Cloning, Characterization, and Expression in Escherichia coli of Three Creatine Kinase Muscle Isoenzyme cDNAs from Carp (Cyprinus carpio) Striated Muscle*

In vertebrates, the creatine kinase isoenzyme family consists of four types of isoforms: cytosolic muscle type (M-CK), cytosolic brain type (B-CK), mitochondrial ubiquitous, acidic type (Miu-CK), and mitochondrial sarcomeric, basic type (Mis-CK). Until recently, the existence of more than one subisoform of CK isoenzyme has been demonstrated only in fishes by starch gel electrophoresis. We report herein the isolation of three full-length cDNAs that correspond to three closely related creatine kinase M-CK genes from common carp ( Cyprinus carpio ), designated the M1-CK, M2-CK, and M3-CK genes. Using oligonucleotide probes that correspond to the same region but with the most variable sequences, different restricted genomic hybridization patterns have been obtained. These Southern blot results indicate that the three cDNAs come from different genes. Northern blot analysis using probes that correspond to the 3 * untranslated regions further show that all three subisoforms are expressed specifically in carp muscle. The deduced amino acid sequences of these three subisoforms of carp M-CK show about 85% identity to mamma-lian M-CK isoenzyme. Finally, the three cDNAs have been expressed in Escherichia coli with a molecular mass of approximately 43,000 Da, and these recombinant proteins exhibit creatine kinase activity. All of these data suggest that the M-CK

nucleotide sequences; the predicted amino acid sequences of the three subisoforms of M-CK along with data on their tissue distribution; and the E. coli-expressed recombinant M-CK protein catalytic activities. Also, since the common carp is a poikilothermic fish, we have focused our attention on the catalytic activity of the E. coli-expressed recombinant M-CK proteins of carp at various temperatures.

EXPERIMENTAL PROCEDURES
Experimental Animals-Common carp (Cyprinus carpio) were maintained in tanks of circulating aerated water at 25°C for 3 months under a 12-h day length and were fed a commercial diet.
Isolation of cDNA Clones-Poly(A) ϩ RNA was prepared from striated muscle of the common carp by an established procedure (26). The double-stranded cDNA was supplied with EcoRI and XhoI linkers and was inserted into the -ZAP II vector (Stratagene). The library was packaged in the Gigapack II gold packaging extract (Stratagene) and was plated on E. coli XL1-blue MRFЈ cells (27). The probe for screening the carp cDNA library was synthesized by PCR using carp striated muscle cDNA as a template with a sense oligonucleotide primer (CK-5Ј, 5Ј-CAY AAY AAY CAY ATG GCN AA-3Ј, alignment positions amino acids 26 -32 in the M-CK isoenzyme) and an antisense oligonucleotide primer (CK-3Ј, 5Ј-CAT RTT NCC NCC YTT YTG CAT-3Ј, alignment positions amino acids 239 -246 in the M-CK isoenzyme). The amplified products were subcloned into pUC 19 (New England Biolabs) and transformed into JM109 cells. Preliminary Northern blot analysis using the PCR product (663 bp) as a probe revealed that it was expressed in carp striated muscle. After in vitro packaging, 1 ϫ 10 6 primary phages were amplified and plated at a density of ϳ50,000 plaques/plate (15 cm), with 12 plates being screened. The probe was prepared by rediPrime labeling system (Amersham Pharmacia Biotech) with [␣-32 P]dATP (3000 Ci/ mmol, Amersham Pharmacia Biotech). Prehybridization and hybridization were carried out in standard hybridization buffer at 42°C for 16 h (27). Filters were finally washed with 0.1ϫ SSC, 0.1% SDS for 1 h twice at 65°C. Positive clones were isolated, purified, and sequenced.
Sequence Analysis and Computer Homology Search-The nucleotide sequences were determined by doubled-stranded sequencing according to the dideoxy chain termination method using an ABI PRISM dye terminator cycle sequencing kit (Applied Biosystems). The nucleotide sequences were then analyzed with an Applied Biosystems 377A automated DNA sequencer. A sequence homology search and comparison were performed with computer programs of the GCG sequence analysis software package (Genetic Computer Group, Madison, WI). The homologous sequences searched included GenBank TM accession numbers M11660 (dog M-CK), X03233 (mouse M-CK), M10140 (rat M-CK), M14780 (human M-CK), M10012 (chicken M-CK), M11508 (Torpedo M-CK), M14400 (rat B-CK), X59736 (rat Mis-CK), and X57937 (rat Miu-CK).
Southern Hybridization Analysis-Genomic DNA was prepared from carp muscle tissue with the proteinase K method (27). Southern blotting with nitrocellulose membrane (Hybond-C super, Amersham Pharmacia Biotech) was carried out following standard procedures from Maniatis et al. (27). Restriction digestion of genomic and plasmid DNA was carried out overnight according to the manufacturer's specifications (New England Biolabs). Oligodeoxynucleotide probes were designed to cover the region 27 nucleotides after the stop codon of the three CK cDNAs that contained the most variable sequences (for M1-CK, 5Ј-AGC GGG GAG CCC TTC CAT TTT TTT CTA-3Ј; for M2-CK,  5Ј-AGC GGG AGC CCT TCC TCT TTT TTC CTC-3Ј; and for M3-CK,  5Ј-AAT GGC AGA AGT GCT TTT CTT TTT TTA-3Ј). Each probe was 5Ј-end-labeled with [␥-32 P]ATP (5000 Ci/mmol; Amersham Pharmacia Biotech) to a specific activity of approximately 4 ϫ 10 6 cpm/ml using bacteriophage T4 polynucleotide kinase (Promega). The membranes were hybridized at a high stringency temperature (T H ) (27). Prehybridization and hybridization were carried out in standard hybridization buffer at 65°C for 16 h (27). The filters were finally washed with 6ϫ SSC, 0.1% SDS for 30 min at 52°C. The membranes were exposed to a PhosphorImager (Molecular Dynamics).
Northern Hybridization Analysis-Total RNA was isolated by the acid guanidinium isothiocyanate/phenol/chloroform method using TRIzol reagent (Life Technologies, Inc.), electrophoresed on 1.2% agarose-formaldehyde gels, and then transferred onto nitrocellulose membranes (Hybond-C super, Amersham Pharmacia Biotech) (28). Probes of different M-CK subisoforms were prepared with the entire 3Ј-UTRs as templates and labeled by the rediPrime labeling system (Amersham Pharmacia Biotech) with [␣-32 P]dCTP (3000 Ci/mmol; Amersham Pharmacia Biotech). Hybridization was carried out at 42°C overnight in standard hybridization buffer with 50% formamide (28). After hybridization, the membrane was finally washed twice at 68°C for 15 min in 0.1ϫ SSC, 0.1% SDS. The membrane was exposed to a PhosphorImager.
Expression Construction, Preparation, and Isolation of Recombinant Protein-The three full-length carp M-CK cDNA coding regions were amplified by PCR. To facilitate directional subcloning into the expression vector, forward and reverse primers were designed to contain restriction sites for NdeI and EcoRI at the 5Ј-and 3Ј-termini, respectively (BamHI for the 5Ј-terminal of M2-CK). Both restriction sites were absent in the M-CK cDNAs. The cDNA products were ligated into the pET-28a(ϩ) expression vector (Novagen), in which prokaryotic gene expression is driven by the T7 lac promoter. Nucleotide sequencing was carried out to ensure the absence of PCR-induced mutation. Strain BL21 (DE3) pLysS bacteria were transformed with the resulting cDNA constructs and screened for positive clones. The bacterial cultures were grown in LB broth until A 600 ϭ 0.3 and then were induced with isopropyl ␤-D-thiogalactoside. The bacteria were pelleted by centrifugation, resuspended in 20 mM Tris-HCl (pH 7.9) containing 5 mM imidazole and 0.5 M NaCl, and ultrasonicated. Purification of recombinant protein was achieved by chromatography on His-Bind resin (Novagen) following the pET system manual (Novagen). The eluted protein was dialyzed with 30 mM Tris-HCl (pH 7.0), 50 M EDTA, and 5 mM 2-mercaptoethanol and then concentrated 10-fold by a YM-10 filter (Amicom).
Immunoblot Analysis-Polyclonal goat anti-human creatine kinase MM isoenzyme was purchased from BIODESIGN. SDS-polyacrylamide gel electrophoresis was performed according to the method of Laemmli (29). After electrophoresis, the gel was stained with Coomassie Blue (Sigma) and then electrotransferred to a polyvinylidene difluoride membrane filter (Millipore Corp.) (28). The blotted filter was incubated in 20 mM Tris, pH 7.6, 137 mM NaCl, and 3% bovine serum albumin (fraction V, Sigma) for 1 h. The filter was incubated with a 1:1000 dilution of anti-M-CK serum, followed by interaction with anti-goat IgG alkaline phosphatase-conjugated antibody as a secondary antibody (Zymed Laboratories Inc.) and staining with 5-bromo-4-chloro-3-indolylphosphate (Pierce) and nitro blue tetrazolium (Pierce).
Isoenzyme Activity Assay-Protein concentration determination was performed using the Bradford method (30), using bovine serum albumin as the protein standard (Pierce). Creatine kinase activities were determined by the direct enzyme activity assay, which measured the creatine (Cr) concentration formed in the reverse reaction, PCr ϩ ADP 3 Cr ϩ ATP, using the colorimetric detection method (31,32). Samples of the recombinant protein were diluted 1:100 in a buffer containing 30 mM Tris, pH 7.0, 50 M EDTA, and 5 mM 2-mercaptoethanol. Reagents were from the creatine kinase isoenzyme colorimetric detection kit (Sigma Diagnostics) and were used according to the manufacturer's instructions modified for optimum assay conditions. Reaction mixtures containing 20 mM phosphocreatine (PCr) and 4 mM ADP were incubated for 30 min at 25°C. The reaction was stopped by the addition of p-hydroxymercuribenzoate, and then ␣-naphthol solution and diacetyl solution were added and incubated for 20 min at 37°C. The sample was measured at A 520 . CK activity was determined from the creatine standard calibration curve. One enzyme unit corresponds to 1 mol of creatine formed per min at pH 7.0 and 25°C. Under this condition, the analysis of the three subisoforms of carp M-CK isoenzymes at different temperatures was carried out.
Enzyme Kinetic Analysis-The same direct enzyme assay was used in the kinetic experiments. Reaction conditions for each individual kinetic experiment are described under "Results." The influence of PCr and ADP concentrations on creatine kinase activity was studied at 25°C in Trizma (Tris base) buffer (with Mg 2ϩ ion) (Sigma). K m for PCr was determined with various PCr (Sigma) concentrations (from 0.5 to 15 mM) and a fixed ADP (Sigma) concentration (4 mM). K m for ADP was determined with various ADP concentrations (from 0.1 to 10 mM) and a fixed PCr concentration (15 mM). For the determination of all kinetic parameters, initial velocity data were analyzed using the EZ-FIT program package by Perrella (33). Values of K m for PCr and ADP and of V max were calculated by a Lineweaver-Burk plot of three series of independent measurements. S.E. values for calculated K m and V max values are given. could be grouped into three PstI restriction patterns. One clone from each group of the cDNAs, designated M1-CK, M2-CK, and M3-CK was picked, and subjected to nucleotide sequence analysis.
The lengths for these three cDNA inserts, excluding the poly(A), were 1559, 1580, and 1508 bp for M1-CK, M2-CK, and   (Fig. 2). Computer alignment and search were carried out using the programs Pileup and Pretty of the GCG software package and protein sequence data banks. The identity between M1-CK and M2-CK proteins is as high as 96% (Fig. 2, Table I). For the M3-CK protein, the identity with the homologous M1-CK and M2-CK subisoforms is somewhat lower, at approximately 87%, due to differences within sequences of different domains, especially at the N terminus (Fig. 2). Analysis of the deduced amino acid sequences of the three carp muscle subisoforms revealed 83-87% identity with human and other species' creatine kinase muscle isoenzyme (Table I). In mammals and birds, the amino acid identities within each isoenzyme class range from 88 to 99% (24). Evidently, a higher degree of homology can be observed between the B-and the M-CKs (77-82%), while the homology between the cytosolic and the mitochondrial CK isoforms is lower (60 -65%) ( Table I) (24).
A previously defined "CK framework," consisting of the six most conserved sequence blocks and "diagnostic boxes," which are characteristic for any creatine kinase isoenzyme and which may serve to distinguish this isoenzyme from all others, could be observed in the protein sequences of the three carp M-CKs (24). The six highly conserved blocks are boxed in the carp M-CK consensus sequences in Fig. 2. These conserved sequences are flanked by regions that are less conserved, among them the N and C termini (Fig. 2). The absolutely conserved  Cys-283, alkylation of which is always paralleled by a very pronounced or even complete loss of enzymatic activity, is also present (41). A putative adenine nucleotide binding motif (glycine-rich loop) LGXGXXGXV and the absolutely conserved seven-amino acid sequence Cys-Pro-Ser-Asn-Leu-Gly-Thr, referred to as the active site, are also observed ( Fig. 2) (42,43). Analysis of Carp M-CK Genomic DNA Variants-Detection of the carp M-CK genes was by Southern hybridization. The results of the hybridization of the three restricted M-CK genes are shown in Fig. 3. The oligonucleotide probes specific for each of the three genes hybridized strongly to their respective target DNAs. The respective gene-specific probes hybridized to a 6.5-kb HindIII fragment, a 3.8-kb PstI fragment, and a 4.7-kb PvuII fragment for the carp M1-CK gene; a 5.2-kb HindIII fragment, a 2.2-kb PstI fragment, and a 3.4-kb PvuII fragment for the carp M2-CK gene; and a 16.2-kb HindIII fragment, a 5.8-kb PstI fragment, and a 3.5-kb PvuII fragment for the carp M3-CK gene (Fig. 3). In the standard plasmid DNA lanes that carried the different cDNAs and the restricted genomic lanes, only slight cross-hybridization could be observed between M1-CK and M2-CK and between M2-CK and M3-CK (Fig. 3). These results demonstrate that under the hybridization conditions used, M1-CK, M2-CK, and M3-CK genes could be detected by Southern analysis with these gene-specific oligonucleotide probes. Therefore, there are at least three M-CK genes in the carp genome.
Tissue Distribution of Carp M1-CK, M2-CK, and M3-CK mRNAs-Tissue distribution of the three carp M1-CK, M2-CK, and M3-CK mRNAs was characterized by Northern blot anal-ysis (Fig. 4). These three subisoform mRNAs (about 1.6 kb) in carp tissues were abundantly expressed in red and white muscles, whereas faint signals were observed in heart and no transcripts were detected in brain, kidney, gill, or liver. This distribution is similar to those of other vertebrate M-CK isoenzymes, which were found to express abundantly in striated muscle but with no expression in the brain (44). These results indicate that carp M1-CK, M2-CK, and M3-CK are subisoforms of M-CK isoenzymes.
Expression of the Three Carp M-CK Recombinant Subisoforms-To express and purify the recombinant carp M-CK proteins, a 1-L culture of BL21 (DE3) pLysS cells harboring the expression plasmid pETM1, pETM2, or pETM3 was grown and harvested. The SDS-polyacrylamide gel depicting the purified recombinant proteins is shown in Fig. 5A. Western analysis showed that they were all immunoreactive with anti-human M-CK antibody (Fig. 5B).  (Table III). Specific activities were determined under standard conditions of 20 mM phosphocreatine, 4 mM ADP, and 4 mM MgCl 2 . Kinetic parameters K m and V max were calculated from three sets of independent experiments, and in each set of experiments, three independent measurements were taken, with the phosphocreatine concentrations being varied between 0.5 and 15 mM while with constant concentrations of ADP at 4 mM and of MgCl 2 at 4 mM. The enzyme amount in our assay was 0.04 g, and the enzyme activity was linear up to 0.04 g of enzyme. Also, the enzyme activity was linear up to 30 min under the assay conditions; therefore, the end point creatine value was taken at 30 min for this reaction (data not shown). kb) in carp tissues were abundantly expressed in red and white muscle, whereas very faint signals were observed in heart; and no transcripts were detected in the brain, kidney, gill, or liver. Ethidium bromidestained 18S rRNAs are intended to show the loading differences in different lanes, and the images have been converted artificially from white to black.

Different Enzyme Catalytic Activities and Kinetic Parameters of Carp M-CK
M3-CK (70.8 Ϯ 0.8 units/mg) recombinant protein was lower (Table III). Also, while the K m values of ADP were determined, the values fluctuated widely from experiment to experiment. This assay method, for technical reasons, might not be suitable for accurate determination of K m values of ADP of CKs, since these K m values might be far below 100 M (45).
When specific activities were measured at different temperatures, the M1-CK protein was found to exhibit its highest specific activity at 37°C (64 unit/mg) and then to fall to around 56% at 20°C, and at 10°C only 18% activity was retained (Fig.  6). The temperature dependence of the M2-CK protein was very similar to that of human M-CK. The highest specific activity appeared at 37°C (63 unit/mg), was maintained at 85% activity from 30 to 25°C, and then was reduced further to 66% activity at 20°C and finally to 28% activity at 10°C (Fig. 6). In contrast, the temperature dependence of the M3-CK protein-specific activities was quite different. The highest activity was measured at 25°C (21 units/mg), and at 37 and 10°C, 41 and 28% activities were measured, respectively (Fig. 6). DISCUSSION In this report, we describe the cloning, characterization, and expression of three creatine kinase isoenzymes from common carp striated muscle. According to the nucleotide and amino acid sequence homology comparison, the three carp M-CKs should belong to cytosolic M-CKs. Southern hybridization analysis using oligonucleotides specific to individual cDNA as probes indicates that three carp M-CK cDNAs are encoded by different genes. From Northern blot analysis, mRNAs for M-CKs were detected in striated muscle (red, white, and cardiac muscle), and the expression levels varied in each muscle. However, with our probe design we could not tell whether there is any quantitative difference in expression levels of each enzyme. The Northern signal is faint in carp heart, and we speculate that there is the possibility that a mixture of MB and BB forms exists in heart (25). Another more interesting possibility is that another heart-specific M-CK subisoform exists whose DNA sequence in the corresponding position might be sufficiently different so that it only weakly cross-hybridized with our probes. All of these results indicate that these three carp CKs are muscle creatine kinase subisoforms.
In the experiments where the M-CK expression plasmids were transformed and expressed in E. coli cells, we detected the expressed recombinant proteins by immunoblot analysis with human muscle creatine kinase antibody. The three recombinant proteins exhibited somewhat different K m values, despite the amino acid homology between M1-CK and M2-CK being as high as 96%. As for V max values, M1-CK and M2-CK showed similar values, while that of M3-CK was 30% lower. The K m of carp M-CK has previously been found to be around 2.84 mM for PCr, and in light of our present findings, it is possible that this previous value was a measurement of a mixture of M-CKs (46). Finally, in the temperature dependence-specific activity studies, the decreasing trends of carp M1-CK and M2-CK enzyme activity when temperatures decreased from 37 to 10°C were somewhat similar. Yet, more interestingly, the specific activities of M3-CK recombinant protein showed a plateau between 30 and 20°C. All of these results suggest that these carp muscle creatine kinase subisoforms possess quite different enzyme properties.   Taken together, the three carp M-CKs are a new combination of cytosolic CK subisoforms that may have overlapping but not necessarily redundant physiological enzyme functions. Whether their mRNA expression patterns are similar or different at different developmental stages or at different environmental temperatures is an interesting question. Likewise, it is important to learn the different mechanisms or physiological roles played by the three carp M-CKs, if any, in energy homeostasis during environmental adaptation in ectothermic animals. Since there are multiple subisoforms of M-CK in carp, it would be important to learn whether heterodimers of M-CKs exist in a single carp muscle cell and whether there are differential subcellular localizations of these subisoforms.
Cold is a major environmental problem for all living organisms. Poikilothermic animals respond adaptively to chronic cold by a suite of cellular responses that compensate to varying extents for the rate-depressing effects of cooling. In ectothermic fish, body temperature is totally dependent on ambient temperature, and yet, even at low temperatures, carp have evolved strategies to maintain growth and swimming ability. Various review articles concerning possible acclimation mechanisms have been published (47,48). Hazel and Prosser (49) suggested that acclimation might involve a temperature-dependent synthesis of new proteins, which would require the expression of different sets of temperature-specific isoenzyme genes. Also, the nature and significance of changes in enzymatic myosin ATPase activity and the recruitment of different muscle fiber types in relation to acclimation and environmental temperature in carp have been reviewed (50,51). It has been reported that there are different myosin heavy chain isoform genes that are expressed at warm and cold environmental temperatures (52,53). Myofibrillar creatine kinase and myosin ATPase are associated with the same microenvironment and provide and utilize ATP for muscle contraction. Since M3-CK exhibited a specific activity peak at 25°C, we could imagine that some cold-specific subisoform carp M-CKs should exist to maintain normal muscle ability at lower environmental temperatures.
In conclusion, we have cloned three carp M-CK cDNAs encoded by different genes. The three M-CK subisoforms are not due to alternative initiation sites, alternative splicing, posttranslational glycosylation, phosphorylation, or proteolytic modification. These results suggest that multiple genes give rise to creatine kinase heterogeneity in carp.