+ and the Ca2+ receptor calmodulin are involved in regulating gene transcription. For example, expression of the c-fos gene is mediated by Ca2+signals through two DNA regulatory elements, the cyclic AMP-response element and the serum-response element. Increase in nuclear Ca2+ concentration stimulates cyclic AMP-response element-dependent gene expression, whereas elevation of cytosolic Ca2+ activates transcription via the serum-response element (3Hardingham G.E. Chawla S. Johnson C.M. Bading H. Nature. 1997; 385: 260-265Crossref PubMed Scopus (642) Google Scholar). Thus, nuclear and cytoplasmic Ca2+ control transcription by distinct mechanisms. Certain transcription factors are selectively activated in response to distinct Ca2+ signal duration and amplitude. NF-κB and c-Jun N-terminal kinase are activated by a large transient cytoplasmic Ca2+ rise, whereas NFAT is activated by a low, sustained plateau (4Dolmetsch R.E. Lewis R.S. Goodnow C.C. Healy J.I. Nature. 1997; 386: 855-858Crossref PubMed Scopus (1562) Google Scholar). Therefore, different types of oscillating Ca2+signals can modulate downstream transcription factor activity. Ca2+ can also directly bind and regulate transcription factors. For example, the DREAM protein contains four EF-hand motifs and represses transcription (5Carrion A.M. Link W.A. Ledo F. Mellstrom B. Naranjo J.R. Nature. 1999; 398: 80-84Crossref PubMed Scopus (492) Google Scholar) as DREAM affinity for DNA is reduced upon binding to Ca2+. Calmodulin modulates the nuclear activity of various proteins, like the mammalian family of nuclear Ca2+/calmodulin-dependent protein kinases (6Heist E.K. Schulman H. Cell Calcium. 1998; 23: 103-114Crossref PubMed Scopus (86) Google Scholar). When activated by Ca2+/calmodulin, calmodulin-kinase II can specifically decode the frequency of Ca2+ spikes into distinct levels of kinase activity (7de Koninck P. Schulman H. Science. 1998; 279: 227-230Crossref PubMed Scopus (1086) Google Scholar) and phosphorylates a large number of target proteins. In plants, recent advances revealed that post-translational modifications of CaM53, a novel petunia calmodulin isoform, could modify the subcellular localization of the protein and direct it to the nucleus or the plasma membrane (8Rodriguez-Concepcion M. Yalovsky S. Zik M. Fromm H. Gruissem W. EMBO J. 1999; 18: 1996-2007Crossref PubMed Scopus (130) Google Scholar). In addition, certain transcription factors of the basic helix-loop-helix family were shown to bind calmodulin, thus inhibiting their DNA-binding properties by masking the DNA-binding domain (9Onions J. Hermann S. Grundstrom T. Biochemistry. 2000; 39: 4366-4374Crossref PubMed Scopus (37) Google Scholar, 10Onions J. Hermann S. Grundstrom T. J. Biol. Chem. 1997; 272: 23930-23937Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar, 11Corneliussen B. Holm M. Waltersson Y. Onions J. Hallberg B. Thornell A. Grundstrom T. Nature. 1994; 368: 760-764Crossref PubMed Scopus (144) Google Scholar). Therefore, interaction of calmodulin with transcription factors is a mechanism by which transcriptional activity may be regulated in response to Ca2+ signals originating from a variety of stimuli. We used protein-protein interaction for library screenings to identify plant calcium/calmodulin-binding proteins. One family of calmodulin-binding proteins, designated the calmodulin-binding transcription activator (CAMTA)1 family, which has been identified in the course of this study, resembles a group of putative transcription activators recently identified in the human genome (12Lander E.S. Linton L.M. Birren B. Nusbaum C. Zody M.C. Baldwin J. Devon K. Dewar K. Doyle M. FitzHugh W. Funke R. Gage D. Harris K. Heaford A. Howland J. et al.Nature. 2001; 409: 860-921Crossref PubMed Scopus (17950) Google Scholar). These were reported to contain a novel DNA-binding domain termed CG-1, a transcription factor immunoglobulin (TIG)-like DNA-binding domain, and ankyrin repeats. However, the properties of these proteins as transcription activators have never been tested, and the extent of their distribution in eukaryotes has not been investigated. Here we investigated the properties of members of this family of putative transcription factors from Arabidopsisand humans, demonstrating the ability of both to activate transcription in yeast cells. We also expanded the bioinformatic analysis of this protein family to reveal their occurrence and domain organization in multicellular organisms. A Brassica napus library from leaves of drought-stressed plants (13Downing W.L. Mauxion F. Fauvarque M.O. Reviron M.P. de Vienne D. Vartanian N. Giraudat J. Plant J. 1992; 2: 685-693PubMed Google Scholar) was kindly provided by J. Giraudat (Institut des Sciences Végétales, CNRS, Gif-sur-Yvette, France). Expression library screening was performed with35S-labeled recombinant calmodulin from petunia (CaM81; GenBankTM accession number S70768) as a probe (14Baum G. Chen Y. Arazi T. Takatsuji H. Fromm H. J. Biol. Chem. 1993; 268: 19610-19617Abstract Full Text PDF PubMed Google Scholar). Several databases were used to retrieve and compare sequences: the GenBankTM data base (NCBI server; www.ncbi.nlm.nih.gov), the Arabidopsis Genome Initiative data base (TAIR server; www.arabidopsis.org), and the Berkeley fly data base (BDGP server; www.fruitfly.org). Domain identification and comparisons were done with the InterPro data base (www.ebi.ac.uk/interpro; Ref. 15Apweiler R. Attwood T.K. Bairoch A. Bateman A. Birney E. Biswas M. Bucher P. Cerutti L. Corpet F. Croning M.D. Durbin R. Falquet L. Fleischmann W. Gouzy J. Hermjakob H. Hulo N. Jonassen I. Kahn D. Kanapin A. Karavidopoulou Y. Lopez R. Marx B. Mulder N.J. Oinn T.M. Pagni M. Servant F. Nucleic Acids Res. 2001; 29: 37-40Crossref PubMed Scopus (822) Google Scholar). Multiple sequence alignments were assembled with the ClustalX program (16Thompson J.D. Gibson T.J. Plewniak F. Jeanmougin F. Higgins D.G. Nucleic Acids Res. 1997; 25: 4876-4882Crossref PubMed Scopus (35598) Google Scholar). Intron/exon junctions were predicted with the NetGene2 program in Arabidopsis thaliana (17Hebsgaard S.M. Korning P.G. Tolstrup N. Engelbrecht J. Rouze P. Brunak S. Nucleic Acids Res. 1996; 24: 3439-3452Crossref PubMed Scopus (662) Google Scholar) and NNSPLICE0.9 program in Drosophila melanogaster (18Reese M.G. Eeckman F.H. Kulp D. Haussler D. J. Comput. Biol. 1997; 4: 311-323Crossref PubMed Scopus (1394) Google Scholar). Expressed sequence tags (ESTs) corresponding to complete cDNA clones were obtained from the Arabidopsis Biological Resource Center (ABRC at TAIR server) for AtCAMTA1 (clone H9D3T7) and AtCAMTA5 (clone 4G3T7P) and the Kazusa DNA Research Institute for AtCAMTA2 (clone AV528637), HsCAMTA1 (clone KIAA0833), and HsCAMTA2 (clone KIAA0909). TheSf9 cell line of Spodoptera frugiperda(19O'Reilly D.R. Miller L.K. Luckow V.A. Baculovirus Expression Vectors: A Laboratory Manual. W. H. Freeman and Co., New York1992Google Scholar) was maintained as a monolayer culture at 27 °C in Grace medium (19O'Reilly D.R. Miller L.K. Luckow V.A. Baculovirus Expression Vectors: A Laboratory Manual. W. H. Freeman and Co., New York1992Google Scholar) supplemented with 10% fetal calf serum (Biological Industries). Cells were recultured every 4 days to maintain a density ranging from ∼5 × 105 to 2.5 × 106 cells/ml. The full-length AtCAMTA1 was excised from the EST clone H9D3T7 (ABRC) with SalI and XbaI and inserted in a pFastBac1 donor plasmid (Invitrogen; Ref. 20Luckow V.A. Lee S.C. Barry G.F. Olins P.O. J. Virol. 1993; 67: 4566-4579Crossref PubMed Google Scholar) downstream of the promoter of the viral polyhedrin gene. Similarly, the BnCAMTA cDNA sequence, coding for a partial BnCAMTA protein (Ile1–Lys688), was amplified by PCR with aPfu DNA polymerase (Promega) and cloned into theEcoRI and SalI sites of pFastBac1. Plasmids were then transformed into DH10BAC Escherichia coli cells (Invitrogen) for transposition into the Bacmid. The screening and isolation of recombinant Bacmid DNA were done according to the manufacturer's instructions. Sf9 cells were transfected with recombinant Bacmid DNA using CellFECTIN (Invitrogen). Recombinant baculoviruses were harvested 72 h after the start of transfection. Sf9 cells were layered at a density of 5 × 106 cells/90-mm plate and infected with high titer recombinant baculoviruses. After 3 days of incubation at 27 °C, cells were harvested by centrifugation at 500 ×g for 10 min, washed once with phosphate-buffered saline centrifuged for 10 min at 500 × g and resuspended (1 ml/plate) in extraction buffer containing 100 mm Tris-HCl, pH 7.5, 10% glycerol, 1 mm EDTA and 1 mm PMSF. Cells were broken in liquid nitrogen, or by the addition of 0.5% Nonidet P-40. Cell lysates were centrifuged at 4 °C, 14,000 ×g for 15 min, and the supernatant was collected. Protein concentrations were determined with a Bradford reagent (Bio-Rad). To prepare polyclonal antibodies against the N-terminal part of AtCAMTA1, the CG-1 domain of AtCAMTA1 (Val2–Lys148) was fused in frame to the GST coding sequence in the BamHI and EcoRI sites of the pGEX-3X vector (Amersham Biosciences). To prepare antibodies against the ANK-repeat region, the corresponding sequence from BnCAMTA (Gln588–Gly687) was amplified as described above and subcloned in the NdeI and SalI sites of a pET12c vector (Novagen, Inc.). These constructs were introduced in E. coli strain BL21(DE3)pLysS to produce the recombinant proteins as described (14Baum G. Chen Y. Arazi T. Takatsuji H. Fromm H. J. Biol. Chem. 1993; 268: 19610-19617Abstract Full Text PDF PubMed Google Scholar). Inclusion bodies from the insoluble fraction of the bacterial cells, containing most of the recombinant proteins, were purified and solubilized in sample buffer (21Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207472) Google Scholar), and proteins were separated by SDS-PAGE. An acrylamide band containing the recombinant protein (either the ANK repeat region or the N-terminal part of AtCAMTA1 fused to GST) was excised from the gel, crushed, and mixed (1:1) with complete Freund's adjuvant (Sigma). Three ml of the mixture containing 100 μg of recombinant protein were injected into two rabbits. Each rabbit was given four booster injections about 2 weeks apart. The rabbits were bled about 10 days after each injection. The serum containing anti-CG-1 antibodies was depleted from the antibodies against GST by passing it on a GST column (Pierce). Double stranded calf thymus DNA-cellulose (Sigma) and heparin-Sepharose CL-6B (Amersham Biosciences) were pre-equilibrated with the following buffer: 25 mmHEPES-KOH, pH 7.5, 40 mm KCl, 0.1 mm EDTA, 10% glycerol, 1 mm dithiothreitol, and 1 mm PMSF. Soluble fraction proteins obtained from Sf9 insect cells were dialyzed against this buffer at 4 °C using VSWP-25 filters (Millipore Corp.) and loaded on either column. After washing with 10 column volumes of buffer, proteins were eluted with the same buffer containing KCl as indicated. Aerial parts from 4-week-old A. thaliana Columbia ecotype plants were grown in vitro under the following conditions: photoperiod, 16-h day (100–150 microeinsteins/m2/s)/8-h night; temperature, 20 °C day/15 °C night; humidity, 70%. They were frozen and ground in liquid nitrogen to a fine powder with a mortar and pestle. All subsequent steps were carried out at 4 °C. Part of this powder was homogenized with plant extraction buffer (100 mm Tris-HCl, pH 7.5, 10 mm EDTA, 10 mm β-mercaptoethanol, 10% glycerol, 1 mm PMSF, 2 μg/ml leupeptin, 2 μg/ml pepstatin, and 2 μg/ml aprotinin). This extract was filtered through two layers of Miracloth and centrifuged at 10,000 × gfor 15 min. The insoluble and soluble fractions were collected. The rest of the powder was mixed with nuclei isolation buffer (22Foster R. Gasch A. Kay S. Chua N.-H. Koncz C. Chua N.-H. Schell J. Methods in Arabidopsis Research. World Scientific Publishing Co. Pte. Ltd., Singapore1992: 378-392Crossref Google Scholar) (1m hexylene glycol, 10 mm PIPES-KOH, pH 7, 10 mm MgCl2, 0.2% Triton X-100, 5 mmβ-mercaptoethanol, and 1 mm PMSF) and filtered through two layers of Miracloth and one 100-μm nylon mesh. The extract was centrifuged at 2,000 × g for 10 min. The pellet was resuspended in nuclei wash buffer (0.5 m hexylene glycol, 10 mm PIPES-KOH, pH 7, 10 mm MgCl2, 0.2% Triton X-100, 5 mm β-mercaptoethanol, and 1 mm PMSF) and centrifuged again at 3,000 ×g for 10 min. The pellet was then washed two more times and finally resuspended in 5 ml of nuclei wash buffer. Nuclei were further purified in a discontinuous Percoll gradient (23Luthe D.S. Quatrano R.S. Plant Physiol. 1980; 65: 305-308Crossref PubMed Google Scholar). The gradient contained 5-ml layers of 40, 60, and 80% (v/v) Percoll on a 5-ml layer of 2 m sucrose cushion. The Percoll contained 0.5m hexylene glycol, 10 mm PIPES-KOH, pH 7, 10 mm MgCl2, and 0.2% Triton X-100. The gradient was centrifuged at 4,000 × g in a Sorvall HB4 swinging bucket rotor for 30 min. Most of the nuclei banded in the 80% Percoll, just above the sucrose cushion. They were removed with a Pasteur pipette, washed twice with 15 ml of nuclei wash buffer to remove Percoll, and centrifuged again at 3,000 × g for 10 min. This nuclei-enriched fraction was resuspended in protein sample buffer (21Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207472) Google Scholar) to be loaded onto SDS-PAGE gels for Western blot analysis. The completeAtCAMTA1, HsCAMTA1, and HsCAMTA2cDNA plus 12 different AtCAMTA1 regions (corresponding to residues 1–147, 1–230, 1–680, 1–820, 148–1008, 231–1008, 681–1008, 821–1008, 231–680, 231–397, 398–566, and 567–680 for constructs 2–13, respectively) were fused in frame with the coding sequence of the DNA-binding domain of LexA in the pEG202 vector (OriGene) carrying the HIS3 selectable marker. These plasmids were then introduced in EGY48 yeast strain (MATa trp1 ura3 his3 LEU2::plexAop6-LEU2; Ref. 24Estojak J. Brent R. Golemis E.A. Mol. Cell. Biol. 1995; 15: 5820-5829Crossref PubMed Scopus (450) Google Scholar) by lithium acetate transformation, together with the pSH18.34 vector (OriGene) carrying the URA3 selectable marker and the lacZreporter gene fused to eight lexA operators. Yeast transformants were selected on plates containing complete minimal (CM) dropout medium without Ura and His but with Glc as a unique carbon source. Individual yeast colonies were then transferred to liquid medium (CM dropout +Glc, −His, −Ura) and grown to late log phase. These single-colony-derived cultures were tested for the production of β-galactosidase with the chromogenic substrateo-nitrophenyl-β-d-galactoside (ONPG) or for their ability to grow on plates containing CM dropout medium without uracil, histidine, and leucine. To study the interaction between calmodulin and the calmodulin-binding domain of AtCAMTA1 in vivo, a petunia calmodulin (CaM81; GenBankTM accession number S70768) was fused in frame to the B42 transcription activator domain in the pJG4–5 vector (Origene) that carries the TRP1 selectable marker. This construct was introduced into a EGY48 yeast strain, and transformants were selected as previously described, except that Trp was omitted from the medium. 3 ml of selective medium were inoculated with 15–30 μl of saturated culture grown to late log phase. Yeast cells were grown overnight at 28 °C, under agitation. Cells were then centrifuged 5 min at 2,500 rpm and resuspended in 3 ml of Z buffer (60 mm Na2HPO4·7H2O, 40 mm NaH2PO4·H2O, 10 mm KCl, 1 mmMgSO4·7H2O, 50 mmβ-mercaptoethanol final concentrations; adjusted to pH 7.0) and then placed on ice. A600 was determined for each sample, and the following two reaction tubes were set up (1 ml each) by mixing (a) 100 μl of cells with 900 μl of Z buffer and (b) 50 μl of cells with 950 μl of Z buffer. To break the cells, one drop of 0.1% SDS and two drops of chloroform were added to each sample, which were then vortexed 10–15 s and incubated for 15 min in a 30 °C water bath. 0.2 ml of 4 mg/ml ONPG were added, and samples were vortexed for 5 s and placed in a 30 °C water bath, at which point timing was begun. When a medium yellow color had developed, the reaction was stopped by adding 0.5 ml of 1 m Na2CO3, and the time was noted. Cells were centrifuged for 5 min at 2,500 rpm, and A420 plus A550 of the supernatant were determined. To calculate β-galactosidase units, the following equation was applied, U=(1000∗(A420−(1.75∗A550)))/(t∗v∗A600)(Eq. 1) where t represents the time of reaction (min), v is the volume of culture used in the assay (ml), A600 represents the cell density at the start of the assay; A420 is a combination of absorbence by o-nitrophenol and light scattering by cell debris, and A550 is the light scattering by cell debris. All of the measurements were done in triplicate, and the β-galactosidase units counted results from an average number. Controls used are encoded by plasmids pSH17.4 (positive control) and pRFHM1 (negative control) commercialized by OriGene. To confirm that proteins were expressed at the same level in yeast, cells were disrupted with glass beads (Sigma), and total proteins were extracted and separated on SDS-PAGE. LexA DNA-binding domain fusions were detected by Western blots using a monoclonal antibody raised against LexA (CLONTECH). DNA fragments derived from the AtCAMTA1 cDNA from residues 682–1007, 682–897, 682–869, 863–1007, 913–1007, 823–897, 823–869, and 863–897 (constructs 1–8, respectively) were fused in frame to the coding sequence of GST in a pGEX-3X vector (Amersham Biosciences). Fusion proteins were expressed in E. coli XL1-Blue strain. After induction of the expression, total proteins were extracted, separated by SDS-PAGE, and electrotransferred to nitrocellulose membranes. [35S]Calmodulin was prepared and used as described (25Arazi T. Baum G. Snedden W.A. Shelp B.J. Fromm H. Plant Physiol. 1995; 108: 551-561Crossref PubMed Scopus (108) Google Scholar). Following autoradiography of blots, immunodetection of proteins by anti-GST antibodies on the same blots was performed as described (25Arazi T. Baum G. Snedden W.A. Shelp B.J. Fromm H. Plant Physiol. 1995; 108: 551-561Crossref PubMed Scopus (108) Google Scholar). For nondenaturing PAGE, samples containing 120 pmol of bovine brain calmodulin (Sigma) and different quantities of high pressure liquid chromatography-purified synthetic peptides in 100 mmTris-HCl (pH 7.2), and 0.1 mm CaCl2, making a total volume of 30 μl, were incubated for 1 h at room temperature. Samples were analyzed by nondenaturing gel electrophoresis as previously described (25Arazi T. Baum G. Snedden W.A. Shelp B.J. Fromm H. Plant Physiol. 1995; 108: 551-561Crossref PubMed Scopus (108) Google Scholar). For fluorescence measurements of peptide interactions with dansyl-calmodulin, dansylated bovine calmodulin (400 nm; Sigma) was incubated with different concentrations of synthetic peptide in 50 mm Tris-HCl (pH 7.5), 150 mm NaCl, and 0.5 mm CaCl2. After each addition of peptide, the calmodulin/peptide solution was mixed and incubated for 5 min at 23 °C. Emission fluorescence at 480 nm was then measured using a SLM AMINCO 8000 fluorimeter (SLM Instruments); excitation wavelength was at 340 nm. Each measurement was the average of three readings. The apparent dissociation constant (Kd) was determined as described (26Arazi T. Kaplan B. Fromm H. Plant Mol. Biol. 2000; 42: 591-601Crossref PubMed Scopus (85) Google Scholar). Plant adaptation to environmental stresses is mediated by Ca2+-signaling (27Knight H. Knight M.R. Trends Plant Sci. 2001; 6: 262-267Abstract Full Text Full Text PDF PubMed Scopus (806) Google Scholar) and Ca2+-responsive proteins, among them calmodulin and calmodulin-related proteins (28Snedden W.A. Fromm H. New Phytol. 2001; 151: 35-66Crossref PubMed Scopus (387) Google Scholar). To isolate cDNAs encoding calmodulin-binding proteins with a possible role in plant response to abiotic stress, we used 35S-labeled recombinant calmodulin as a probe to screen cDNA expression libraries derived from plants exposed to various stress conditions. In particular, screening of a cDNA library from Brassica napus leaves (see “Experimental Procedures”) resulted in the isolation of one clone that contained putative domains with DNA-binding properties and a domain that proved to function as a transcription activator, as will be shown. This clone was designated BnCAMTA (for B. napus calmodulin-binding transcription activator). Based on the domain organization and amino acid sequence of BnCAMTA, we identified members of the CAMTA family in various eukaryotes (Fig.1A): in nematodes, flies, and humans and also in other plants, including Arabidopsis. The latter's genome has six highly similar CAMTA genes designated AtCAMTA1–AtCAMTA6. In human, two homologous cDNA clones have been identified, designated HsCAMTA1(GenBankTM accession number XM_042323) and HsCAMTA2 (GenBankTM accession number XM_053753). These were isolated from a population of size-fractionated human brain mRNAs (29Nagase T. Ishikawa K. Suyama M. Kikuno R. Hirosawa M. Miyajima N. Tanaka A. Kotani H. Nomura N. Ohara O. DNA Res. 1998; 5: 355-364Crossref PubMed Scopus (209) Google Scholar). Gene expression profiles revealed that they are expressed in all human organs tested but highly expressed in the brain (29Nagase T. Ishikawa K. Suyama M. Kikuno R. Hirosawa M. Miyajima N. Tanaka A. Kotani H. Nomura N. Ohara O. DNA Res. 1998; 5: 355-364Crossref PubMed Scopus (209) Google Scholar). Interestingly, only one CAMTA gene was identified in the complete genomes of both Caenorhabditis elegans and D. melanogaster. In contrast, no members of the CAMTA family have been found in the complete genomes of S. cerevisiae and prokaryotes. Importantly, to date, no function has been attributed to any of the CAMTA genes identified in any organism, although based on domain organization, their relationship to transcription factors has been suggested (12Lander E.S. Linton L.M. Birren B. Nusbaum C. Zody M.C. Baldwin J. Devon K. Dewar K. Doyle M. FitzHugh W. Funke R. Gage D. Harris K. Heaford A. Howland J. et al.Nature. 2001; 409: 860-921Crossref PubMed Scopus (17950) Google Scholar). Alignment of the amino acid sequences of CAMTAs using the ClustalX program (16Thompson J.D. Gibson T.J. Plewniak F. Jeanmougin F. Higgins D.G. Nucleic Acids Res. 1997; 25: 4876-4882Crossref PubMed Scopus (35598) Google Scholar) and comparisons with protein domain databases (InterPro data base; Ref. 15Apweiler R. Attwood T.K. Bairoch A. Bateman A. Birney E. Biswas M. Bucher P. Cerutti L. Corpet F. Croning M.D. Durbin R. Falquet L. Fleischmann W. Gouzy J. Hermjakob H. Hulo N. Jonassen I. Kahn D. Kanapin A. Karavidopoulou Y. Lopez R. Marx B. Mulder N.J. Oinn T.M. Pagni M. Servant F. Nucleic Acids Res. 2001; 29: 37-40Crossref PubMed Scopus (822) Google Scholar) revealed different types of conserved regions in all CAMTAs (Fig. 1A). The conserved domains include the following: (a) a bipartite nuclear localization signal (NLS) in the N-terminal part of all CAMTA proteins; (b) a TIG domain reported to be involved in nonspecific DNA contacts in various transcription factors, like those of the Rel/NF-κB family or NFAT (30Aravind L. Koonin E.V. J. Mol. Biol. 1999; 287: 1023-1040Crossref PubMed Scopus (374) Google Scholar); (c) ankyrin (ANK) repeats, known to be involved in protein-protein interactions (31Sedgwick S.G. Smerdon S.J. Trends Biochem. Sci. 1999; 24: 311-316Abstract Full Text Full Text PDF PubMed Scopus (663) Google Scholar, 32Rubtsov A.M. Lopina O.D. FEBS Lett. 2000; 482: 1-5Crossref PubMed Scopus (110) Google Scholar) and present in a large number of functionally diverse proteins; and (d) IQ motifs, known as calmodulin-binding sites (33Rhoads A.R. Friedberg F. FASEB J. 1997; 11: 331-340Crossref PubMed Scopus (749) Google Scholar, 34Bähler M. Rhoads A. FEBS Lett. 2002; 513: 107-112Crossref PubMed Scopus (372) Google Scholar), localized in the C-terminal part of CAMTAs. These vary in number from zero (CeCAMTA) to five (AtCAMTA6). Although spacing is highly variable, overall domain organization is conserved in all proteins. We also identified a highly conserved uncharacterized domain of about 130 amino acid residues designated CG-1, containing the predicted bipartite NLS (Fig. 1B). The CG-1 domain is named after a partial cDNA clone isolated from parsley (Petroselinum crispum L.) encoding a sequence-specific DNA-binding protein (35da Costa e Silva O. Plant Mol. Biol. 1994; 25: 921-924Crossref PubMed Scopus (39) Google Scholar). Bioinformatic studies have recently revealed CG-1 domains in the human proteins HsCAMTA1 and HsCAMTA2 (12Lander E.S. Linton L.M. Birren B. Nusbaum C. Zody M.C. Baldwin J. Devon K. Dewar K. Doyle M. FitzHugh W. Funke R. Gage D. Harris K. Heaford A. Howland J. et al.Nature. 2001; 409: 860-921Crossref PubMed Scopus (17950) Google Scholar). To test the possible occurrence of similar domains in other proteins, we compared the CG-1 amino acid sequence against the nucleotide databases translated in all six re
Referência(s)