- Email: [email protected]
ethanolamine kinase: their sequence comparison to the respective rat homologs1
Biochimica et Biophysica Acta 1393 (1998) 179^185
Short sequence-paper
Molecular cloning of mouse choline kinase and choline/ethanolamine kinase: their sequence comparison to the respective rat homologs1 Chieko Aoyama a
a;b
, Kinichi Nakashima a , Kozo Ishidate
a;
*
Department of Molecular Cell Biology, Medical Research Institute, Tokyo Medical and Dental University, 2-3-10 Kanda-surugadai, Chiyodaku, Tokyo 101-0062, Japan b Department of Hygienic Chemistry, Kyoritsu College of Pharmacy, Minatoku, Tokyo 105, Japan Received 23 March 1998; accepted 11 May 1998
Abstract Complementary DNAs homologous to a rat 42-kDa choline/ethanolamine kinase [C. Aoyama et al., Biochim. Biophys. Acta 1390 (1998) 1^7] and to a 50-kDa choline kinase [T. Uchida and S. Yamashita, J. Biol. Chem. 267 (1992) 10156^10162] were isolated from a 17-day post coitum mouse embryo cDNA library and their sequences were compared with the two murine species, respectively. The nucleotide sequence homology (within the coding sequence) between mouse and rat 50-kDa choline kinases (96.0%) was considerably higher than that between their 42-kDa choline/ethanolamine kinases (92.4%). Northern blot and RT-PCR studies on several rat tissues demonstrated that both isozymes are expressed ubiquitously with the highest level in testis. z 1998 Elsevier Science B.V. All rights reserved. Keywords: Choline kinase; Ethanolamine kinase; cDNA cloning; Isozyme; Phosphocholine; Phosphoethanolamine; mRNA expression
Choline kinase (CK) (ATP:choline phosphotransferase, EC 2.7.1.32) and ethanolamine kinase (EK) (ATP:ethanolamine phosphotransferase, EC 2.7.1.82) catalyze the phosphorylation of choline and ethanolamine by ATP yielding phosphocholine/ phosphoethanolamine and ADP. This enzyme step Abbreviations: CK, choline kinase; EK, ethanolamine kinase; PC, phosphatidylcholine; PE, phosphatidylethanolamine; ORF, open reading frame; DIG, digoxigenin; UTR, untranslated region; RT-PCR, reverse transcription-polymerase chain reaction; G3PDH, glyceraldehyde-3-phosphate dehydrogenase * Corresponding author. Fax: +81 (3) 5280 8065; E-mail: [email protected] 1 The nucleotide sequence data reported in this paper have appeared in the DDBJ, EMBL and GenBank nucleotide sequence databases with the following accession numbers: AB011000 and AB011001 for the mouse 42-kDa choline/ethanolamine kinase and AB011002 for the mouse 50-kDa choline kinase.
commits choline/ethanolamine to the CDP-choline/ CDP-ethanolamine pathway for the biosynthesis of phosphatidylcholine (PC)/phosphatidylethanolamine (PE) in all animal cells. In most (if not all) mammalian tissues, both reactions have been shown to be catalyzed by the same enzyme protein(s). It has also been demonstrated that the enzyme does not exist in one particular active form but exists as several isozymes, some of which have been puri¢ed from rat tissues to apparent homogeneity [1^3] and their cDNA cloned [4,5]. In recent years, phosphocholine (and probably also phosphoethanolamine) has been proposed to be a possible second messenger in the process leading to the induction of DNA synthesis by several growth factors (reviewed in [6]). In addition, certain chemical carcinogens [7,8] and/or ras/raf oncogene transfection [9^12] have been shown to elevate CK/EK ac-
0005-2760 / 98 / $19.00 ß 1998 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 5 - 2 7 6 0 ( 9 8 ) 0 0 0 6 2 - 9
BBALIP 50409 29-7-98
180
C. Aoyama et al. / Biochimica et Biophysica Acta 1393 (1998) 179^185
tivity in the cell. Although the exact mechanism leading to activation or induction of CK/EK as well as the exact site of phosphocholine participation in mitogenic signal pathway remain to be determined, isolation and characterization of each CK/EK gene would further prove the critical role of each isozyme in PC/PE biosynthesis and/or in mitogenic signal transduction in mammalian cells. In a previous paper [4], we reported cDNA cloning for a 42-kDa rat kidney CK/EK. Its sequence was shown to have only limited homology with that for a 50-kDa form of rat liver CK reported by Uchida and Yamashita [5], indicating that these two isoforms in the rat must be derived from separate genes. In addition, the existence of a third isoform of CK/EK was suggested by our earlier studies. The activity of the latter isoform was shown to be strongly induced in liver when the rat was treated with certain chemical carcinogens [7,13] and a hepatotoxin, carbon tetrachloride (CCl4 ) [13,14]. The inducible form could be distinguished immunologically from the other two (42-kDa and 50-kDa) constitutively expressed forms [15]. Thus, it now becomes most likely that there exist at least three CK (CK/EK) isozymes in the rat, i.e. a 42-kDa, a 50-kDa and an inducible form. In contrast, no molecular information has so far been reported for any CK or EK in the mouse. It would be desirable to clone the mouse gene for future studies on characterization as well as evaluation of the physiological function of each CK/EK isozyme in a whole animal model. Here we report cDNA cloning and sequencing of the mouse homologs for the 42-kDa CK/EK and 50-kDa CK2 . In addition, the level of mRNA expression for the two isozymes was compared in several rat tissues. A cDNA library from 17-day Swiss Webster/NIH mouse whole embryo constructed in Vgt10 (oligo(dT)+random primed, CLML5000a from Clontech) was screened with a DIG-labeled (PCR DIG Probe Synthesis Kit from Boehringer Mannheim) probe which was prepared by PCR with either 42-kDa CK/EK cDNA (clone 3-1 [4]) or reverse-transcribed (SuperScript Preampli¢cation System for First
2 The nomenclature for the 50-kDa CK or 42-kDa CK/EK in this paper is essentially according to that in the nucleotide sequence databases (DDBJ, EMBL and GenBank).
Strand cDNA Synthesis from Gibco-BRL) rat kidney cDNA as a template. The primers used were 5PGTACCTAAAGCAGATCCAGG-3P (upper) and 5P-GAACCGAGATTGGGCGTATT-3P (lower) for the 42-kDa CK/EK and 5P-ACTGGAGCAGTTTATCCC-3P (upper) and ACCAAGCTTCCTCTTCTG-3P (lower) for the 50-kDa CK. The latter primers were chosen based on the sequence reported by Uchida and Yamashita [5]. The sizes of resulting DIGlabeled PCR products were 553 bp and 757 bp for the 42-kDa CK/EK and 50-kDa CK, respectively. Several positive clones were isolated with either of the probes. Their sequences were determined (ABI PRISM 377 DNA Sequencer) after subcloning each cDNA insert into pT7Blue (from Novagen). For the 50-kDa CK, a clone (termed m3-3) containing almost full length cDNA (2430 bp) could be obtained (Fig. 1). After the ¢rst stop codon (TGA) at nucleotides 49^51, there was a long ORF (nucleotide positions 163^1470) which could encode a polypeptide comprising 435 amino acid residues with a calculated molecular size of 49 920. The entire nucleotide sequence and the coding sequence were found to have 94.3% and 96.0% homology (by GENETYXSV), respectively, to those for the rat 50-kDa CK. It should be noted that an extremely long 3P-UTR exists in their cDNA sequences for both mouse and rat 50-kDa CK isozymes. On the other hand, we could not obtain a single clone which contained the entire cDNA sequence for the 42-kDa CK/EK, but obtained two independent clones (m6-1 and m2-1) including partially overlapping sequences which together gave a composite sequence of full length cDNA (2263 bp). Clone m6-1 had a probable 169bp intron sequence (nucleotide positions 746^747) at the identical site as was found previously in the isolated rat cDNA clones for 42-kDa CK/EK (clones 21-2, 22-3 and 24-4 [4]). The other clone, m2-1, was found to have another three probable introns, 181 bp (880^881), 83 bp (1035^1036) and 122 bp (1117^ 1118), in addition to the 169-bp one, which were predicted to be introns by comparison with the rat 42-kDa CK/EK cDNA sequence. In the composite mouse 42-kDa CK/EK cDNA sequence without introns (1708 bp, Fig. 2), there was found a long ORF (nucleotide positions 300^1484) after the ¢rst stop codon TGA at 261^263, which could encode a polypeptide comprising 394 amino acid residues with a
BBALIP 50409 29-7-98
C. Aoyama et al. / Biochimica et Biophysica Acta 1393 (1998) 179^185
181
Fig. 1. Nucleotide sequence of a cDNA clone m3-3 and the deduced amino acid sequence for the mouse 50-kDa CK. Nucleotide and amino acid positions are indicated in the right margin. Amino acids are denoted by the one-letter code. An in-frame stop codon TAG in the 5P non-coding region and a putative polyadenylation signal AATAAA are boxed. Brenner's [16] consensus amino acid sequence for phosphotransferase (HXDhXXXNhhhTD, where h stands for large hydrophobic amino acids, FLIMVWY) is underlined with several key amino acids outlined.
calculated molecular size of 45 125. The entire nucleotide sequence and the coding sequence homologies were 91.3% and 92.4%, respectively, to those for the rat 42-kDa CK/EK.
The predicted amino acid sequences for both mouse and rat 50-kDa CK and 42-kDa CK/EK are compared in Fig. 3A,B. It has been shown previously that there exist eight highly conserved domains (d-1
Fig. 2. A composite nucleotide sequence of cDNA clones m6-1 and m2-1 (without introns) and the deduced amino acid sequence for the mouse 42-kDa CK/EK. The arrows indicate the positions where probable intron sequences are inserted in clone m2-1. Other indications are as described in the legend to Fig. 1.
BBALIP 50409 29-7-98
182
C. Aoyama et al. / Biochimica et Biophysica Acta 1393 (1998) 179^185
BBALIP 50409 29-7-98
C. Aoyama et al. / Biochimica et Biophysica Acta 1393 (1998) 179^185
183
Fig. 3. Comparison of the deduced amino acid sequences of the mouse 50-kDa CK (A) and 42-kDa CK/EK (B) to those of the respective rat homologs. Amino acids are denoted by the one-letter code and their positions are indicated in the right margin. The mismatched amino acids between the two murine species are black-boxed. The previously discovered highly conserved domains between the 50-kDa CK and 42-kDa CK/EK from the rat [4] are boxed with numbering (d-1 to d-8).
6
to d-8, marked in boxes) in the predicted amino acid sequences between 50-kDa CK and 42-kDa CK/EK from the rat [4]. Most (if not all) of these domains are considered to be critical for the expression of CK/EK activity. The entire amino acid sequence of 50-kDa CK shows 97.9% matching (426 out of 435 amino acid residues) between the two murine species (Fig. 3A). No mismatch could be found within the highly conserved domains 1^8. On the other hand, there was relatively less matching (95.2%) in amino acid sequences of the 42-kDa CK/EK between the two species (Fig. 3B). Nineteen out of 394 amino acid residues were mismatched and two of them were found within domains 5 and 8, respectively. These results clearly demonstrated that the 50-kDa CK is much more conserved than the 42-kDa CK/ EK between the two murine species. This fact may indicate that there is, or has been evolutionally, a
critical di¡erence in their physiological functions between the two CK (CK/EK) isozymes. Next, the expression and/or abundance of mRNA for 42-kDa CK/EK and 50-kDa CK was compared in several rat tissues to understand the possible difference in their functions. First analysis was carried out by Northern blot in which an aliquot (usually 30 Wg) of freshly isolated total RNA (extracted with ISOGEN, a product from Nippon Gene Co., Japan) from several rat tissues was run in 1.0% agarose gel, then transferred onto a positively charged nylon membrane (Hybond N from Amersham). The membrane was subsequently hybridized with a freshly radio-labeled (Rediprime DNA Labelling System from Amersham, RPN 1633) probe, then subjected to autoradiography. The fragments of cDNA used as probes were the same (553 bp for 42-kDa CK/EK and 757 bp for 50-kDa CK) as those
Fig. 4. The size of mRNA for the 42-kDa CK/EK and 50-kDa CK in rat testis (A) and the tissue distribution of both types of CK (CK/EK) mRNA in the rat (B). (A) Freshly isolated total RNA (30 Wg) from rat testis was run in 1% agarose gel, then transferred onto a positively charged nylon membrane. The membrane was hybridized with a freshly prepared radioactive cDNA probe for either 42-kDa CK/EK (I) or 50-kDa CK (II) as described in the text, then subjected to autoradiography. The positions of 28S and 18S rRNAs are indicated. (B) Total RNAs from rat brain (lane 1), kidney (lane 2), lung (lane 3), testis (lane 4), normal liver (lane 5) and CCl4 -induced liver (lane 6) were reverse-transcribed with oligo(dT) as a primer, then the resultant ¢rst-strand cDNAs were used as templates for PCR with the gene-speci¢c primers (see text) for the 42-kDa CK/EK (a), 50-kDa CK (b) or G3PDH (c). The results from 25 (left), 30 (center) and 35 (right) PCR cycles are shown in each column. Annealing temparature was 54³C for all experiments. The approximate sizes of the respective PCR products are shown in the right margin.
BBALIP 50409 29-7-98
184
C. Aoyama et al. / Biochimica et Biophysica Acta 1393 (1998) 179^185
used for the plaque hybridization mentioned above. The hybridization temperature, 65³C, was chosen where cross-hybridization did not occur to a signi¢cant level. As shown in Fig. 4A, the major signal for the 42-kDa CK/EK was found to locate at the position just below a 18S rRNA band (1.9 kb) and that for the 50-kDa CK located at the middle of 28S (4.7 kb) and 18S rRNA bands, either of which roughly corresponded to the mRNA size predicted from the isolated cDNA size (1679 bp for the 42-kDa CK/EK [4] and 2540 bp for the 50-kDa CK [5]). While a Northern blot analysis clearly demonstrated that mRNAs for both 42-kDa and 50-kDa CK/EK isozymes are highly expressed in testis, the signals from other rat tissues (brain, liver, kidney, spleen and lung) were very low and their abundance could not e¤ciently be compared among the tissues. Thus, we next examined RT-PCR in which an aliquot of total RNA (usually 1 Wg) from rat tissues was ¢rst reversetranscribed with oligo(dT) as a primer, then a resultant ¢rst strand cDNA was ampli¢ed for several round of cycles with the isozyme-speci¢c primers. The primers for each isozyme were exactly the same as those used for the preparation of DIG-labeled probes mentioned above. As G3PDH primers for a control experiment, 5P-ACCACAGTCCATGCCATCAC-3P (upper) and 5P-TCCACCACCCTGTTGCTGTA-3P (lower) were used. In addition to the normal rat tissues, the RNA sample from CCl4 treated liver (1 ml/kg of CCl4 in corn oil had been administered intraperitoneally 9 h before the animal was killed as described previously [14]) was also examined because CCl4 was found to induce CK activity severalfold in this tissue. As shown in Fig. 4B, RT-PCR study revealed again that expression of mRNA for either isozyme was highest in testis. Among other tissues examined, relatively higher expression of 50-kDa CK mRNA was detected in both brain and CCl4 -treated liver whereas almost the same level of 42-kDa CK/EK mRNA was shown to be expressed. It should be noted that, in lung, there was found to exist another mRNA fragment of a larger size (approximately 740 bp) for 42-kDa CK/ EK. This may indicate that a unique splicing mechanism for the 42-kDa CK/EK mRNA could be involved in this tissue. The overall results clearly indicate that both 42-kDa CK/EK and 50-kDa CK isozymes appear to be ubiquitously expressed en-
zymes and their expression occurs at the highest level in testis of the rat tissues examined. Finally, on the basis of cDNA sequences for mouse 42-kDa CK/EK and 50-kDa CK presented in this report, we are currently isolating their genomic clones. This work was supported in part by a Grant-inAid (8672503) for Scienti¢c Research from the Ministry of Education, Science, Sports and Culture of Japan. K.N. was supported by Research Fellowships of the Japan Society for the Promotion of Science for Young Scientists. C.A. is a visiting graduate student from Kyoritsu College of Pharmacy, Japan.
References [1] K. Ishidate, Y. Nakazawa, Choline/ethanolamine kinase from rat kidney, Methods Enzymol. 209 (1992) 121^134. [2] T.J. Porter, C. Kent, Choline/ethanolamine kinase from rat liver, Methods Enzymol. 209 (1992) 134^146. [3] T. Uchida, S. Yamashita, Choline/ethanolamine kinase from rat brain, Methods Enzymol. 209 (1992) 147^153. [4] C. Aoyama, K. Nakashima, M. Matsui, K. Ishidate, Complementary DNA sequence for a 42-kDa rat kidney choline/ ethanolamine kinase, Biochim. Biophys. Acta 1390 (1998) 1^ 7. [5] T. Uchida, S. Yamashita, Molecular cloning, characterization, and expression in Escherichia coli of a cDNA encoding mammalian choline kinase, J. Biol. Chem. 267 (1992) 10156^ 10162. [6] K. Ishidate, Choline/ethanolamine kinase from mammalian tissues, Biochim. Biophys. Acta 1348 (1997) 70^78. [7] K. Ishidate, M. Tsuruoka, Y. Nakazawa, Induction of choline kinase by polycyclic aromatic hydrocarbon carcinogens in rat liver, Biochem. Biophys. Res. Commun. 96 (1980) 946^952. [8] Z. Kiss, M. Tomono, Wortmannin inhibits carcinogenstimulated phosphorylation of ethanolamine and choline, FEBS Lett. 358 (1995) 243^246. [9] I.G. Macara, Elevated phosphocholine concentration in rastransformed NIH 3T3 cells arises from increased choline kinase activity, not from phosphatidylcholine breakdown, Mol. Cell. Biol. 9 (1989) 325^328. [10] D. Teegarden, E.J. Taparowsky, C. Kent, Altered phosphatidylcholine metabolism in C3H10T1/2 cells transfected with the Harvey-ras oncogene, J. Biol. Chem. 265 (1990) 6042^ 6047. [11] S. Ratnam, C. Kent, Early increase in choline kinase activity upon induction of the H-ras oncogene in mouse ¢broblast cell lines, Arch. Biochem. Biophys. 323 (1995) 313^322. [12] Z. Kiss, K.S. Crilly, Ha-ras stimulates uptake and phospho-
BBALIP 50409 29-7-98
C. Aoyama et al. / Biochimica et Biophysica Acta 1393 (1998) 179^185 rylation of ethanolamine: inhibition by wortmannin, FEBS Lett. 357 (1995) 279^282. [13] K. Tadokoro, K. Ishidate, Y. Nakazawa, Evidence for the existence of isozymes of choline kinase and their selective induction in 3-methylcholanthrene- or carbon tetrachloridetreated rat liver, Biochim. Biophys. Acta 835 (1985) 501^513. [14] K. Ishidate, S. Enosawa, Y. Nakazawa, Actinomycin D-sensitive induction of choline kinase by carbon tetrachloride
185
intoxication in rat liver, Biochem. Biophys. Res. Commun. 111 (1983) 683^689. [15] K. Ishidate, K. Nakagomi, Y. Nakazawa, Complete puri¢cation of choline kinase from rat kidney and preparation of rabbit antibody against rat kidney choline kinase, J. Biol. Chem. 259 (1984) 14706^14710. [16] S. Brenner, Phosphotransferase sequence homology, Nature 329 (1987) 21.
BBALIP 50409 29-7-98
Short sequence-paper
Molecular cloning of mouse choline kinase and choline/ethanolamine kinase: their sequence comparison to the respective rat homologs1 Chieko Aoyama a
a;b
, Kinichi Nakashima a , Kozo Ishidate
a;
*
Department of Molecular Cell Biology, Medical Research Institute, Tokyo Medical and Dental University, 2-3-10 Kanda-surugadai, Chiyodaku, Tokyo 101-0062, Japan b Department of Hygienic Chemistry, Kyoritsu College of Pharmacy, Minatoku, Tokyo 105, Japan Received 23 March 1998; accepted 11 May 1998
Abstract Complementary DNAs homologous to a rat 42-kDa choline/ethanolamine kinase [C. Aoyama et al., Biochim. Biophys. Acta 1390 (1998) 1^7] and to a 50-kDa choline kinase [T. Uchida and S. Yamashita, J. Biol. Chem. 267 (1992) 10156^10162] were isolated from a 17-day post coitum mouse embryo cDNA library and their sequences were compared with the two murine species, respectively. The nucleotide sequence homology (within the coding sequence) between mouse and rat 50-kDa choline kinases (96.0%) was considerably higher than that between their 42-kDa choline/ethanolamine kinases (92.4%). Northern blot and RT-PCR studies on several rat tissues demonstrated that both isozymes are expressed ubiquitously with the highest level in testis. z 1998 Elsevier Science B.V. All rights reserved. Keywords: Choline kinase; Ethanolamine kinase; cDNA cloning; Isozyme; Phosphocholine; Phosphoethanolamine; mRNA expression
Choline kinase (CK) (ATP:choline phosphotransferase, EC 2.7.1.32) and ethanolamine kinase (EK) (ATP:ethanolamine phosphotransferase, EC 2.7.1.82) catalyze the phosphorylation of choline and ethanolamine by ATP yielding phosphocholine/ phosphoethanolamine and ADP. This enzyme step Abbreviations: CK, choline kinase; EK, ethanolamine kinase; PC, phosphatidylcholine; PE, phosphatidylethanolamine; ORF, open reading frame; DIG, digoxigenin; UTR, untranslated region; RT-PCR, reverse transcription-polymerase chain reaction; G3PDH, glyceraldehyde-3-phosphate dehydrogenase * Corresponding author. Fax: +81 (3) 5280 8065; E-mail: [email protected] 1 The nucleotide sequence data reported in this paper have appeared in the DDBJ, EMBL and GenBank nucleotide sequence databases with the following accession numbers: AB011000 and AB011001 for the mouse 42-kDa choline/ethanolamine kinase and AB011002 for the mouse 50-kDa choline kinase.
commits choline/ethanolamine to the CDP-choline/ CDP-ethanolamine pathway for the biosynthesis of phosphatidylcholine (PC)/phosphatidylethanolamine (PE) in all animal cells. In most (if not all) mammalian tissues, both reactions have been shown to be catalyzed by the same enzyme protein(s). It has also been demonstrated that the enzyme does not exist in one particular active form but exists as several isozymes, some of which have been puri¢ed from rat tissues to apparent homogeneity [1^3] and their cDNA cloned [4,5]. In recent years, phosphocholine (and probably also phosphoethanolamine) has been proposed to be a possible second messenger in the process leading to the induction of DNA synthesis by several growth factors (reviewed in [6]). In addition, certain chemical carcinogens [7,8] and/or ras/raf oncogene transfection [9^12] have been shown to elevate CK/EK ac-
0005-2760 / 98 / $19.00 ß 1998 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 5 - 2 7 6 0 ( 9 8 ) 0 0 0 6 2 - 9
BBALIP 50409 29-7-98
180
C. Aoyama et al. / Biochimica et Biophysica Acta 1393 (1998) 179^185
tivity in the cell. Although the exact mechanism leading to activation or induction of CK/EK as well as the exact site of phosphocholine participation in mitogenic signal pathway remain to be determined, isolation and characterization of each CK/EK gene would further prove the critical role of each isozyme in PC/PE biosynthesis and/or in mitogenic signal transduction in mammalian cells. In a previous paper [4], we reported cDNA cloning for a 42-kDa rat kidney CK/EK. Its sequence was shown to have only limited homology with that for a 50-kDa form of rat liver CK reported by Uchida and Yamashita [5], indicating that these two isoforms in the rat must be derived from separate genes. In addition, the existence of a third isoform of CK/EK was suggested by our earlier studies. The activity of the latter isoform was shown to be strongly induced in liver when the rat was treated with certain chemical carcinogens [7,13] and a hepatotoxin, carbon tetrachloride (CCl4 ) [13,14]. The inducible form could be distinguished immunologically from the other two (42-kDa and 50-kDa) constitutively expressed forms [15]. Thus, it now becomes most likely that there exist at least three CK (CK/EK) isozymes in the rat, i.e. a 42-kDa, a 50-kDa and an inducible form. In contrast, no molecular information has so far been reported for any CK or EK in the mouse. It would be desirable to clone the mouse gene for future studies on characterization as well as evaluation of the physiological function of each CK/EK isozyme in a whole animal model. Here we report cDNA cloning and sequencing of the mouse homologs for the 42-kDa CK/EK and 50-kDa CK2 . In addition, the level of mRNA expression for the two isozymes was compared in several rat tissues. A cDNA library from 17-day Swiss Webster/NIH mouse whole embryo constructed in Vgt10 (oligo(dT)+random primed, CLML5000a from Clontech) was screened with a DIG-labeled (PCR DIG Probe Synthesis Kit from Boehringer Mannheim) probe which was prepared by PCR with either 42-kDa CK/EK cDNA (clone 3-1 [4]) or reverse-transcribed (SuperScript Preampli¢cation System for First
2 The nomenclature for the 50-kDa CK or 42-kDa CK/EK in this paper is essentially according to that in the nucleotide sequence databases (DDBJ, EMBL and GenBank).
Strand cDNA Synthesis from Gibco-BRL) rat kidney cDNA as a template. The primers used were 5PGTACCTAAAGCAGATCCAGG-3P (upper) and 5P-GAACCGAGATTGGGCGTATT-3P (lower) for the 42-kDa CK/EK and 5P-ACTGGAGCAGTTTATCCC-3P (upper) and ACCAAGCTTCCTCTTCTG-3P (lower) for the 50-kDa CK. The latter primers were chosen based on the sequence reported by Uchida and Yamashita [5]. The sizes of resulting DIGlabeled PCR products were 553 bp and 757 bp for the 42-kDa CK/EK and 50-kDa CK, respectively. Several positive clones were isolated with either of the probes. Their sequences were determined (ABI PRISM 377 DNA Sequencer) after subcloning each cDNA insert into pT7Blue (from Novagen). For the 50-kDa CK, a clone (termed m3-3) containing almost full length cDNA (2430 bp) could be obtained (Fig. 1). After the ¢rst stop codon (TGA) at nucleotides 49^51, there was a long ORF (nucleotide positions 163^1470) which could encode a polypeptide comprising 435 amino acid residues with a calculated molecular size of 49 920. The entire nucleotide sequence and the coding sequence were found to have 94.3% and 96.0% homology (by GENETYXSV), respectively, to those for the rat 50-kDa CK. It should be noted that an extremely long 3P-UTR exists in their cDNA sequences for both mouse and rat 50-kDa CK isozymes. On the other hand, we could not obtain a single clone which contained the entire cDNA sequence for the 42-kDa CK/EK, but obtained two independent clones (m6-1 and m2-1) including partially overlapping sequences which together gave a composite sequence of full length cDNA (2263 bp). Clone m6-1 had a probable 169bp intron sequence (nucleotide positions 746^747) at the identical site as was found previously in the isolated rat cDNA clones for 42-kDa CK/EK (clones 21-2, 22-3 and 24-4 [4]). The other clone, m2-1, was found to have another three probable introns, 181 bp (880^881), 83 bp (1035^1036) and 122 bp (1117^ 1118), in addition to the 169-bp one, which were predicted to be introns by comparison with the rat 42-kDa CK/EK cDNA sequence. In the composite mouse 42-kDa CK/EK cDNA sequence without introns (1708 bp, Fig. 2), there was found a long ORF (nucleotide positions 300^1484) after the ¢rst stop codon TGA at 261^263, which could encode a polypeptide comprising 394 amino acid residues with a
BBALIP 50409 29-7-98
C. Aoyama et al. / Biochimica et Biophysica Acta 1393 (1998) 179^185
181
Fig. 1. Nucleotide sequence of a cDNA clone m3-3 and the deduced amino acid sequence for the mouse 50-kDa CK. Nucleotide and amino acid positions are indicated in the right margin. Amino acids are denoted by the one-letter code. An in-frame stop codon TAG in the 5P non-coding region and a putative polyadenylation signal AATAAA are boxed. Brenner's [16] consensus amino acid sequence for phosphotransferase (HXDhXXXNhhhTD, where h stands for large hydrophobic amino acids, FLIMVWY) is underlined with several key amino acids outlined.
calculated molecular size of 45 125. The entire nucleotide sequence and the coding sequence homologies were 91.3% and 92.4%, respectively, to those for the rat 42-kDa CK/EK.
The predicted amino acid sequences for both mouse and rat 50-kDa CK and 42-kDa CK/EK are compared in Fig. 3A,B. It has been shown previously that there exist eight highly conserved domains (d-1
Fig. 2. A composite nucleotide sequence of cDNA clones m6-1 and m2-1 (without introns) and the deduced amino acid sequence for the mouse 42-kDa CK/EK. The arrows indicate the positions where probable intron sequences are inserted in clone m2-1. Other indications are as described in the legend to Fig. 1.
BBALIP 50409 29-7-98
182
C. Aoyama et al. / Biochimica et Biophysica Acta 1393 (1998) 179^185
BBALIP 50409 29-7-98
C. Aoyama et al. / Biochimica et Biophysica Acta 1393 (1998) 179^185
183
Fig. 3. Comparison of the deduced amino acid sequences of the mouse 50-kDa CK (A) and 42-kDa CK/EK (B) to those of the respective rat homologs. Amino acids are denoted by the one-letter code and their positions are indicated in the right margin. The mismatched amino acids between the two murine species are black-boxed. The previously discovered highly conserved domains between the 50-kDa CK and 42-kDa CK/EK from the rat [4] are boxed with numbering (d-1 to d-8).
6
to d-8, marked in boxes) in the predicted amino acid sequences between 50-kDa CK and 42-kDa CK/EK from the rat [4]. Most (if not all) of these domains are considered to be critical for the expression of CK/EK activity. The entire amino acid sequence of 50-kDa CK shows 97.9% matching (426 out of 435 amino acid residues) between the two murine species (Fig. 3A). No mismatch could be found within the highly conserved domains 1^8. On the other hand, there was relatively less matching (95.2%) in amino acid sequences of the 42-kDa CK/EK between the two species (Fig. 3B). Nineteen out of 394 amino acid residues were mismatched and two of them were found within domains 5 and 8, respectively. These results clearly demonstrated that the 50-kDa CK is much more conserved than the 42-kDa CK/ EK between the two murine species. This fact may indicate that there is, or has been evolutionally, a
critical di¡erence in their physiological functions between the two CK (CK/EK) isozymes. Next, the expression and/or abundance of mRNA for 42-kDa CK/EK and 50-kDa CK was compared in several rat tissues to understand the possible difference in their functions. First analysis was carried out by Northern blot in which an aliquot (usually 30 Wg) of freshly isolated total RNA (extracted with ISOGEN, a product from Nippon Gene Co., Japan) from several rat tissues was run in 1.0% agarose gel, then transferred onto a positively charged nylon membrane (Hybond N from Amersham). The membrane was subsequently hybridized with a freshly radio-labeled (Rediprime DNA Labelling System from Amersham, RPN 1633) probe, then subjected to autoradiography. The fragments of cDNA used as probes were the same (553 bp for 42-kDa CK/EK and 757 bp for 50-kDa CK) as those
Fig. 4. The size of mRNA for the 42-kDa CK/EK and 50-kDa CK in rat testis (A) and the tissue distribution of both types of CK (CK/EK) mRNA in the rat (B). (A) Freshly isolated total RNA (30 Wg) from rat testis was run in 1% agarose gel, then transferred onto a positively charged nylon membrane. The membrane was hybridized with a freshly prepared radioactive cDNA probe for either 42-kDa CK/EK (I) or 50-kDa CK (II) as described in the text, then subjected to autoradiography. The positions of 28S and 18S rRNAs are indicated. (B) Total RNAs from rat brain (lane 1), kidney (lane 2), lung (lane 3), testis (lane 4), normal liver (lane 5) and CCl4 -induced liver (lane 6) were reverse-transcribed with oligo(dT) as a primer, then the resultant ¢rst-strand cDNAs were used as templates for PCR with the gene-speci¢c primers (see text) for the 42-kDa CK/EK (a), 50-kDa CK (b) or G3PDH (c). The results from 25 (left), 30 (center) and 35 (right) PCR cycles are shown in each column. Annealing temparature was 54³C for all experiments. The approximate sizes of the respective PCR products are shown in the right margin.
BBALIP 50409 29-7-98
184
C. Aoyama et al. / Biochimica et Biophysica Acta 1393 (1998) 179^185
used for the plaque hybridization mentioned above. The hybridization temperature, 65³C, was chosen where cross-hybridization did not occur to a signi¢cant level. As shown in Fig. 4A, the major signal for the 42-kDa CK/EK was found to locate at the position just below a 18S rRNA band (1.9 kb) and that for the 50-kDa CK located at the middle of 28S (4.7 kb) and 18S rRNA bands, either of which roughly corresponded to the mRNA size predicted from the isolated cDNA size (1679 bp for the 42-kDa CK/EK [4] and 2540 bp for the 50-kDa CK [5]). While a Northern blot analysis clearly demonstrated that mRNAs for both 42-kDa and 50-kDa CK/EK isozymes are highly expressed in testis, the signals from other rat tissues (brain, liver, kidney, spleen and lung) were very low and their abundance could not e¤ciently be compared among the tissues. Thus, we next examined RT-PCR in which an aliquot of total RNA (usually 1 Wg) from rat tissues was ¢rst reversetranscribed with oligo(dT) as a primer, then a resultant ¢rst strand cDNA was ampli¢ed for several round of cycles with the isozyme-speci¢c primers. The primers for each isozyme were exactly the same as those used for the preparation of DIG-labeled probes mentioned above. As G3PDH primers for a control experiment, 5P-ACCACAGTCCATGCCATCAC-3P (upper) and 5P-TCCACCACCCTGTTGCTGTA-3P (lower) were used. In addition to the normal rat tissues, the RNA sample from CCl4 treated liver (1 ml/kg of CCl4 in corn oil had been administered intraperitoneally 9 h before the animal was killed as described previously [14]) was also examined because CCl4 was found to induce CK activity severalfold in this tissue. As shown in Fig. 4B, RT-PCR study revealed again that expression of mRNA for either isozyme was highest in testis. Among other tissues examined, relatively higher expression of 50-kDa CK mRNA was detected in both brain and CCl4 -treated liver whereas almost the same level of 42-kDa CK/EK mRNA was shown to be expressed. It should be noted that, in lung, there was found to exist another mRNA fragment of a larger size (approximately 740 bp) for 42-kDa CK/ EK. This may indicate that a unique splicing mechanism for the 42-kDa CK/EK mRNA could be involved in this tissue. The overall results clearly indicate that both 42-kDa CK/EK and 50-kDa CK isozymes appear to be ubiquitously expressed en-
zymes and their expression occurs at the highest level in testis of the rat tissues examined. Finally, on the basis of cDNA sequences for mouse 42-kDa CK/EK and 50-kDa CK presented in this report, we are currently isolating their genomic clones. This work was supported in part by a Grant-inAid (8672503) for Scienti¢c Research from the Ministry of Education, Science, Sports and Culture of Japan. K.N. was supported by Research Fellowships of the Japan Society for the Promotion of Science for Young Scientists. C.A. is a visiting graduate student from Kyoritsu College of Pharmacy, Japan.
References [1] K. Ishidate, Y. Nakazawa, Choline/ethanolamine kinase from rat kidney, Methods Enzymol. 209 (1992) 121^134. [2] T.J. Porter, C. Kent, Choline/ethanolamine kinase from rat liver, Methods Enzymol. 209 (1992) 134^146. [3] T. Uchida, S. Yamashita, Choline/ethanolamine kinase from rat brain, Methods Enzymol. 209 (1992) 147^153. [4] C. Aoyama, K. Nakashima, M. Matsui, K. Ishidate, Complementary DNA sequence for a 42-kDa rat kidney choline/ ethanolamine kinase, Biochim. Biophys. Acta 1390 (1998) 1^ 7. [5] T. Uchida, S. Yamashita, Molecular cloning, characterization, and expression in Escherichia coli of a cDNA encoding mammalian choline kinase, J. Biol. Chem. 267 (1992) 10156^ 10162. [6] K. Ishidate, Choline/ethanolamine kinase from mammalian tissues, Biochim. Biophys. Acta 1348 (1997) 70^78. [7] K. Ishidate, M. Tsuruoka, Y. Nakazawa, Induction of choline kinase by polycyclic aromatic hydrocarbon carcinogens in rat liver, Biochem. Biophys. Res. Commun. 96 (1980) 946^952. [8] Z. Kiss, M. Tomono, Wortmannin inhibits carcinogenstimulated phosphorylation of ethanolamine and choline, FEBS Lett. 358 (1995) 243^246. [9] I.G. Macara, Elevated phosphocholine concentration in rastransformed NIH 3T3 cells arises from increased choline kinase activity, not from phosphatidylcholine breakdown, Mol. Cell. Biol. 9 (1989) 325^328. [10] D. Teegarden, E.J. Taparowsky, C. Kent, Altered phosphatidylcholine metabolism in C3H10T1/2 cells transfected with the Harvey-ras oncogene, J. Biol. Chem. 265 (1990) 6042^ 6047. [11] S. Ratnam, C. Kent, Early increase in choline kinase activity upon induction of the H-ras oncogene in mouse ¢broblast cell lines, Arch. Biochem. Biophys. 323 (1995) 313^322. [12] Z. Kiss, K.S. Crilly, Ha-ras stimulates uptake and phospho-
BBALIP 50409 29-7-98
C. Aoyama et al. / Biochimica et Biophysica Acta 1393 (1998) 179^185 rylation of ethanolamine: inhibition by wortmannin, FEBS Lett. 357 (1995) 279^282. [13] K. Tadokoro, K. Ishidate, Y. Nakazawa, Evidence for the existence of isozymes of choline kinase and their selective induction in 3-methylcholanthrene- or carbon tetrachloridetreated rat liver, Biochim. Biophys. Acta 835 (1985) 501^513. [14] K. Ishidate, S. Enosawa, Y. Nakazawa, Actinomycin D-sensitive induction of choline kinase by carbon tetrachloride
185
intoxication in rat liver, Biochem. Biophys. Res. Commun. 111 (1983) 683^689. [15] K. Ishidate, K. Nakagomi, Y. Nakazawa, Complete puri¢cation of choline kinase from rat kidney and preparation of rabbit antibody against rat kidney choline kinase, J. Biol. Chem. 259 (1984) 14706^14710. [16] S. Brenner, Phosphotransferase sequence homology, Nature 329 (1987) 21.
BBALIP 50409 29-7-98