calmodulin-dependent protein kinase Iδ causes developmental abnormalities in zebrafish, Danio rerio

Archives of Biochemistry and Biophysics 517 (2012) 71–82

Contents lists available at SciVerse ScienceDirect

Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi

Knockdown of two splice variants of Ca2+/calmodulin-dependent protein kinase Id causes developmental abnormalities in zebrafish, Danio rerio q Yukako Senga, Tadashi Nagamine, Isamu Kameshita, Noriyuki Sueyoshi ⇑ Department of Life Sciences, Faculty of Agriculture, Kagawa University, Kagawa 761-0795, Japan

a r t i c l e

i n f o

Article history: Received 25 July 2011 and in revised form 4 November 2011 Available online 15 November 2011 Keywords: CaMKId Embryogenesis Gene knockdown Isoform Protein phosphorylation Zebrafish

a b s t r a c t We isolated cDNA clones for zebrafish Ca2+/calmodulin-dependent protein kinase I (zCaMKI) d isoforms by expression screening using cDNA library from embryos at 72-h post-fertilization (hpf). There are two splice variants with different C-terminal sequences, comprising of 392 and 368 amino acids, and they are designated zCaMKId-L (long form) and zCaMKId-S (short form), respectively. Although recombinant zCaMKId-L and zCaMKId-S expressed in Escherichia coli showed essentially the same catalytic properties including substrate specificities, they showed different spatial and temporal expression. Western blotting analysis using the isoform-specific antibodies revealed that zCaMKId-L clearly appeared from 36 hpf but zCaMKId-S began to appear at 60 hpf and thereafter. zCaMKId-S was predominantly expressed in brain, while zCaMKId-L was widely distributed in brain, eye, ovary and especially abundantly expressed in skeletal muscle. The gene knockdown of zCaMKId using morpholino-based antisense oligonucleotides induced significant morphological abnormalities in zebrafish embryos. Severe phenotype of embryos exhibited short trunk, kinked tail and small heads. These phenotypes could be rescued by coinjection with the recombinant zCaMKId, but not with the kinase-dead mutant. These results clearly indicate that the kinase activity of zCaMKId plays a crucial role in the early stages in the embryogenesis of zebrafish. Ó 2011 Elsevier Inc. All rights reserved.

Introduction Protein kinases are known to play pivotal roles in various signaling pathways and to participate in diverse cellular processes including proliferation, development and differentiation [1]. To explore a wide variety of protein kinases expressed in various cells and tissues, we developed unique expression screening of protein kinases with the aid of kinase-specific monoclonal antibodies [2,3]. By introducing this technique, we isolated various known and novel protein kinases from mouse brain [2], Xenopus laevis embryo [4], root-nodule of Lotus japonicus [5] and basidiomycete mushroom Coprinopsis cinerea [6,7]. To investigate protein kinases involved in the developmental processes in embryogenesis, we performed expression screening of protein kinases using cDNA library from zebrafish embryos at 72 hpf. With this screening, we obtained Ca2+/calmodulin(CaM)1-dependent protein kinase I (CaM-

q The nucleotide sequence reported in this paper has been submitted to the GenBankTM/EBI Data Bank with an Accession Nos. BAG70944, BAG70945. ⇑ Corresponding author. Address: Department of Life Sciences, Faculty of Agriculture, Kagawa University, Ikenobe 2393, Miki-cho, Kagawa 761-0795, Japan. Fax: +81 87 891 3114. E-mail address: [email protected] (N. Sueyoshi). 1 Abbreviations used: CaM, calmodulin; CaMKI, Ca2+/calmodulin-dependent protein kinase I; CaMKK, Ca2+/calmodulin-dependent protein kinase kinase; CREB, cyclic AMP-responsive element-binding protein; DIG, digoxigenin; hpf, hours post-fertilization; MBP, myelin basic protein; MO, morpholino oligonucleotide.

0003-9861/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2011.11.008

KI), CaMKIV [8] and other protein kinases, which may function in the developing embryos. A variety of cellular functions are regulated by Ca2+ signalings through CaMK cascades, which consist of multifunctional CaMKs; CaMKI, CaMKII and CaMKIV. CaMKII is regulated by Ca2+/CaMdependent autophosphorylation [9–12], while CaMKI and CaMKIV are activated by phosphorylation of Thr residues in the activation loops by an upstream kinase, CaMK kinase (CaMKK) [13–16]. Although these CaMKs are known to be abundantly expressed in the central nervous systems and phosphorylate downstream targets, much smaller number of evidences has been documented concerning CaMKI as compared to the case of CaMKII. CaMKI has been shown to comprise a family of four isoforms (a, b, c and d) encoded by separate genes. Spatiotemporal expression of four isoforms of CaMKI in mouse brain was extensively investigated by the previous studies [17]. Each isoform exhibited different temporal expression during developmental stages and different regional expression in the brain, suggesting that these isoforms may share different functional roles in the central nervous systems. Although expression of CaMKI isoforms in the other tissues than brain have been reported [18–21], their functions still remain to be elucidated. In the present study, we found that two splice variants of CaMKId with different C-terminal sequences were present in the zebrafish embryos at 72 hpf. Therefore, we expressed these two isoforms, CaMKId-S and CaMKId-L, in Escherichia coli and

72

Y. Senga et al. / Archives of Biochemistry and Biophysics 517 (2012) 71–82

investigated their biochemical properties. These isoforms began to appear at the different stages in the early embryogenesis and showed different tissue distribution in the adult fish. The gene knockdown of zCaMKId by morpholino-based antisense oligonucleotides induced significant morphological abnormalities in zebrafish embryos having small heads and short bodies, and this phenotype could be rescued by coinjection with recombinant zCaMKId. These results implicate that CaMKId is indispensable for some of the processes in zebrafish embryogenesis. Materials and methods Materials ATP, Cy3-labeled anti-mouse IgG, bovine serum albumin, histone type IIA from calf thymus, a-casein from bovine milk, myelin basic protein (MBP) from bovine brain, antibody against cyclic AMP-responsive element-binding protein (CREB) and antibody against phospho-CREB were purchased from Sigma Chemicals. [c-32P]ATP (111 TBq/mmol) and Alexa Fluor 488 anti-rabbit IgG were purchased from PerkinElmer and Invitrogen, respectively. Goat anti-mouse IgG and goat anti-rabbit IgG, conjugated with horseradish peroxidase, were obtained from ICN Pharmaceuticals. Myosin light chain was prepared as described previously [22]. Recombinant rat CaM [23], mouse CaMKKa [4] and rat CaMKIa [24] were expressed in E. coli and purified as described previously. Antibodies against CaMKI and phospho-CaMKI at Thr-177 were produced by immunizing mice with purified CaMKI (for anti-CaMKI) or antigenic phosphopeptide (for anti-phospho-CaMKI) [25]. Fish maintenance Zebrafish, Danio rerio, were maintained at 26 °C and embryos were collected from natural crosses of wild-type fish. Embryos were maintained in E3 medium (5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2 and 0.33 mM MgSO4) at 28 °C. Embryos were staged according to hpf at 28 °C and morphological criteria [26]. Preparation of crude extracts from zebrafish embryos and adult tissues Crude extracts from the embryos at each sampling point were prepared as follows: dechorionated embryos (N = 40) were suspended in 200 ll of ice cold homogenizing buffer containing 5 mM Tris–HCl buffer (pH 7.5), 0.5 mM EGTA, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride (PMSF), 1 mM dithiothreitol, 10 lg/ml chymostatin, 10 lg/ml pepstatin, 10 lg/ml leupeptin and 10 lg/ml antipain, and homogenized by sonication (Tomy, Handy Sonic UR-20P). Crude extracts from adult tissues were prepared as above except that five volumes of homogenizing buffer were used and homogenized by Polytron (Kinematica). The homogenates were centrifuged at 20,000g at 4 °C for 20 min, and the supernatants thus obtained were used as crude extracts. cDNA cloning of zCaMKId A cDNA library was constructed with mRNA isolated from the zebrafish embryo at 72 hpf using a kZAPII cDNA Synthesis Kit (Stratagene). Screening of protein kinases was carried out using Multi-PK antibodies directed against highly conserved regions in Ser/Thr protein kinases [2]. Cloning, sequencing and data analyses of the cDNA clones corresponding to various protein kinases were carried out as described previously [2,4,5]. A SMART RACE cDNA Amplification Kit (Clontech) was used to obtain a full-length coding sequence for CaMKId using primers based on a cDNA clone. The 50 -RACE first-strand cDNA was synthesized from mRNA iso-

lated from the zebrafish embryo (72 hpf) with SMARTScribe reverse transcriptase using a SMARTer™ RACE cDNA Amplification Kit (Clontech). The 50 -end of the cDNA was amplified by PCR with a gene-specific antisense primer (50 -TGT AAG CAA TAA CTC CAA TGG ACC AGC AG-30 ) and a universal primer mix with the 50 -RACE first-strand cDNA as a template using an Advantage 2 PCR Kit (Clontech). The 50 -RACE PCR product was cloned into pGEM-T Easy (Promega), and the DNA sequences determined. A sense primer (50 GAA TGC GGG CAA CGA TTA AAA CC-30 ) and an antisense primer (50 -GCA TCA ATC AGG TGA GCA GTA TCA CAG-30 ) were designed from the outside sequences of the open reading frame. Full-length cDNA was prepared by PCR using these primers and 50 -RACE ready cDNA library as a template with PrimeSTAR DNA polymerase (TaKaRa Bio). Two PCR products, shorter one and longer one, were cloned into pGEM-T Easy, and designated as pGEMzCaMKId-L and pGEMzCaMKId-S, respectively. Construction of plasmids To generate the plasmid pETzCaMKId-S, the following primers were used for PCR with pGEMzCaMKId-S as a template: sense primer (50 -GCT AGC ATG GCC AGG GGG AGC GAG GA-30 ) and antisense primer (50 -CTC GAG GAC AAA GAT AAC ACC CTC TTC AA30 ). In the case of pETzCaMKId-L, the following primers were used for PCR with pGEMzCaMKId-L as a template: sense (50 -GCT AGC ATG GCC AGG GGG AGC GAG GA-30 ) and antisense primer (50 CTC GAG TTT GGA GCC GGT GAT GAT GG-30 ). The NheI (underlined)–XhoI (double-underlined) fragment was inserted into the NheI–XhoI sites of pET-23a(+) (Novagen). To generate the plasmid pETmCaMKId the following primers were used for PCR with mouse brain 50 -RACE ready cDNA library as a template: sense primer (50 GCT AGC ATG GCC CGG GAG AAC GGC GA-30 ) and antisense primer (50 -CTC GAG CTT GCT TCC AGT GTG CCC TGT TG-30 ). For mammalian cells, the following plasmids were prepared. pczCaMKId-S was generated by PCR using specific primers (50 AAG CTT GTT ATG GCC AGG GGG AGC GAG GA-30 and 50 -CTC GAG CGG ACA AAG ATA ACA CCC TCT TCA AGG GAG TC-30 ) and pETzCaMKId-S as a template. The HindIII (underlined)–XhoI (double-underlined) fragment was inserted into HindIII–XhoI sites of pcDNA3.1(+)/myc-His B (Invitrogen). pczCaMKId-L was generated by PCR with specific primers (50 -AAG CTT GTT ATG GCC AGG GGG AGC GAG GA-30 and 50 -CTC GAG CGT TTG GAG CCG GTG ATG ATG GTG G-30 ) using pETzCaMKId-L as a template. pcmCaMKId was generated by PCR with primers (50 -AAG CTT GTT ATG GCC CGG GAG AAC GGC GA-30 and 50 -CTC GAG CGC TTG CTT CCA GTG TGC CCT GTT GTC AC-30 ) using pETmCaMKId (WT) as a template. Expression and purification of recombinant CaMKI pETCaMKIds were introduced into E. coli strain BL21 (DE3). The transformed bacteria were grown at 37 °C to an A600 of 1.0, and then isopropyl-b-D-thiogalactopyranoside was added to a final concentration of 0.1 mM. After 6 h at 37 °C, the bacteria were harvested by centrifugation and suspended in buffer A (20 mM Tris–HCl (pH 7.5), 150 mM NaCl, 0.05% Tween 40 and 1 mM PMSF). After sonication, cell debris was removed by centrifugation (20,000g) at 4 °C for 10 min, and supernatant thus obtained was loaded on a HiTrap Chelating HP column (GE Healthcare Bio-Sciences) pre-equilibrated with buffer A. The column was washed with buffer A, buffer A containing 20 mM imidazole, buffer A containing 50 mM imidazole, and then eluted with buffer A containing 200 mM imidazole. The purified fractions were pooled, dialyzed against 20 mM Tris–HCl (pH 7.5) containing 150 mM NaCl 0.05% Tween 40 and 1 mM 2-mercaptoethanol, and used for the characterization of the enzyme.

Y. Senga et al. / Archives of Biochemistry and Biophysics 517 (2012) 71–82

Production of antibodies An antigenic peptide (CKSQSVDVSSVHRNDSLEEGV) corresponding to the C-terminal sequence of zCaMKId-S (amino acid residues 346–366) and an antigenic peptide (CSPKQTPNSSTRTGKPERQT) corresponding to the C-terminal sequence of zCaMKId-L (amino acid residues 361–379) were synthesized using a Shimadzu PSSM-8 automated peptide synthesizer. These peptides were purified by reverse-phase HPLC on a C18 column and confirmed its purity by time-of-flight mass spectrometry. Purified peptides were coupled to keyhole limpet hemocyanin by a heterobifunctional reagent, N-(6-maleimidocaproyloxy)succinimide (Dojindo Laboratories), through their N-terminal cysteinyl residues as described previously [2]. Antibodies against zCaMKIdS and zCaMKId-L were produced by immunizing rabbits with the peptide conjugates essentially according to the procedure described previously [27]. SDS–polyacrylamide gel electrophoresis (SDS–PAGE) and Western blotting SDS–PAGE was performed essentially according to the method of Laemmli [28] on slab gels consisting of a 10% or 12% acrylamide separation gel and a 3% stacking gel. The resolved proteins were electrophoretically transferred to nitrocellulose membranes (Hybond-ECL, GE Healthcare Bio-Sciences) and immunoreactive protein bands were detected essentially according to the method described previously [4]. CaM-overlay assay Digoxigenin-labeled CaM (DIG-CaM) was prepared as follows: CaM (240 lg) was dissolved in 250 ll of phosphate-buffered saline (PBS) containing 1 mM CaCl2, and then 250 lg digoxigenin-3-Omethylcarbonyl-e-aminocaproic acid-N-hydroxysuccinimide ester (Roche Diagnostics) in 50 ll H2O was added. The mixture was incubated at 20 °C for 2 h, and unreacted reagent and CaCl2 were removed by dialysis against PBS with three changes. DIG-CaM solution was aliquoted and stored at 80 °C until use. CaM-overlay assay was carried out as follows: Protein samples were resolved on SDS–PAGE and electrophoretically transferred to nitrocellulose membranes. The membranes were incubated in washing buffer (50 mM Tris–HCl (pH 7.4), 150 mM NaCl and 0.05% Tween 20) containing 5% skim milk for 1 h at room temperature, and then washed three times with washing buffer containing either 1 mM CaCl2 or 1 mM EGTA. The membranes were incubated with DIG-CaM (0.7 lg/ml) in washing buffer containing 1 mM CaCl2 or 1 mM EGTA at room temperature for 1 h. After the membranes were rinsed three times with washing buffer in the presence of 1 mM CaCl2 or 1 mM EGTA, they were incubated with anti-DIG antibody conjugated with horseradish peroxidase (Novagen) at a dilution of 1:5000. The membranes were washed twice with washing buffer and then twice with washing buffer without Tween 20, each containing either 1 mM CaCl2 or 1 mM EGTA. CaM-binding proteins were visualized using chemiluminescent substrate, SuperSignal West Dura Extended Duration Substrate (PIERCE). Phosphorylation of recombinant CREB by CaMKId Phosphorylation of recombinant rat CREB was carried out at 30 °C for 30 min in a solution (30 ll) containing 40 mM HEPES– NaOH (pH 8.0), 5 mM Mg(CH3COO)2, 0.1 mM EGTA, 2 mM dithiothreitol (DTT), 1 mM CaCl2, 1 mM CaM and 100 lM ATP in the presence or absence of 5 lg/ml CaMKId and 0.5 lg/ml CaMKK. The reaction was initiated by the addition of CaMKId and terminated by the addition of 30 ll of 2 SDS–PAGE sample buffer.

73

The samples were then subjected to SDS–PAGE and analyzed by Western blotting using an anti-phospho-CREB antibody. Protein kinase assay Prior to the kinase assay, CaMKId was activated by phosphorylation with recombinant CaMKK. Phosphorylation of CaMKId was carried out at 30 °C for 30 min in a reaction mixture (10 ll) consisting of 40 mM HEPES–NaOH (pH 8.0), 5 mM Mg(CH3COO)2, 0.1 mM EGTA, 2 mM DTT, 1 mM CaCl2, 1 lM CaM, 10 lg/ml CaMKId, 1 lg/ ml CaMKK and 100 lM ATP. The reaction was initiated by the addition of CaMKId and terminated by 10-fold dilution with ice-cold 50 mM Tris–HCl (pH 7.5), 0.05% Tween 40, 2 mM DTT and 2 mM EDTA. For phosphorylation of protein substrates, the indicated proteins (100 lg/ml) was incubated at 30 °C for 30 min in a solution (10 ll) containing 40 mM HEPES–NaOH (pH 8.0), 5 mM Mg(CH3COO)2, 0.1 mM EGTA, 2 mM DTT, 1 mM CaCl2, 1 lM CaM, 100 lM [c-32P]ATP and 1 lg/ml activated CaMKId. The reaction was initiated by the addition of CaMKId, and terminated by the addition of 10 ll of 2 SDS–PAGE sample buffer. After SDS–PAGE, the phosphorylated proteins were detected by autoradiography. RT-PCR Total RNA was prepared from representative developmental stages of zebrafish (N = 40) using 1 ml of TRIzol reagent (Invitrogen) by homogenization (polytron, Kinematica). cDNAs were synthesized from 0.7 lg of the total RNA with SuperScript III FirstStand Synthesis System for RT-PCR (Invitrogen) and used as templates. A sense primer (50 -GCT AGC ATG GCC AGG GGG AGC GAG GA-30 ) and an antisense primer (50 -CTC GAG GAC AAA GAT AAC ACC CTC TTC AA-30 ) were used to amplify zCaMKId-S cDNA and zCaMKId-L cDNA was amplified using a sense primer (50 -GCT AGC ATG GCC AGG GGG AGC GAG GA-30 ) and an antisense primer (50 -CGT TTG TTT CGG CGA ACA GTC A-30 ) with rTaq polymerase (TaKaRa). The amplified fragments were subjected to 0.7% agarose gel electrophoresis in TAE buffer and stained with ethidium bromide. An actin gene (GenBank ID: AY222742) was used as an internal standard for the RT-PCR analyses. Cell culture and transfection COS7 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM, Wako) containing 10% heat inactivated fetal calf serum. Cells were grown at 37 °C in a humidified incubator with a 5% CO2/95% air atmosphere. Transfection of COS7 cells was performed using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions. COS7 cells were plated at 2  105 in a 35mm dish in 2 ml of DMEM containing 10% fetal calf serum. After 24 h of culture, cells were transfected by incubation for 24 h in 0.8 ml of DMEM containing 5% fetal calf serum, 4 ll of Lipofectamine 2000 and 2 lg of plasmid DNA. Immunocytochemistry of CaMKId Transfected cells were cultured on 0.1 mg/ml poly-L-lysine coated cover glass and treated with 10% formalin in PBS for 20 min. After being rinsed with PBS, formalin-fixed cells were permeabilized with 0.1% Triton X-100 in PBS for 5 min. After treatment in PBS containing 1% bovine serum albumin, the samples were incubated with anti-myc antibody (mouse IgG, Invitrogen) in PBS containing 1% bovine serum albumin at room temperature for 2 h followed by incubation with Cy3-labeled anti-mouse IgG at room temperature for 2 h, and observed by a confocal laserscanning microscope (TCS SP, Leica).

74

Y. Senga et al. / Archives of Biochemistry and Biophysics 517 (2012) 71–82

Whole mount immunostaining of CaMKId Embryos were incubated in E3 medium containing 0.1 mM phenylthiourea to block melanin production. After being rinsed with PBS containing 0.05% Tween20 (PBS-Tween), trichloroacetic acid-fixed embryos were permeabilized with 0.125% trypsin in PBS containing 0.8% Triton X-100 (PBS-Triton) for 5 min. Embryos were blocked with 10% goat serum in PBS-Triton for 1 h at room temperature. An anti-zCaMKId-S or an anti-zCaMKId-L antibodies (in PBS-Triton containing 1% goat serum) were added, and the embryos were incubated for 16 h at 4 °C. After washing three times with PBS-Triton, embryos were incubated with Alexa Fluor 488 anti-rabbit IgG (in PBS-Triton containing 1% goat serum) for 16 h at 4 °C. After washing six times for 20 min with PBS-Triton and three times for 5 min with PBS-Tween, the specimens were examined using a fluorescence stereoscopic microscope (SMZ1500, Nikon). Whole mount in situ hybridization Whole mount in situ hybridization analysis was carried out as described previously [29–31]. Embryos were incubated in E3 medium containing 0.1 mM phenylthiourea to block melanin production. The sense (as a negative control) and the antisense RNA probes, labeled with DIG-UTP, were synthesized using the DIG RNA Labeling mix (Roche Diagnostics) and zCaMKId cDNA (484– 1177) subcloned into pGEM-T Easy as a template. The hybridization was detected by anti-DIG antibody conjugated with alkaline phosphatase using NBT-BCIP as a substrate. Morpholino injections Morpholino oligonucleotides (MO), stable nucleic acid analog, were purchased from GeneTools and solubilized in sterilized water at a concentration of 5 lg/ll. The resulting stock solution was diluted to working concentrations in sterilized water before injection. Zebrafish embryos were injected at 1–4 cell stage with MO using a micromanipulator and microinjector (Narishige) according to the method of Nasevicius and Ekker [32]. Injected embryos were cultured in E3 medium at 26 °C. The sequences of MO used in the present study were; zCaMKId AS-MO, 50 -GCT GCC TCC GAA TAA CCT TCT TAA C-30 and zCaMKId 5mis-MO, 50 -GCT CCC TCC CAA TAA GCT ACA TAA C-30 (underlines show the sites of mismatch). Other methods Protein concentrations were determined by the method of Bensadoun and Weinstein using bovine serum albumin (BSA) as a standard [33]. Nucleotide sequences were determined by the dideoxynucleotide chain termination method with a BigDye Terminator Cycle Sequencing Ready Reaction Kit Ver.3.1 (Applied Biosystems) and a DNA Sequencer (model 3100, Applied Biosystems). Results cDNA cloning of zebrafish CaMKId In the previous study, we developed a unique expression cloning technique using monoclonal antibodies directed to highly conserved sequences (subdomain VIB) in protein kinases, and isolated cDNA clones for various Ser/Thr kinases [2,4–8]. When a cDNA expression library from zebrafish embryo at 72 hpf was used for this screening, 14 positive clones were obtained from 1.5  106 plaques. The positive clones thus obtained were sequenced and analyzed by BLAST homology search, and among them, the

nucleotide sequence of one clone, designated ZEZ082, was not found in the database. Since this clone appeared to encode a part of the d isoform of zCaMKI, we obtained 50 - and 30 -clones for zCaMKId and determined their entire nucleotide sequences. The PCR analysis of zCaMKId using zebrafish brain cDNA revealed two PCR products with different sizes. The sequence of the longer product was found to contain an additional sequence composed of 29bp DNA, which we designated a-domain sequence (Fig. 1A). The adomain belongs to the 30 -end of intron 10 in the shorter product and to the 50 -end of exon 11 in the longer product as illustrated in Fig. 1B. An open reading frame of the shorter transcript was found to contain 1176 nucleotides encoding a protein of 392 amino acid residues, designated as zCaMKId-L, with a predicted molecular weight of 44,126. On the other hand, the longer one had an insertion sequence of 29 bases between the 1077th and 1078th bases, and the position of the stop codon changed by frame shift (Fig. 1B). As a result, this gene encodes a protein of 368 amino acid residues, designated as CaMKId-S, with a predicted molecular weight of 41,572. The N-terminal sequence (1–359 amino acids) of zCaMKId-L and zCaMKId-S was completely identical, and contained all of the 12 highly conserved subdomains characteristic of protein kinases. Amino acid sequence of the catalytic domain of zCaMKId showed 88% identity with that of mouse CaMKId (Fig. 1C and D). The CaM-binding domain in the C-terminal region and the phosphorylatable Thr residue in the activation loop by CaMKK are highly conserved. Expression and characterization of zCaMKId To characterize zCaMKId isoforms, zCaMKId-L, zCaMKId-S and mouse CaMKId were produced using E. coli expression system and purified. The recombinant CaMKId isoforms were expressed in soluble forms in E. coli, and these kinases could readily be purified using a nickel affinity column. Since the amino acid sequence of zCaMKId showed the highest homology with that of mouse CaMKId, the purified CaMKId isoforms were electrophoresed on SDS–PAGE and compared with each other (Fig. 2A). These recombinant proteins could be immunostained with an antiCaMKI antibody (Fig. 2A, lower panel). CaM-overlay experiment was carried out to examine CaM-binding activity of these recombinant proteins. As shown in Fig. 2B, DIG-labeled CaM bound to all of the recombinant proteins when incubated in the presence of 1 mM CaCl2 but not in the presence of 1 mM EDTA, indicating that these isoforms are a member of the CaM-dependent protein kinase family. Mammalian CaMKI and CaMKIV are known to be phosphorylated and activated by an upstream kinase, CaMKK [13–16]. Therefore, in order to examine whether zCaMKId-L and zCaMKId-S can be phosphorylated by CaMKK, recombinant CaMKId isoforms were incubated with mouse CaMKK in the presence or absence of Ca2+/ CaM. As shown in Fig. 2C, the Thr residues in the activation loops were phosphorylated only in the presence of CaMKK. The protein kinase activity of CaMKId toward CREB was also examined using the reaction mixtures with and without CaMKK. CREB was phosphorylated by zCaMKId-S, zCaMKId-L and mouse CaMKId in the presence of Ca2+/CaM and CaMKK, but not in the absence of CaMKK (Fig. 2C). Phosphorylation activity of zCaMKId isoforms toward CREB was stimulated approximately 10-fold by treatment with CaMKK in the presence of Ca2+/CaM (Fig. 2D). In Fig. 3, substrate specificities of zCaMKId-S, zCaMKId-L, mouse CaMKId and rat CaMKIa were compared with each other. Prior to use, CaMKIs were activated by phosphorylation with CaMKK in the presence of cold ATP. Five different protein substrates; MBP, myosin light chain, casein, histone and CREB, were used as substrates. MBP and histone as well as CREB were phosphorylated significantly in the presence of Ca2+/CaM and CaMKK. Although casein

Y. Senga et al. / Archives of Biochemistry and Biophysics 517 (2012) 71–82

75

A

B

C

D

Fig. 1. Alignment of deduced amino acid sequences of zCaMKId-S and zCaMKId-L with that of mouse CaMKId and splicing patterns of zCaMKId-S and zCaMKId-L. (A) Partial exon–intron structure in the region of the zCaMKId gene containing exon 10, intron 10, a domain and exon 11. The length of the intron 10 and a domain of the zCaMKId gene are indicated. (B) Schematic representation of zCaMKId-S and zCaMKId-L gene structure. Alternative splicing events are indicated. Exons and alternative splicing exon are shown by open boxes and gray box, respectively. (C) Primary structure of CaMKId. The kinase domain, CaM-binding domain and the C-terminal unique sequences are shown by gray boxes, black boxes and hatched boxes, respectively. The ATP binding site and phosphorylation site by CaMKK are shown by asterisks and arrowheads, respectively. The solid underlines in the C-terminal region show the peptide sequence used for anti-zCaMKId-S and zCaMKId-L antibody production. (D) Alignment of zCaMKId-S (Accession No.: BAG70944), zCaMKId-L (Accession No.: BAG70945) and mouse CaMKId (Accession No.: BC141413). Amino acid sequence of zCaMKId-S (residues 1–368), zCaMKId-L (residues 1–392) and mCaMKId (residues 1–385) are aligned using CLUSTAL W. Twelve subdomains specific to protein kinases are shown by solid underlines. The ATP-binding site and the phosphorylation site by CaMKK are shown by an asterisk and an arrowhead, respectively. The putative CaM-binding domains are boxed.

76

Y. Senga et al. / Archives of Biochemistry and Biophysics 517 (2012) 71–82

(kDa)

1

2

B

3

1

2

3

47

Ca2+

47

anti-CaMKI

47

EGTA

zC aM KIδ -S

mC aM KIδ

CBB

C

CaMKI CaMKK

(kDa)

D

(kDa)

45

zC aM KIδ -L

A

47

anti-CaMKI

47

anti-P-CaMKI

47

anti-CREB

47

anti-P-CREB

1.4

P-CREB/CREB

1.2 1.0 0.8 0.6 0.4 0.2 0

mC aM KIδ

zC aM KI δ -L

zC aM KIδ -S

CaMKK

Fig. 2. Purification of recombinant CaMKId isoforms and determination of their activities. (A) Electrophoretic analysis of purified CaMKId. His6-tagged CaMKId isoforms expressed in E. coli were purified using a HiTrap Chelating HP column. Approximately 500 ng (upper panel) or 50 ng (lower panel) of purified zCaMKId-S (lane 1), zCaMKId-L (lane 2) and mouse CaMKId (lane 3) were subjected to 10% SDS–PAGE and stained with Coomassie brilliant blue (upper panel), or detected by Western blotting analysis with anti-CaMKI antibody (lower panel). (B) CaM-overlay assay of CaMKId isoforms. Approximately 10 ng of purified zCaMKId-S (lane 1), zCaMKId-L (lane 2) and mouse CaMKId (lane 3) were subjected to 10% SDS–PAGE, transferred to a nitrocellulose membrane, and detected by DIG-CaM in the presence of 1 mM CaCl2 (upper panel) or 1 mM EGTA (lower panel). (C) Phosphorylation of CREB by CaMKId. Phosphorylation of CaMKId (25 ng) and phosphorylation activities toward CREB (125 ng) were compared in the presence (+) and absence () of CaMKK (2.5 ng). After 30 min of phosphorylation at 30 °C, the reaction was stopped by the addition of the same volume of 2 SDS–PAGE sample buffer, subjected to SDS–PAGE, and the phosphorylations of CaMKId and CREB were detected by Western blotting with anti-phospho-CaMKI and anti-phospho-CREB antibodies, respectively. (D) Quantitation of CREB phosphorylation. The reaction was carried out as in (C) and phospho-CREB was quantitated by Scion Image software. Data are means ± SD from three independent determinations.

served as a moderate substrate for rat CaMKIa, it served as a poor substrate for CaMKId isoforms. Phosphate incorporation into protein substrates was quantitated by an imaging analyzer and presented as the relative ratios of phosphorylation by setting CREB phosphorylation as 1.0 (Fig. 3B). Both zCaMKId-L and zCaMKId-S showed somewhat lower kinase activities as compared to rat CaMKIa. Subcellular localization, temporal and spatial expression of zCaMKId Although CaMKIa is known to be localized in cytoplasm [34], CaMKIb is localized in both cytoplasm and nucleus of neuronal cells [35] and CaMKIc is mainly bound to membrane compartments [20]. To examine subcellular localization of zCaMKId, myctagged zCaMKId-S and zCaMKId-L were transiently expressed in COS7 cells and detected by indirect immunofluorescence. As revealed by fluorescence microscopic analysis, both zCaMKId-S and

zCaMKId-L appeared to be localized in the cytosol, as in case of mammalian CaMKId (Fig. 4). RT-PCR analyses of various adult tissues revealed that zCaMKId-S mRNA is specifically expressed in the brain (Fig. 5A, upper panel). On the other hand, the transcript of zCaMKId-L was widely detected in various tissues (Fig. 5A, middle panel). In order to elucidate the role of zCaMKId in the early development of zebrafish, we examined the temporal and spatial expression patterns of zCaMKId genes by RT-PCR and whole mount in situ hybridization. RT-PCR revealed that mRNA of zCaMKId-S clearly appeared after 72 hpf (Fig. 5B, upper panel). On the other hand, the transcript of zCaMKId-L, but not zCaMKId-S, was detectable in just-fertilized eggs (0 hpf), and this signal disappeared at 12 hpf, a pattern of which is consistent with the typical expression pattern of maternally transcribed RNA (Fig. 5B, middle panel). A secondary rise at 24 hpf with weak level indicates the onset of zCaMKId-L gene by embryonic transcription. After that, mRNA of zCaMKId-L was significantly expressed at 48 hpf and thereafter

77

Y. Senga et al. / Archives of Biochemistry and Biophysics 517 (2012) 71–82

A

MBP 1

(kDa)

2

3

MLC 4

1

2

3

4

histone

casein

CREB

1 2 3 4

1 2 3 4

1 2 3 4

45 CREB

36 29

casein

20

MBP MLC histone

B

MBP

PSL /CREB

20

2.0

15

1.5 10 1.0 5 0

0.5

1

2

3

0

4

histone

0.5

PSL /CREB

MLC

2.5

0.08

0.3

0.06

0.2

0.04

0.1

0.02

1

2

3

2

0

4

3

4

3

4

casein

0.10

0.4

0

1

1

2

Fig. 3. Substrate specificity of CaMKI. (A) Comparative analysis of substrate specificity of zCaMKId-S, zCaMKId-L, mouse CaMKId and rat CaMKIa. Prior to use, recombinant CaMKIs (100 ng) were phosphorylated by CaMKK (10 ng) in the presence of 0.5 mM CaCl2 and 1 lM CaM in the standard phosphorylation mixture (10 ll) containing 100 lM ATP at 30 °C for 30 min. MBP, myosin light chain (MLC), histone, casein and CREB were then incubated with phosphorylated zCaMKId-S (lane 1), zCaMKId-L (lane 2), mouse CaMKId (lane 3) and rat CaMKIa (lane 4) (10 ng) in the standard phosphorylation mixture (10 ll) containing 100 lM [c-32P]ATP. After incubation at 30 °C for 30 min, reactions were stopped by the addition of 10 ll of 2 SDS–PAGE sample buffer, and phosphorylated proteins were resolved on 12% SDS–PAGE and detected by autoradiography. (B) Quantitative result of (A). The phosphorylation of substrate protein was quantitated by a bioimaging analyzer BAS1800 (Fuji Film). The phosphorylation activities toward CREB were arbitrarily set to 1.0 and the activities against the other proteins were expressed as relative values.

A

B zCaMKIδ-S

C zCaMKIδ-L

mCaMKIδ

Fig. 4. Subcellular localization of zCaMKId-S, zCaMKId-L and mouse CaMKId. COS7 cells were transfected with myc-tagged zCaMKId-S (A), zCaMKId-L (B) and mouse CaMKId (C). Transiently expressed myc-tagged proteins were stained by means of indirect immunofluorescence with anti-myc antibody and visualized by a confocal laser-scanning microscope (TCS SP, Leica).

(Fig. 5B, middle panel). In in situ hybridization experiments, positive signals were observed in the wide range of central nervous system, especially in forebrain (Fig. 5D and E, arrowheads) and hindbrain (Fig. 5E, arrows). The signals were not observed when the sense probe was used in place of the antisense probe (Fig. 5C).

We also examined the expression of zCaMKId during the developmental stages of embryogenesis by Western blotting analysis using the isoform-specific antibodies to zCaMKId-L and zCaMKId-S. As shown in Fig. 6A, zCaMKId-L was clearly detected in the embryos at 36 hpf but not at 24 hpf (Fig. 6A, lower panel). In good agreement

Muscle

Fin

Ovary

Eye

Brain

A

Testis

Y. Senga et al. / Archives of Biochemistry and Biophysics 517 (2012) 71–82

Heart

78

Gene knockdown of zCaMKId

zCaMKIδ-S zCaMKIδ-L Actin

B

0

12

24

48

72

96 120 (hpf)

zCaMKIδ-S zCaMKIδ-L Actin sense RNA probe

C

F). zCaMKId-L protein was highly expressed in muscle at 120 hpf (Fig. 6E), while zCaMKId-S protein was hardly detectable (Fig. 6F).

antisense RNA probe

D

E

48 hpf

Fig. 5. Expression of zCaMKId gene transcript in zebrafish embryos. (A) RT-PCR analysis of zCaMKId mRNAs expressed in adult tissues (A) and embryos at indicated stages (B). The RT-PCR was carried out using total RNA (0.7 lg). Expression of zCaMKId-S (upper panels), zCaMKId-L (middle panels) and actin gene as an internal standard (lower panels) are shown. The PCR products were stained with ethidium bromide. (C–E) Zebrafish embryos at 48 hpf were analyzed by whole mount in situ hybridization with DIG-labeled RNA probes and detected by anti-DIG antibody conjugated to alkaline phosphatase. Sense RNA probe (C) and antisense RNA probe (D and E). (E) Dorsal view of (D). Arrowheads and arrows indicate forebrain and hindbrain, respectively.

with the result of Fig. 5B, zCaMKId-L was present in just-fertilized eggs and undetectable at 12 hpf, suggesting a maternal origin (Fig. 6A, lower panel, lane 3). On the other hand, zCaMKId-S was detected at 60 hpf and thereafter, but the expression level appeared to be much lower than zCaMKId-L (Fig. 6A, upper panel). The tissue distribution was also examined by Western blotting analysis using various tissue extracts of adult zebrafish. When the tissue extracts from adult zebrafish was analyzed by immunoblotting with anti-zCaMKId-S and anti-zCaMKId-L antibodies, an immunoreactive band corresponding to the size of zCaMKId-S was specifically detected in brain but not in the other tissues at all (Fig. 6B, upper panel). In contrast, zCaMKId-L was distributed in various tissues such as brain, eye, ovary and especially abundant in skeletal muscle (Fig. 6B, lower panel). When anti-zCaMKId-L antibody was preincubated with the antigenic peptide used for immunization, the reactive bands detected in brain and skeletal muscle were totally diminished (Fig. 6C and D), thereby eliminating the possibility that the reactive bands were caused by non-specific binding of the antibody. It should be noted that both isoforms were detected from supernatant fraction but not in the pellet fraction, indicating that CaMKId is not a membrane-bound CaM kinase (not shown). Whole-mount immunostaining was also carried out to investigate the expression patterns of zCaMKId-S and zCaMKId-L proteins in skeletal muscle (Fig. 6E and

During embryogenesis, the expression of zCaMKId increased gradually after 24 hpf (Figs. 5A and 6A). This led us to analyze the role of zCaMKId isoforms in zebrafish embryogenesis. In an attempt to understand more precisely how zCaMKId is related to embryonic development, we performed functional gene knockdown experiments in zebrafish using antisense morpholino-modified oligonucleotide (AS-MO) that was targeted to the 50 -noncoding sequence common to both zCaMKId-S and zCaMKId-L (Fig. 7A). AS-MO has been reported to act as effective and specific translational inhibitors in the zebrafish embryo [32]. We also used control MO with five mismatch nucleotides (5mis-MO), which would not inhibit translation of CaMKId. AS-MO and 5misMO were injected to 1–4 cell stage of embryos and observed their developments every 12 h (Fig. 7B). Injection of zCaMKId AS-MO to 1–4 cell stage embryos (approximately 2.4 ng/embryo) resulted in knockdown of zCaMKId protein as demonstrated by Western blotting using affinity-purified anti-zCaMKId-S and anti-zCaMKId-L antibodies (Fig. 7C, lanes 3 and 4). In contrast, injection of 5misMO had no effect on the expression of zCaMKId. The knockdown of the zCaMKId gene with AS-MO increased in the number of embryos with severe morphological abnormalities with small head and eye as compared to the control embryos at 24 hpf (Fig. 7B). Severe phenotypes also exhibited short trunks and kinked tails. These morphological abnormalities were further confirmed by whole mount in situ hybridization using DIG-labeled RNA probes for well-characterized gene markers; rx1 for retina, eng2a for midbrain-hindbrain boundary and emx1 for forebrain. Injection of AS-MO but not 5mis-MO resulted in a reduction in the sizes of eye (rx1) and brain (emx1 and eng2a), probably due to retardation in the development of brain and eye (Fig. 7D). Among 100 embryos tested, less than 5% of embryos were abnormal when 5mis-MO was injected into 1–4 cell stage. In contrast, more than 70% of embryos were abnormal when AS-MO was injected at the concentrations of 3 lg/ll (approximately 2.4 ng/ embryo), while about 30% of embryos were abnormal when 1 lg/ ll of AS-MO was injected (Fig. 8). The occurrence of abnormal embryos was closely correlated with the amount of AS-MO that inhibits the production of zCaMKId (Fig. 8A). To clarify whether kinase activity of zCaMKId is a prerequisite for normal embryogenesis, we prepared kinase-dead mutants of zCaMKId (zCaMKId-L(KD) and zCaMKId-S(KD)), and examined whether coinjection of the mutant kinases with AS-MO could rescue the abnormal phenotype. As shown in Fig. 8B, coinjection of recombinant zCaMKId-L(WT) or zCaMKId-S(WT) with AS-MO reduced the gross abnormality rate from 90% to 10% and to 37%, respectively, as compared to control coinjection with BSA. Coinjection of the kinase-dead mutants, however, failed to rescue the phenotype (Fig. 8B). These results clearly demonstrated that the catalytic activity of zCaMKId is indispensable for the normal embryogenesis of zebrafish. Discussion We isolated cDNA clones for Ser/Thr protein kinases by expression screening using cDNA library from zebrafish embryos at 72 hpf. Among them, we isolated cDNA for CaMKIV, CaMKIId, and new d isoforms of CaMKI, but not other isoforms of CaM kinase. There are two splice variants of CaMKId, zCaMKId-S and zCaMKId-L, having different C-terminal sequences. zCaMKId-S (368 amino acids) is shorter than zCaMKId-L (392 amino acids) by 24 amino

Y. Senga et al. / Archives of Biochemistry and Biophysics 517 (2012) 71–82

A

zCaMKIδ

S

(kDa)

L

zCaMKIδ

S

L

0 12 24 36 48 60 72 (hpf)

47

anti-zCaMKIδ-S

47

L

7

8

9

Muscle

S

L

6

Fin

S

5

Ovary

(kDa)

4

Testis

zCaMKIδ zCaMKIδ

3

Heart

2

Eye

1

Brain

anti-zCaMKIδ-L 1 2 anti-CaMKI

B 47

anti-zCaMKIδ-S

47

anti-zCaMKIδ-L 1 2 anti-CaMKI

C

79

(kDa)

1

1

2

2

3

3

4

5

D

47

6

7

8

9

1

2

3

(kDa) 47

peptide(-)

E

peptide(+)

F

zCaMKIδ-L, 120 hpf

zCaMKIδ-S, 120 hpf

Fig. 6. Expression of zCaMKId in zebrafish. (A) Expression of zCaMKId during zebrafish embryogenesis. Crude extracts (150 lg for zCaMKId-S, 50 lg for zCaMKId-L) from zebrafish embryos at the indicated stages were subjected to 10% SDS–PAGE and transferred to a nitrocellulose membrane for detection with affinity-purified anti-zCaMKId-S (upper panel) and anti-zCaMKId-L (lower panel) antibodies. His6-tagged zCaMKId-S (2.5 ng) and zCaMKId-L (2.5 ng) are shown in lanes 1 and 2. To show the migration positions of both isoforms, the same blot was also stained with anti-CaMKI antibody (left panels) (B) Tissue distribution of zCaMKId. Crude extracts (150 lg for zCaMKId-S, 50 lg for zCaMKId-L) of the brain (lane 3), eye (lane 4), heart (lane 5), testis (lane 6), ovary (lane 7), fin (lane 8) and skeletal muscle (lane 9) prepared from adult zebrafish were subjected to 10% SDS–PAGE and detected by Western blotting analysis with affinity-purified anti-zCaMKId-S (upper panel) and anti-zCaMKId-L (lower panel) antibodies. Purified His6-tagged zCaMKId-S (2.5 ng) and zCaMKId-L (2.5 ng) are shown in lanes 1 and 2. (C and D) Purified zCaMKId-L (lane 1, 5 ng), crude extracts from zebrafish brain (lane 2, 50 lg) and skeletal muscle (lane 3, 1 lg) were electrophoresed, and the reactive proteins were immunologically detected by affinity-purified anti-zCaMKId-L antibody (C). As a control, the blot membranes were incubated with anti-zCaMKId-L antibody that had been preincubated with 0.5% skim milk containing 2.5 mg/ml immunizing peptide (D). (E and F) Whole mount immunostaining of CaMKId proteins in skeletal muscle. The endogenous zCaMKIds in zebrafish skeletal muscle were stained with affinitypurified anti-zCaMKId-L antibody (E) or anti-zCaMKId-S antibody (F), respectively. Scale bar = 0.1 mm.

acids due to the frame shift of this gene as illustrated in Fig. 1. Entire amino acid sequence of zCaMKId-L shows 79% identity with that of mouse CaMKId. zCaMKId isoforms were expressed in E. coli and recombinant enzymes were compared with mammalian CaMKI isoforms. All the CaMKI isoforms exhibited essentially the same catalytic properties; they were activated by phosphorylation with CaMKK in the presence of Ca2+/CaM and activated CaMKIs efficiently phosphorylated CREB as a substrate. Although all the CaMKI isoforms preferentially phosphorylated CREB, minor differences in substrate specificities were observed between CaMKIa and CaMKId isoforms (Fig. 3). However, we could not find any differences in the enzymatic properties between zCaMKId-S and zCaMKId-L, though the C-terminal sequences of zCaMKId isoforms are different. These isoforms expressed in different tissues might be regulated by different regulatory proteins that could interact with their unique C-terminal region. Investigation of binding proteins to the C-terminal region of these proteins is now in progress in this laboratory.

CaMKI comprises a family of four isoforms (a, b, c and d) encoded by separate genes. Spatial and temporal expression of these isoforms in mouse brain was extensively studied and documented previously [17]. CaMKIa and CaMKIb are rather widely distributed throughout the brain, while expressions of CaMKIc and CaMKId are restricted in the specific regions in the brain. By RT-PCR analysis, mRNA level of mouse CaMKId was found to increase gradually from P5 to P21. Furthermore, the in situ hybridization histochemistry demonstrated that CaMKId was distributed in forebrain and most abundantly in hippocampal pyramidal neurons. Up to date, physiological roles of CaMKI have been mainly studied in conjunction with its neuronal functions [36,37]. Although CaMKI is known to be widely distributed in various tissues in addition to brain [18,21,38], physiological function of CaMKI in peripheral tissues is not known. In the present study, we found that there are two zCaMKId isoforms, zCaMKId-S and zCaMKId-L, which showed different tissue distribution. zCaMKId-S was specifically expressed in brain, while zCaMKId-L was abundantly expressed in skeletal

80

Y. Senga et al. / Archives of Biochemistry and Biophysics 517 (2012) 71–82

A

B

5mis-MO

AS-MO Mild

Severe

D

5mis-MO

12 hpf

24 hpf

36 hpf

48 hpf

60 hpf

47

rx1 anti-zCaMKIδ-S

47

anti-zCaMKIδ-L 2

3

4

emx1

1

AS-MO

eng2a

AS-MO (Abn)

AS-MO (total)

5mis-MO

(kDa)

Uninjected

C

Fig. 7. Gene knockdown of zCaMKId during zebrafish embryogenesis. (A) Sequences of zCaMKId mRNA around the 50 initiation Met (double underlined), AS-MO and 5mis-MO used in this study. Underlines in the 5mis-MO indicate mismatch nucleotides compared to AS-MO. (B) Phenotypes of zebrafish embryos injected with AS-MO or 5mis-MO. ASMO or 5mis-MO was injected into 1–4 cell stage embryos at a concentration of 3 lg/ll (2.4 ng/embryo). Representative embryos exhibiting normal, mild and severe phenotypes at the indicated hpf are shown. (C) Detection of zCaMKId in the MO-injected embryos. Lysates were prepared at 60 hpf from embryos without injection (uninjected), injected with 5mis-MO (3 lg/ll), AS-MO (3 lg/ll, total) or AS-MO injected embryos exhibiting abnormal morphology (3 lg/ll, Abn). Protein extracts (150 lg for zCaMKId-S, 50 lg for zCaMKId-L) were subjected to 10% SDS–PAGE and detected by Western blotting analysis with anti-zCaMKId-S (upper panel) and anti-zCaMKId-L (lower panel) antibodies. (D) Whole mount in situ hybridization analysis of various marker genes in 5mis-MO (left panels) and AS-MO (right panels) embryos at 24 hpf: rx1, expression marker gene for the retina (upper panels); eng2a, expression marker gene for the mid-hindbrain boundary (middle panels); emx1, expression marker gene for the forebrain (lower panels). Scale bar = 0.2 mm.

muscle, ovary and eye as well as in brain. Expression of the d isoform in skeletal muscle was confirmed not only by Western blotting using tissue extracts but also RT-PCR and whole mount in situ immunostaining experiments. When the expression of zCaMKId-L was compared on the basis of mRNA level and protein level, some discrepancies were observed; mRNA of zCaMKId-L was detected in all the tissues including heart, testis and fin by RT-PCR (Fig. 5A), but zCaMKId-L protein was undetectable in heart, testis and fin when examined by Western blotting (Fig. 6B). Furthermore, although the expression level of zCaMKId-L protein in

muscle was much higher than that in brain, mRNA levels of zCaMKId-L in these tissues are comparable. These results suggest that the expression of zCaMKId-L might be regulated at the post-transcriptional and/or the post-translational level in various tissues. CaMKI is known to be a multifunctional protein kinase and can phosphorylate a number of substrates including synapsin I and II [39], CREB [40], myosin II [41] and Numb family proteins [42]. Suizu et al. reported that myosin regulatory light chain was phosphorylated by CaMKI both in vitro and in vivo [41]. Therefore, CaMKI may play some roles in mitosis through dynamic reorganization

Y. Senga et al. / Archives of Biochemistry and Biophysics 517 (2012) 71–82

Phenotype (%)

A

100 80 Normal 60 Mild 40 Severe

20 0 AS-MO 1 μg/μl)

)

100

* *

80 60 40 20

+ KIδ-S(KD)

+ KIδ-S (WT)

+ KIδ-L(KD)

+ KIδ-L(WT)

+ BSA

AS-MO + KIδ-S(KD)

+ KIδ-S (WT)

+ KIδ-L(KD)

+ KIδ-L (WT)

5mis-MO + BSA

0

Uninjected

Abnormal (%)

)

B

AS-MO 3 μg/μl)

)

Uninjected 5mis-MO 3 μg/μl)

Fig. 8. Phenotype of zebrafish embryos by knockdown of zCaMKId and rescue by coinjection with recombinant CaMKId. (A) Dose-dependent knockdown of zCaMKId by AS-MO. AS-MO (1 lg/ll or 3 lg/ll) or 5mis-MO (3 lg/ll) was injected into 1–4 cell stage embryos. After 60 hpf, embryos were classified into normal, mild and severe phenotypes. Percentage ratios of each phenotype were calculated from three independent experiments. (B) Appearance of abnormal embryos after coinjection of MOs with recombinant CaMKId isoforms. 5mis-MO (3 lg/ll) or AS-MO (3 lg/ll) were coinjected with 50 ng/embryo of BSA, zCaMKId-L(WT), zCaMKId-L(KD), zCaMKId-S(WT) and zCaMKId-S(KD) as indicated. Number of abnormal embryos were counted after 60 h. Data are shown by means ± SD from three independent experiments. ⁄P < 0.01.

of actin filaments in dividing cells. To further elucidate physiological function of zCaMKId in muscle, it is necessary to identify the endogenous target of zCaMKId-L in the developing muscle in zebrafish. In this study, we found that both zCaMKId-S and zCaMKId-L were localized exclusively in cytoplasm (Fig. 4). CaMKIb was reported to be localized in both cytoplasm and nucleus [35] and CaMKIc was bound to the Golgi and plasma membranes through its C-terminal prenylation motif [20]. In contrast, CaMKIa was predominantly localized in cytoplasm, since it had nuclear export signal. Nuclear export signal sequence of CaMKIa was identified as 312VVRHMRKLQL321, where underlined amino acids are critical residues for its cytoplasmic localization [34]. In case of zCaMKId, corresponding sequence around this region was found to be 317VIRHMRRLQL326, critical amino acids of which are highly conserved, and this sequence may function as nuclear export signal in zCaMKId. Although CaMKI isoforms showed essentially the same catalytic properties including substrate specificities, they could target different substrate proteins by exhibiting different subcellular localization. By the gene knockdown experiments with AS-MO, both zCaMKId-S and zCaMKId-L were eliminated at the same time, because these isoforms are encoded by the same gene. Since we could not knockdown these isoforms separately, we observed the effects of gene knockdown of both isoforms. The gene knockdown of zCaMKId by AS-MO caused severe phenotypes of embryos with small head and eye, short trunk and kinked tail. zCaMKId-S was ex-

81

pressed exclusively in brain, while zCaMKId-L was widely distributed in brain, eye, ovary and skeletal muscle. Underdevelopment of brain might be caused by knockdown of both isoforms, and other severe phenotypes in body shape may be attributed to the knockdown of CaMKId-L. In the gene knockdown experiments, coinjection of either zCaMKId-L or zCaMKId-S with AS-MO rescued the embryos from severe phenotypes. However, zCaMKId-L showed more significant effect on the rescue experiments than zCaMKId-S did (Fig. 8B). In consideration of the fact that zCaMKId-L was expressed from much earlier stage than CaMKId-S and more abundantly and widely in various tissues than CaMKId-S, it is rational to speculate that zCaMKId-L has more critical roles during the embryogenesis of zebrafish. In this study, we found that two splice variants, zCaMKId-S and zCaMKId-L, were expressed during embryogenesis of zebrafish. Gene knockdown experiments and rescue experiments suggested that these CaMKId isoforms play crucial roles in the early developmental stages of embryogenesis. Up to date, there have been very few documents regarding functional roles of CaMKI in the other tissues than the central nervous system. Therefore, identification of substrate proteins and regulatory proteins of zCaMKId-L in the other tissues in zebrafish is the next issue to be resolved. Acknowledgments We thank Dr. Yasushi Shigeri (AIST) for peptide synthesis. This work was supported in part by Grants-in-aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan and by the fund for Kagawa University Young Scientists 2010-2011. References [1] G. Manning, G.D. Plowman, T. Hunter, S. Sudarsanan, Trends Biochem. Sci. 27 (2002) 514–520. [2] I. Kameshita, T. Tsuge, T. Kinashi, S. Kinoshita, N. Sueyoshi, A. Ishida, S. Taketani, Y. Shigeri, Y. Tatsu, N. Yumoto, K. Okazaki, Anal. Biochem. 322 (2003) 215–224. [3] S. Sugiyama, N. Sueyoshi, Y. Shigeri, Y. Tatsu, N. Yumoto, A. Ishida, T. Taniguchi, I. Kameshita, Anal. Biochem. 347 (2005) 112–120. [4] S. Kinoshita, N. Sueyoshi, H. Shoju, I. Suetake, M. Nakamura, S. Tajima, I. Kameshita, J. Biochem. 135 (2004) 619–630. [5] I. Kameshita, T. Nishida, S. Nakamura, Y. Sugiyama, N. Sueyoshi, Y. Umehara, M. Nomura, S. Tajima, J. Biochem. 137 (2005) 33–39. [6] K. Kaneko, Y. Yamada, N. Sueyoshi, A. Watanabe, Y. Asada, I. Kameshita, Biochim. Biophys. Acta 1790 (2009) 71–79. [7] K. Kaneko, Y. Sugiyama, Y. Yamada, N. Sueyoshi, A. Watanabe, Y. Asada, A. Ishida, I. Kameshita, Biochim. Biophys. Acta 1810 (2011) 620–629. [8] T. Nimura, Y. Sugiyama, N. Sueyoshi, Y. Shigeri, A. Ishida, I. Kameshita, J. Biochem. 147 (2010) 857–865. [9] S.S. Hook, A.R. Means, Annu. Rev. Pharmacol. Toxicol. 41 (2001) 471–505. [10] H. Fujisawa, J. Biochem. 129 (2001) 193–199. [11] T.R. Soderling, J.T. Stull, Chem. Rev. 101 (2001) 2341–2352. [12] A. Hudmon, H. Schulman, Annu. Rev. Biochem. 71 (2002) 473–510. [13] S. Okuno, T. Kitani, H. Fujisawa, J. Biochem. 116 (1994) 923–930. [14] H. Tokumitsu, H. Enslen, T.R. Soderling, J. Biol. Chem. 270 (1995) 19320– 19324. [15] A.M. Edelman, K.I. Mitchelhill, M.A. Selbert, K.A. Anderson, S.S. Hook, D. Stapleton, E.G. Goldstein, A.R. Means, B.E. Kemp, J. Biol. Chem. 271 (1996) 10806–10810. [16] M. Matsushita, A.C. Nairn, J. Biol. Chem. 273 (1998) 21473–21481. [17] A. Kamata, H. Sakagami, H. Tokumitsu, Y. Owada, K. Fukunaga, H. Kondo, Neurosci. Res. 57 (2007) 86–97. [18] Y. Naito, Y. Watanabe, H. Yokokura, R. Sugita, M. Nishio, H. Hidaka, J. Biol. Chem. 272 (1997) 32704–32708. [19] O-P. Loseth, L. de Lecea, M. Calbet, P.E. Danielson, V. Gautvik, P.I. Hovring, S.I. Walaas, K.M. Gautvik, Brain Res. 869 (2000) 137–145. [20] S. Takemoto-Kimura, H. Terai, M. Takemoto, S. Ohmae, S. Kikumura, E. Segi, Y. Arakawa, T. Furuyashiki, S. Narumiya, H. Bito, J. Biol. Chem. 278 (2003) 18597– 18605. [21] Y. Ishikawa, H. Tokumitsu, H. Inuzuka, M. Murata-Hori, H. Hosoya, R. Kobayashi, FEBS Lett. 550 (2003) 57–63. [22] O. Miyano, I. Kameshita, H. Fujisawa, J. Biol. Chem. 267 (1992) 1198–1203. [23] N. Hayashi, M. Matsubara, A. Takasaki, K. Titani, H. Taniguchi, Protein Expr. Purif. 12 (1998) 25–28.

82

Y. Senga et al. / Archives of Biochemistry and Biophysics 517 (2012) 71–82

[24] Y. Tada, T. Nimura, N. Sueyoshi, A. Ishida, Y. Shigeri, I. Kameshita, Arch. Biochem. Biophys. 452 (2006) 174–185. [25] N. Sueyoshi, T. Takao, T. Nimura, Y. Sugiyama, T. Numano, Y. Shigeri, T. Taniguchi, I. Kameshita, A. Ishida, Biochem. Biophys. Res. Commun. 363 (2007) 715–721. [26] C.B. Kimmel, W.W. Ballard, S.R. Kimmel, B. Ullmann, T.F. Schilling, Dev. Dyn. 203 (1995) 253–310. [27] I. Kameshita, S. Kinoshita, Y. Shigeri, Y. Tatsu, N. Yumoto, A. Ishida, J. Biochem, Biophys. Methods 60 (2004) 13–22. [28] U.K. Laemmli, Nature 227 (1970) 680–685. [29] T. Nimura, N. Sueyoshi, A. Ishida, Y. Yoshimura, M. Ito, H. Tokumitsu, Y. Shigeri, N. Nozaki, I. Kameshita, Arch. Biochem. Biophys. 457 (2007) 205–216. [30] N. Sueyoshi, T. Nimura, A. Ishida, T. Taniguchi, Y. Yoshimura, M. Ito, Y. Shigeri, I. Kameshita, Arch. Biochem. Biophys. 488 (2009) 48–59. [31] M. Westerfield, The Zebrafish Book. A Guide for the Laboratory Use of Zebrafish (Danio rerio), 4th ed., University of Oregon Press, Eugene (1996). [32] A. Nasevicius, S.C. Ekker, Nat. Genet. 26 (2000) 216–220.

[33] A. Bensadoun, D. Weinstein, Anal. Biochem. 70 (1976) 241–250. [34] D.R. Stedman, N.V. Uboha, T.T. Stedman, A.C. Nairn, M.R. Picciotto, FEBS Lett. 566 (2004) 275–280. [35] Rina-Susilowati, A.A. Juseuf, H. Sakagami, S. Kikkawa, H. Kondo, Y. Minami, T. Terashima, Brain Res. 911 (2001) 1–11. [36] G.A. Wayman, S. Kaech, W.F. Grant, M. Davare, S. Impey, H. Tokumitsu, N. Nozaki, G. Banker, T.R. Soderling, J. Neurosci. 24 (2004) 3786–3794. [37] J.M. Schmitt, E.S. Guire, T. Saneyoshi, T.R. Soderling, J. Neurosci. 25 (2005) 1281–1290. [38] M.R. Picciotto, M. Zoli, G. Bertuzzi, A.C. Nairn, Synapse 20 (1995) 75–84. [39] A.C. Nairn, P. Grreengard, J. Biol. Chem. 262 (1987) 7273–7281. [40] M. Sheng, M.A. Thompson, M.E. Greenberg, Science 252 (1991) 1427–1430. [41] F. Suizu, Y. Fukuta, K. Ueda, T. Iwasaki, H. Tokumitsu, H. Hosoya, Biochem. J. 367 (2002) 335–345. [42] H. Tokumitsu, N. Hatane, H. Inuzuka, Y. Sueyoshi, S. Yokokura, T. Ichimura, N. Nozaki, R. Kobayashi, J. Biol. Chem. 280 (2005) 35108–35118.