HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 279, Issue 17, 17801-17809, April 23, 2004

This Article Abstract Full Text (PDF) All Versions of this Article: 279/17/17801    most recent M401382200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Yuan, C. Articles by Kent, C. Search for Related Content PubMed PubMed Citation Articles by Yuan, C. Articles by Kent, C. Social Bookmarking                      What's this?

Identification of Critical Residues of Choline Kinase A2 from Caenorhabditis elegans^*

Chong Yuan and Claudia Kent

From the Department of Biological Chemistry, University of Michigan Medical Center, Ann Arbor, Michigan 48109-0606

Received for publication, February 8, 2004

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Choline kinase catalyzes the phosphorylation of choline by ATP, the first committed step in the CDP-choline pathway for phosphatidylcholine biosynthesis. To begin to elucidate the mechanism of catalysis by this enzyme, choline kinase A-2 from Caenorhabditis elegans was analyzed by systematic mutagenesis of highly conserved residues followed by analysis of kinetic and structural parameters. Specifically, mutants were analyzed with respect to K_m and k_cat values for each substrate and Mg²⁺, inhibitory constants for Mg²⁺ and Ca²⁺, secondary structure as monitored by circular dichroism, and sensitivity to unfolding in guanidinium hydrochloride. The most severe impairment of catalysis occurred with the modification of Asp-255 and Asn-260, which are located in the conserved Brenner's phosphotransferase motif, and Asp-301 and Glu-303, in the signature choline kinase motif. For example, mutation of Asp-255 or Asp-301 to Ala eliminated detectable catalytic activity, and mutation of Asn-260 and Glu-303 to Ala decreased k_cat by 300- and 10-fold, respectively. Additionally, the K_m for Mg²⁺ for mutants N260A and E303A was approximately 30-fold higher than that of wild type. Several other residues (Ser-86, Arg-111, Glu-125, and Trp-387) were identified as being important: Catalytic efficiencies (k_cat/K_m) for the enzymes in which these residues were mutated to Ala were reduced to 2-25% of wild type. The high degree of structural similarity among choline kinase A-2, aminoglycoside phosphotransferases, and protein kinases, together with the results from this mutational analysis, indicates it is likely that these conserved residues are located at the catalytic core of choline kinase.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Phosphatidylcholine is the most abundant phospholipid in eukaryotic cells. It serves not only as a structural foundation for membrane bilayers, but also as a precursor for several lipid messengers (1, 2). In eukaryotes, there are two major routes for biosynthesis of phosphatidylcholine: the CDP-choline, or Kennedy, pathway and the pathway for the methylation of phosphatidylethanolamine. In the CDP-choline pathway, choline is phosphorylated by choline kinase, then the phosphocholine reacts with CTP to make CDP-choline, and finally the CDP-choline is transferred to sn-1,2-diacylglycerol to form phosphatidylcholine (3). The methylation pathway uses three molecules of S-adenosylmethionine in succession to convert phosphatidylethanolamine to phosphatidylcholine. In yeast the relative contribution of the two pathways to production of phosphatidylcholine depends on the supply of exogenous choline (4). In mammals, the CDP-choline pathway is responsible for phosphatidylcholine production in all cells and tissues, although the methylation of phosphatidylethanolamine also has a major role in liver (5).

Choline kinase (ATP:choline phosphotransferase, EC 2.7.1.32 [EC] ) catalyzes the initial step in the CDP-choline pathway. Since the discovery of choline kinase in 1953 (6), various forms of choline kinase from eukaryotes have been purified and characterized (7-10). These enzymes are specific for ATP as the phosphate donor, but often can use both choline and ethanolamine as substrates. Enzymes that are specific for phosphorylation of ethanolamine are also known (11). Choline kinase can be regulatory for phosphatidylcholine biosynthesis under certain conditions (12). Increased choline kinase activity is seen during early stages of mitogenic stimulation (13). In addition, elevated levels of choline kinase and its product, phosphocholine, have been found in human cancers, and the stimulation of choline kinase activity and/or expression by a ras oncogene has been proposed to be one of the mechanisms of this phenomenon (14-16). Therefore, inhibition of choline kinase has been targeted as a potential antitumor strategy (17). Moreover, recent evidence shows that choline kinase can be regulated at the level of protein itself, as the yeast choline kinase is activated by phosphorylation in vitro or in vivo (18, 19).

Biochemical evidence for multiple isoforms of choline kinase in mammals was verified by the identification of two mammalian genes for choline kinase, and , and splice variants thereof (8). Soybean also contains at least two genes encoding choline kinase (20). We have recently identified seven choline kinase-like genes in Caenorhabditis elegans, and have verified that at least four of these genes encode active choline kinases (21). The seven genes were divided into three families, A-C, with the A family being the most similar to mammalian choline kinases. We have purified and characterized two of the C. elegans isoforms, choline kinase A-2 (CKA-2)¹ and choline kinase B-2 (21).

Despite the significance of this enzyme in lipid metabolism and diverse cellular processes, the enzyme has received relatively little attention in the literature, especially with respect to enzymology. For example, there are no studies on the catalytic mechanism of choline/ethanolamine kinases. We have chosen to pursue the characterization of CKA-2 from C. elegans as a model choline kinase. This enzyme is as catalytically active as choline kinase from mammals and yeast and is quite similar to them in primary structure. Moreover, CKA-2 has been amenable to crystallization, and the crystal structure of CKA-2 has been determined recently (22). Surprisingly, the three-dimensional structure of CKA-2 is quite similar to those of protein kinases and aminoglycoside phosphotransferases, despite very low similarity in amino acid sequences among these enzymes. In this report, we have employed CKA-2 to probe the importance of key, conserved residues in catalysis by choline kinase. Additionally, guided by a comparison of the structures of CKA-2 and the well understood aminoglycoside phosphotransferase APH(3')-IIIa (23), we discuss the contributions of these specific residues to the catalysis of CKA-2 at the molecular level.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Materials—All oligonucleotide primers for mutagenesis were from Invitrogen. Restriction endonucleases were from New England Biolabs. Protease inhibitor (Complete Mini EDTA-Free) was from Roche Molecular Biochemicals. Guanidinium chloride, ATP, imidazole, and choline were from Sigma. Nickel-nitrilotriacetic acid-agarose was from Qiagen. [methyl-¹⁴C]Choline was from Amersham Biosciences. Invitrogen was the source of the Bac-to-Bac baculovirus expression system, Sf-9 cells adapted for serum-free growth, and Sf-900 II serum-free medium.

Generation of Recombinant Baculovirus of CKA-2 and Its Variants—A 1.2-kb DNA fragment encoding C. elegans CKA-2 (21) was cloned into the modified baculovirus donor plasmid pFastBacHT (pFBHT) using NdeI and XhoI sites, generating the sequence MSYYH₆DYDIPTTENLYFQGAMDPEFKGLLRH-CKA2, which contained 6 His residues and an rTEV protease cleavage site (underlined) at the amino terminus. This plasmid was then used as the template for all site-directed mutagenesis. The point mutations were introduced by overlap extension PCR (24). The codon GCN was used for conversion to Ala, GAA(G) for conversions to Glu, CAA(G) for conversions to Gln, GAC(T) for conversions to Asp, AAC(T) for conversions to Asn, and ACC for conversion to Thr. The resulting mutant fragments were placed into pFBHT using NdeI and XhoI. The sequences of all constructs were verified by automated DNA sequencing in the DNA sequencing core at the University of Michigan.

The preparation of virus encoding CKA-2 and its variants followed the instruction manual for the Bac-to-Bac baculovirus expression system, except that two rounds of virus amplification and 6 days for each round were carried out to reach the needed levels of protein expression.

Protein Expression and Purification—Sf-9 cells at the initial density of 1.0 x 10⁶ cells/ml were infected with recombinant virus and harvested at 72 h after infection by centrifugation at 4000 x g for 10 min at 4 °C. The cell pellets were frozen at -80 °C and thawed to facilitate lysis.

All subsequent steps were performed at 4 °C. Cell pellets from 500-ml cultures were resuspended in 8 ml of homogenization buffer (20 mM Tris-Cl, pH 8.0, 5 mM 2-mercaptoethanol) containing protease inhibitors (21) and then lysed by homogenization. Following centrifugation at 20,000 x g for 20 min, clear supernatants were collected and adjusted to 300 mM KCl, 1 mM imidazole, and 5% glycerol. The resultant supernatants were loaded onto the nickel-nitrilotriacetic acid-agarose column pre-equilibrated with buffer A (20 mM Tris-Cl, pH 8.0, 5 mM 2-mercaptoethanol, 5% glycerol) containing 300 mM KCl and1 mM imidazole. The column was washed sequentially with buffer B (buffer A containing 1 M KCl, 20 mM galactose, and 20 mM imidazole) to remove the majority of contaminants, then buffer C (buffer A containing 300 mM KCl and 40 mM imidazole). The desired enzymes were eluted with buffer D (buffer A containing 300 mM KCl and 100 mM imidazole). The purified enzymes were dialyzed against buffer E (20 mM Tris-Cl, pH 8.0, 40 mM KCl, 1 mM dithiothreitol, 5% glycerol) to remove the imidazole and then stored at 4 °C. The purity of enzymes was identified by SDS-polyacrylamide gel electrophoresis. The concentration of enzyme protein was determined by the Bradford method (25) using bovine serum albumin as standard.

Kinetic Analysis—Steady-state kinetic analysis was performed as described previously (21). Briefly, the 20-µl reaction mixture contained 50 mM Tris-glycine, pH 10, 11 mM MgSO₄, 1 mM dithiothreitol, 200 µg/ml bovine serum albumin, and enzyme with varying concentrations of ATP and [¹⁴C]choline. Reactions were carried out at 37 °C for 10 min, except as noted in Tables I, II, III, and stopped by adding 2 µl of acetic acid. Radiolabeled phosphocholine was separated from radiolabeled choline by paper chromatography in ethanol/ammonium hydroxide/isopropanol (6.5/3.5/2). The amount of labeled product was determined by scintillation counting. Enzyme assays for wild type and all mutants were linear with time and protein concentration. The kinetic parameters k_cat and K_m with respect to ATP and choline were determined from secondary plots by fitting the data to the Michaelis-Menten equation (24, 26).

(Eq. 1)

(Eq. 2)

In Equations 1 and 2, [Mg²⁺] is the concentration of calculated free Mg²⁺.

In Ca²⁺ inhibition experiments, the concentrations of ATP (10 mM), choline (10 mM), and Mg²⁺ (11 mM) were fixed, and the concentration of Ca²⁺ was varied from 1 to 20 mM. The IC₅₀ for Ca²⁺ was calculated by fitting to Equation 3.

(Eq. 3)

In Equation 3, v₀ is the initial velocity in the absence of Ca²⁺, and [I] is the concentration of Ca²⁺.

All data fitting was performed with Kaleidagraph software (Synergy).

Circular Dichroism (CD)—CD studies were carried out on an Aviv CD-spectropolarimeter at 25 °C using a cuvette with a 1-mm path length. Proteins (150 µg/ml) were in buffer F (10 mM Tris-Cl, 10 mM KCl, 1 mM EDTA, 1 mM dithiothreitol, 5% glycerol, pH 8.0). All the solutions were filtered through a 0.22-µm membrane prior to use. Spectra were recorded over 200-250 nm in 1-nm increments (1 s averaging time). All data were the averages of three blank-corrected spectra.

Unfolding Analysis—Protein (60 µg/ml) was incubated overnight at 4 °C in buffer F containing guanidinium chloride (GdmCl) at pH 8.0 at various concentrations. Circular dichroism was used to monitor the changes induced by GdmCl. Spectra over 215-230 nm were recorded with the same scanning parameters as indicated above. The ellipticity at 222 nm under different concentration of GdmCl was recorded for the unfolding transition profiles. Subsequently, all the data were normalized to relative fractions for the purpose of comparison.

(Eq. 4)

_N is the ellipticity of the protein in the native conformation (in the absence of GdmCl), _D is the ellipticity of the protein in the unfolded state (at 4 M GdmCl in our case), and is the ellipticity at a given concentration of GdmCl.

Gel Filtration—Analysis of protein size by gel filtration was performed by fast protein liquid chromatography on a Superpose-12 HR (Amersham Biosciences) column at 4 °C. Samples of 200 µl (100 µg/ml) were loaded onto the column pre-equilibrated with buffer G (20 mM Tris-Cl, pH 8.0, 150 mM KCl, 1 mM dithiothreitol, 5% glycerol) at a flow rate of 0.4 ml/min and monitored at 280 nm. The standard markers for molecular weight calibration were bovine thyroglobulin (670,000), gamma globin (158,000), bovine serum albumin (67,000), chicken ovalbumin (44,000), horse myoglobin (17,000), and vitamin B-12 (1350).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Similarity among Choline/Ethanolamine Kinases—To search for critical residues in CKA-2, we compared the primary structure of CKA-2 with that of other choline or ethanolamine kinases from various sources. The alignment in Fig. 1 revealed a large number of highly conserved residues, with two noticeable clusters. The first cluster, (C/S)HXDhX₃N (where h is a large hydrophobic residue) is similar to Brenner's phosphotransferase motif, hXHXDhX₃N, which is shared by many protein kinases and aminoglycoside phosphotransferases (28-30). The other cluster, which could be considered a signature motif for the choline/ethanolamine kinase family (8), is (LIV)X₂ID(FWY)E(YF)X₃NX₃(FWY)DX₆E. The conserved hydrophilic and aromatic residues were chosen for mutagenesis because the side chains of these residues are likely to be involved in catalysis and/or to make major contributions to the binding of substrates.

View larger version (120K): [in this window] [in a new window]   FIG. 1. Primary structural alignment of choline/ethanolamine kinases, APH(3')-IIIa and PKA. The alignment was obtained with the Clustal algorithm in the MegAlign program provided by the DNAstar software package. Residues shaded in black are identical or highly conserved. CKA-2 and CKB-2 are choline kinases from C. elegans (21), and EKC is an ethanolamine kinase from C. elegans (termed CKC-1 in Ref. 21). CK1 and CK are from mouse (8), CK-sp is choline kinase from Streptococcus pneumoniae (H. A. Campbell and C. Kent, unpublished results; GenBankTM accession no. AAK94073 [GenBank] , CK1-Gm and CK3-Gm are from soybean (Glycine max) (20), CKI1-sc (45) and EKI2-sc (46) are the choline and ethanolamine kinase from Saccharomyces cerevisiae, EKI1-hs is sequence of human (Homo sapiens) (47), and EK-Dm is ethanolamine kinase from fruit fly (Drosophila melanogaster) (48). The alignment with APH(3')-IIIa (49) and PKA (cAPK) (50) is based on their structure. The identical or conserved residues among APH(3')-IIIa and PKA are shaded in gray. The conserved Glu residues in PKA and APH(3')-IIIa in the open box are not consistent with Glu-125 in the structure of CKA-2. The regions marked with bars on their top are the nucleotide positioning loop, Brenner's phosphotransferase motif, and the choline kinase motif, respectively.

Kinetic Analysis of Wild Type and Mutant CKA-2—Steady-state kinetic measurements were performed to compare the catalytic properties of various mutants with those of the wild type enzyme. First, the properties of the His-tagged wild type enzyme were compared with those of native CKA-2. The kinetic parameters (Table I), pH optimum (pH 10), and dimeric structure as determined by gel filtration were nearly identical, indicating that the overall structure of CKA-2 was not altered appreciably by the His tag. All subsequent studies were performed on His-tagged mutant and wild type forms. For presentation of the results, mutants generated in this study were sorted into three groups based on their positions in the primary structure. The groups were the NH₂-terminal region, the phosphotransferase motif, and the COOH-terminal region which includes the choline kinase motif.

NH₂-terminal Region (Table I)—A Ser or Thr side chain at residue 86 is conserved among choline/ethanolamine kinases. Elimination of this hydroxyl by mutation to Ala had an effect on k_cat (7-fold reduction) and K_m for choline (5-fold increase), which led to a pronounced decrease in catalytic efficiency (k_cat/K_m) for both ATP (9.0% of wild type) and choline (2.7% of wild type). Catalytic efficiencies were more similar to wild type when Ser-86 was replaced by Thr, suggesting that a hydroxyl at position 86 is essential for full activity. A conserved Asn residue neighbors the conserved Ser/Thr; however, modification of Asn-87 to Ala was much less detrimental than mutation of Ser-86 to Ala.

The substitution of Arg-111 with Ala caused a modest reduction in k_cat and a 4-fold increase in K_m for ATP, resulting in a 10-fold reduction in k_cat/K_m for ATP. The replacement of Glu-125 with Ala or Gln resulted in modest effects on both k_cat and K_m for ATP. However, substitution of Glu-125 with Asp was not deleterious to activity. A positive charge at position 149 is conserved only in the choline/ethanolamine kinases; mutation of Arg-149 to Ala did not markedly affect the kinetic parameters of the enzyme.

Phosphotransferase Motif (Table II)—This motif is conserved in many kinases, including the eukaryotic protein kinase superfamily (30), in which the equivalent of Asp-255 and Asn-260 are absolutely conserved. Mutation of these two residues in CKA-2 had very dramatic effects. Enzymatic activity was not detectable in either D255A or D255N. Conserving the negative charge by mutation of Asp-255 to Glu resulted in low but detectable activity, with a reduction in k_cat of approximately 300-fold and an increase in K_m for choline by 5-fold. Mutation of Asn-260 to Ala resulted in a sharp drop in activity, with a 300-fold and reduction in k_cat consistent with the highly conserved nature of this residue.

His-253 is conserved in choline/ethanolamine kinases but not necessarily in kinases of this superfamily. Mutation of His-253 to Ala did not result in appreciable loss in initial activity, but the activity was linear with time only for a little over 30 s, whereas the wild type activity is linear for over 30 min. This instability with respect to activity suggested that His-253 has a structural role in the active site. N254A shows a similar instability, with activity linear only for 5 min. In addition, the k_cat for N254A was reduced by approximately 30-fold.

COOH-terminal Region (Table III)—Although the choline kinase motif is not shared among protein kinases and aminoglycoside phosphotransferases, residue Asp-301 of this motif does appear to be conserved among these other kinases structurally (see "Discussion"). Indeed, mutation of Asp-301 to Ala, Asn, and Glu resulted in a lack of any detectable enzymatic activity, even with 500 times more mutant protein in the assay mixture. This suggests that both the negative charge of Asp-301 and its precise position are critical for the activity of CKA-2. Glu-303 is conserved throughout the choline/ethanolamine kinases but not the other kinases. The Ala substitution for Glu-303 had a marked effect on both k_cat (over 10-fold reduction) and K_m values for both ATP and choline (3- and 10-fold increase, respectively), resulting in a sharp decrease in the catalytic efficiency. The k_cat value was restored to nearly normal by mutation of Glu-303 to either Asp or Gln, although K_m values were still abnormally high. Thus, the carbonyl function of this side chain appears to be necessary for optimal catalysis, whereas the precise position and negative charge affect the K_m.

Mutation of Asn-308 to Ala resulted in a modest decrease in k_cat, but mutation to Asp and Gln resulted in catalytic efficiencies very similar to wild type. Substitution of Glu-320 with Ala resulted in an increased K_m for ATP. The E320A mutant was also somewhat unstable in that its activity was linear for only approximately 5 min. This residue may have both catalytic and structural roles.

Trp-387 is an absolutely conserved aromatic residue among choline/ethanolamine kinases. The k_cat value for the W387A mutant was decreased 30-fold, with no change in K_m values, consistent with an important role for this residue in catalysis.

Structural Analysis of Wild Type and Mutant CKA-2—Mutants with pronounced catalytic defects (S86A, R111A, D255A(E), N260A, D301A, E303A, and W387A) were chosen for structural analysis to determine whether the catalytic defects were caused by extensive structural defects. The far-UV circular dichroic spectra of all these mutants were similar to that of the wild type protein (data not shown), indicating the mutations did not cause profound changes in the secondary structure under native conditions. Analysis of the mutant proteins by gel filtration was used to compare their quaternary structures. All mutated proteins had the same retention volume as the wild type enzyme, corresponding to a molecular weight of 102,000 under native conditions (data not shown). Additionally, with the exception of D255A and D301A, the enzymatic activity of the mutants under assay conditions was linear at 37 °C for at least 30 min, indicating that the thermal stability of these mutants under assay conditions is similar to that of the wild type.

To further assess the conformational stability of each mutant, the sensitivity of the wild type and mutants to guanidinium-induced unfolding was determined by circular dichroism (Fig. 2). The wild type enzyme displayed a biphasic denaturation curve, with some unfolding occurring between 0.2 and 1 M, then further unfolding between 2 and 4 M. Most of the mutants exhibited similar biphasic unfolding curves, suggesting that these mutants are as stable, under these conditions, as the wild type. The first unfolding phase for W387A and D255E occurred at much lower guanidinium concentrations than for the other proteins, suggesting a lower degree of stability for these two mutants. Thus, under native condition, all mutant enzymes retained a high degree of structural integrity, implying that any substantial differences in catalytic efficiency or other kinetic defects were not simply the result of large structural alterations. However, two mutations, W387A and D255E, rendered the enzyme more sensitive to conditions that promote unfolding of the protein.

View larger version (14K): [in this window] [in a new window]   FIG. 2. The unfolding curves of wild type and mutant CKA-2 induced by GdmCl monitored by circular dichroism. The changes of []222 at various concentration of GdmCl were recorded for wild type (), S86A (), R111A (), E125A (), D255A (), D255E (), N260A (), D301A (), E303A (), and W387A ().

View larger version (20K): [in this window] [in a new window]   FIG. 3. The Mg2+-dependent activity of wild type and mutant CKA-2. Initial rates were determined at various free Mg2+ concentrations for wild type (), S86A (, R111A (), E125A (), D255E (), N260A (), E303A (), and W387A (). For the purpose of clarity, the specific activities of W387A, N260A and S86A were multiplied by 10, and that for D255A was multiplied by 100.

View larger version (23K): [in this window] [in a new window]   FIG. 4. Inhibition of wild type and mutant CKA-2 by Ca2+ The specific activities were determined at various concentrations of Ca2+ for wild type (), S86A (), R111A (), E125A (), D255E (), N260A (), E303A (), and W387A (). For the purpose of clarity, the specific activity obtained for wild type, S86A (multiplied by 10), R111A (multiplied by 5), D255E (multiplied by 500), and W387A (multiplied by 10) were plotted using the left y-axis, whereas the specific activity obtained for E125A, N260A (multiplied by 100), and E303A (multiplied by 10) were plotted using the right y-axis.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Phosphotransferase Motif—Asp-255 in CKA-2 is an absolutely conserved component of Brenner's phosphotransferase motif, found in protein kinases, choline/ethanolamine kinases, and aminoglycoside phosphotransferases. That this Asp is a critical residue has been demonstrated in several studies, although investigators have not agreed upon a universal role of this residue. Studies with several protein kinases have implicated this Asp residue as an active site base necessary for deprotonation of the incoming substrate hydroxyl group (31). However, evidence has also been presented that this Asp, rather that acting as a base, is involved in proper orientation of the nucleophilic hydroxyl of substrates, enhancing the electrophilicity of the -phosphate of ATP through direct or indirect hydrogen bonding, or maintaining a proper conformation of the active site (32).

In CKA-2, mutagenesis of Asp-255 to both Ala and Asn results in complete loss of activity even when a 500-fold excess of mutant protein was assayed. The replacement of Asp-255 by Glu afforded some activity, but the k_cat was only 0.1% of the wild type level. Thus, a negative charge at position 255 plays a key role in activity, and the precise position of this charge is necessary for optimal activity. In contrast to its corresponding mutant D190E in APH(3')-IIIa (27), D255E in CKA-2 mutant did not show an alteration in Mg²⁺ dependence or Ca²⁺ inhibition relative to wild type.

Asn-260 is the other highly conserved residue in Brenner's phosphotransferase motif. The amide oxygen of the corresponding Asn in APH(3')-IIIa and ePKs interacts with Mg²⁺ bound to ATP (27, 30). Mutation of Asn-260 in CKA-2 results in a severely impaired enzyme, with k_cat only 0.3% of wild type. This mutation also results in a 36-fold increase in apparent K_m for Mg²⁺, and a slight increase in K_m for ATP, consistent with its role established with other kinases. Inhibition of N260A by Mg²⁺ is virtually nonexistent, which agrees with the analysis of mutant N195A in APH(3')-IIIa, implicating this Asn in the phenomenon of inhibition by high concentrations of Mg²⁺. However, the effect of the N195A mutation in APH(3')-IIIa on k_cat was much less severe than N260A in CKA-2 (27), indicating that Asn-260 may have additional role(s).

Choline Kinase Motif—The sequence (LIV)X₂ID(FWY)E(YF)X₃NX₃(FWY)DX₆E is faithfully conserved in choline/ethanolamine kinases. The first Asp of that sequence appears to assume the role of a conserved Asp found in many kinases: in the DFG triplet of protein kinases (30), and a similar DLG sequence in APH(3')-IIIa (Fig. 1). This Asp coordinates two Mg²⁺ ions via its carboxyl oxygen in binding of the ATP-Mg complex (33, 34). With APH(3')-IIIa the charge neutralization of Mg²⁺ by this Asp is important for generating and stabilizing the transition state, because the substitution of Asp by Asn, a potential Mg²⁺ ligand, lacks any detectable activity (27). Substitution of Asp-301 in CKA-2 with Ala, Asn, or Glu completely abolished choline kinase activity. These results are in agreement with those obtained with APH(3')-IIIa, and indicate that the precise location of the negative carboxylate of Asp-301 is critical for catalysis.

Glu-303 of the choline kinase motif does not have apparent counterparts in ePKs and APH(3')-IIIa. Mutation of Glu-303 with Ala compromised catalytic activity with a 10-fold depression in k_cat and increased K_m for both ATP (3-fold) and choline (10-fold). E303A also exhibited altered sensitivity to both Mg²⁺ and Ca²⁺; high concentrations of Mg²⁺ no longer inhibited, and low concentrations of Ca²⁺ increased activity. Substitutions of Glu-303 with Asp or Gln were not as costly as the Ala mutation with respect to k_cat, but K_m values for ATP and choline remained high. Thus, both the negative charge and precise position of the side chain of Glu-303 are important for full activity. It may be that perturbations in Glu-303 negatively influence the side chain of Glu-301, resulting in kinetic disturbances. Glu-320 appears to have a complex role in catalysis by CKA-2. Mutation of this residue to Ala results in a 26-fold higher K_m for ATP with no significant change in k_cat, suggesting that Glu-320 might participate in ATP binding. However, E320A is active only for a short time relative to wild type, suggesting that this residue may also participate in the structural integrity of the enzyme. Asn-308 appears to be less critical for activity than other highly conserved residues, because mutation to Ala caused only a 3-fold decrease in k_cat, and substitution with Asp or Gln were even less detrimental.

Other Functional Residues—In addition to the conserved motifs described above, CKA-2 has a loop structure located in the proposed ATP binding pocket (Fig. 5), which is structurally equivalent to the ATP binding loop or nucleotide position loop (NPL) in APH(3')-IIIa and other kinases (35, 36). A comparison among these NPL structures reveals that the NPL varies widely in length, secondary structure, conformation, and hydrogen bonding with ATP; therefore, the judgment as to whether an NH₂-terminal region is truly an NPL must be based on a three-dimensional structure (37). NPL structures do have in common their high mobility, and their conformations undergo rearrangement upon binding of ATP. The loop structure of CKA-2 is somewhat similar to that of APH(3')-IIIa in terms of both primary and tertiary structure (22, 38). First, they both have a GMS sequence instead of the classical GXGXXG sequence in most ePKs (Fig. 1). Second, they share a Ser/Thr positioned in the middle of the loop, rather than in the apex of the loop as in most ePKs. Finally, these loops are in the NH₂-terminal domain between the first and second -strands of the antiparallel -sheet. On the basis of these similarities, we postulate that this loop in CKA-2, like its analogous region in APH(3')-IIIa, plays an important part in binding ATP and facilitating phosphoryl transfer, with Ser-86 as an important participant. Steady-state kinetic analysis of the S86A mutant revealed an 11- and 36-fold decrease in k_cat/K_m for ATP and choline, respectively. This mutant also displayed modest alterations in both K_m and K_i for Mg²⁺, which is also observed with its corresponding mutant S27A in APH(3')-IIIa (38). These results support the proposal for a role of Ser-86 in an NPL in CKA-2.

View larger version (45K): [in this window] [in a new window]   FIG. 5. The proposed ATP binding site in CKA-2. A, a close-up view of the ATP binding site in APH(3')-IIIa (Protein Data Bank code 1J7U [PDB] ) and B, the proposed site in CKA-2 (Protein Data Bank code 1NW1 [PDB] ). The backbone of both enzymes is shown as brown ribbon drawings with the exception of the ATP binding loop, which is in cyan. The side chains of conserved residues are yellow ball-and-stick models, ATP is a magenta ball-and-stick model, and the Mg2+ ions are green spheres. Please note that the ATP molecule in B is modeled in; there is no ATP in the actual structure.

Trp-387 is conserved among members of the choline/ethanolamine kinase family, but not the other kinases, suggesting that this residue might play a specific role in binding choline or ethanolamine. In fact, binding sites for choline often contain aromatic residues. The choline binding site of autolysin (lytA), a choline-activated enzyme, comprises three aromatic residues, including at least one tryptophan (42). Structural studies on lytA and the M3C65 antibody, which binds phosphatidylcholine (43), indicate that both hydrophobic and electrostatic forces are responsible for interaction between the protein and choline moiety. The hydrophobic portion of the binding site is mainly constructed of aromatic residues, which also provide the electrostatic component as a cation- interaction between the positively charged nitrogen of the choline and the indole ring of tryptophan. The crystal structure of CKA-2 demonstrates that Trp-387 is located in a pocket that is not found in the active sites of ePKs and APH(3')-IIIa (Fig. 6) (22). Trp-387 lies close to the binding site of the -phosphate of ATP; two other aromatic residues, Trp-390 and Tyr-304, are close to the Trp-387 in this pocket structure. When Trp-387 of CKA-2 was replaced by Ala, k_cat decreased by almost 30-fold, supporting the role of this residue in rate enhancement. However, K_m values for both substrates in W387A were unchanged, suggesting that choline binding is not rate-limiting for the catalytic step that is modified. Further evidence that the role of Trp-387 may be complex, not limited to the substrate binding and catalysis was provided by the GdmCl-induced denaturation studies. The dependence on low concentrations of GdmCl for W387A was quite different from that of wild type, suggesting that Trp-387 is also involved in maintaining the overall stability of the enzyme.

View larger version (75K): [in this window] [in a new window]   FIG. 6. The proposed ATP and choline binding site in CKA-2. The enzyme is shown as indicated in Fig. 5B, plus three aromatic residues as in green ball-and-stick representation.

Three mutants, H253A, N254A, and E320A, exhibited instability in the enzyme assay, with activity linear for only a few seconds. Two other mutants, H181A and D313A, were expressed at low levels, and we were unable to purify them. The crystal structure shows that four of these residues are involved in mutual interactions. The side chain of Asp-313 hydrogenbonds with the main chain of His-253 and Asn-254, whereas His-181 interacts with the carboxylate group of Asp-313 (22). This evidence suggests that these residues have structural roles in maintaining the proper conformation of the enzyme.

Concluding Remarks—In summary, we have defined some residues in CKA-2 that are critical for catalysis, and suggested that these critical residues all make up the proposed catalytic core of CKA-2, which structurally lies in the cleft between NH₂-terminal and COOH-terminal domains. ATP is proposed to bind to CKA-2 in a pocket lined with Ser-86, Arg-111, Asp-255, Asn-260, and Asp-301; in particular, on the one side of the ATP binding pocket, Arg-111 and Ser-86 are responsible for binding to phosphate of ATP; on the other side, Asp-260 and Asp-301 are involved in binding to the ATP-Mg complex via coordinating to the Mg²⁺ ion. Moreover, the choline binding site is proposed to locate near the ATP-binding site and involve several aromatic residues, including Trp-387. CKA-2, the first choline kinase to be crystallized and studied in detail by mutagenesis, will serve as a paradigm for the entire choline kinase families.

The high degree of similarity in the three-dimensional structures of ePKs, aminoglycoside phosphotransferases, and CKA-2 suggests that these three families of enzymes evolved from a common ancestor. Several highly conserved residues apparently line the catalytic cores of these enzymes and have functional conservation, such as nucleotide binding and metal binding. Analysis of the relationship among these disparate groups of proteins will further enhance our understanding of the common features of catalysis of various kinases that use widely different substrates, ranging from peptides to small molecules. On the other hand, mutagenesis studies reveal that some of the highly conserved residues among these kinases have different features, suggesting that, despite the similarity of the frame of the active site core, several residues in the core have evolved to accommodate the different functions for each individual enzyme family. Thus, an immediate goal is to obtain the x-ray structure of CKA-2 in complex with different substrates or inhibitors, which will aid immensely in understanding these differences.

	FOOTNOTES

 
^* This work was supported by National Institutes of Health Grant GM60510. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed. Present address: 1497 Vista Claridad, La Jolla, CA 92037. Tel.: 858-456-0717; E-mail: ckent{at}umich.edu.

¹ The abbreviations used are: CKA-2, choline kinase A-2; APH(3')IIIa, aminoglycoside-3'-phosphotransferase type IIIa; PKA, cAMP-dependent protein kinase; GdmCl, guanidinium hydrochloride; NPL, nucleotide positioning loop; ePK, eukaryotic protein kinase.

	ACKNOWLEDGMENTS

 
We thank Dr. Gerard D. Wright (McMaster University, Hamilton, Ontario, Canada) for providing software; Dr. Zhaohui Xu for providing fast protein liquid chromatography; and Dr. Heidi A. Campbell, Dr. Joel M. Clement, Dr. John Stansberry, Dr. Daniel Peisach, and Patricia Gee for helpful discussions. We also thank Dr. Daniel Peisach for preparing Figs. 5 and 6.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 

1. Kent, C., and Carman, G. M. (1999) Trends Biochem. Sci. 24, 146-150[CrossRef][Medline] [Order article via Infotrieve]
2. Exton, J. H. (1994) Biochim. Biophys. Acta 1212, 26-42[Medline] [Order article via Infotrieve]
3. Kent, C. (1997) Biochim. Biophys. Acta 1348, 79-90[Medline] [Order article via Infotrieve]
4. Carman, G. M., and Zeimetz, G. M. (1996) J. Biol. Chem. 271, 13293-13296[Free Full Text]
5. DeLong, C. J., Shen, Y. J., Thomas, M. J., and Cui, Z. (1999) J. Biol. Chem. 274, 29683-29688[Abstract/Free Full Text]
6. Wittenberg, J., and Kornberg, A. (1953) J. Biol. Chem. 202, 431-444[Free Full Text]
7. Porter, T. J., and Kent, C. (1990) J. Biol. Chem. 265, 414-422[Abstract/Free Full Text]
8. Aoyama, C., Yamazaki, N., Terada, H., and Ishidate, K. (2000) J. Lipid Res. 41, 452-464[Abstract/Free Full Text]
9. Kim, K. H., Voelker, D. R., Flocco, M. T., and Carman, G. M. (1998) J. Biol. Chem. 273, 6844-6852[Abstract/Free Full Text]
10. Aoyama, C., Nakashima, K., Matsui, M., and Ishidate, K. (1998) Biochim. Biophys. Acta 1390, 1-7[Medline] [Order article via Infotrieve]
11. Ishidate, K. (1997) Biochim. Biophys. Acta 1348, 70-78[Medline] [Order article via Infotrieve]
12. Kent, C. (1990) Prog. Lipid Res. 29, 87-105[CrossRef][Medline] [Order article via Infotrieve]
13. Cuadrado, A., Carnero, A., Dolfi, F., Jimenez, B., and Lacal, J. C. (1993) Oncogene 8, 2959-2968[Medline] [Order article via Infotrieve]
14. Ramirez de Molina, A., Penalva, V., Lucas, L., and Lacal, J. C. (2002) Oncogene 21, 937-946[CrossRef][Medline] [Order article via Infotrieve]
15. Ratnam, S., and Kent, C. (1995) Arch. Biochem. Biophys. 323, 313-322[CrossRef][Medline] [Order article via Infotrieve]
16. Teegarden, D., Taparowsky, E. J., and Kent, C. (1990) J. Biol. Chem. 265, 6042-6047[Abstract/Free Full Text]
17. Ramirez de Molina, A., Rodriguez-Gonzalez, A., Penalva, V., Lucas, L., and Lacal, J. C. (2001) Biochem. Biophys. Res. Commun. 285, 873-879[CrossRef][Medline] [Order article via Infotrieve]
18. Kim, K. H., and Carman, G. M. (1999) J. Biol. Chem. 274, 9531-9538[Abstract/Free Full Text]
19. Yu, Y., Sreenivas, A., Ostrander, D. B., and Carman, G. M. (2002) J. Biol. Chem. 277, 34978-34986[Abstract/Free Full Text]
20. Monks, D. E., Goode, J. H., and Dewey, R. E. (1996) Plant Physiol. 110, 1197-1205[Abstract]
21. Gee, P., and Kent, C. (2003) Biochim. Biophys. Acta 1648, 33-42[Medline] [Order article via Infotrieve]
22. Peisach, D., Gee, P., Kent, C., and Xu, Z. (2003) Structure (Camb.) 11, 703-713
23. Hon, W. C., McKay, G. A., Thompson, P. R., Sweet, R. M., Yang, D. S., Wright, G. D., and Berghuis, A. M. (1997) Cell 89, 887-895[CrossRef][Medline] [Order article via Infotrieve]
24. Horton, R. M., Hunt, H. D., Ho, S. N., Pullen, J. K., and Pease, L. R. (1989) Gene (Amst.) 77, 61-68[CrossRef][Medline] [Order article via Infotrieve]
25. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254[CrossRef][Medline] [Order article via Infotrieve]
26. Park, Y. S., Gee, P., Sanker, S., Schurter, E. J., Zuiderweg, E. R., and Kent, C. (1997) J. Biol. Chem. 272, 15161-15166[Abstract/Free Full Text]
27. Boehr, D. D., Thompson, P. R., and Wright, G. D. (2001) J. Biol. Chem. 276, 23929-23936[Abstract/Free Full Text]
28. Brenner, S. (1987) Nature 329, 21[Medline] [Order article via Infotrieve]
29. Bairoch, A., and Claverie, J. M. (1988) Nature 331, 22[CrossRef][Medline] [Order article via Infotrieve]
30. Hanks, S. K., and Hunter, T. (1995) FASEB J. 9, 576-596[Abstract]
31. Zheng, J., Knighton, D. R., ten Eyck, L. F., Karlsson, R., Xuong, N., Taylor, S. S., and Sowadski, J. M. (1993) Biochemistry 32, 2154-2161[CrossRef][Medline] [Order article via Infotrieve]
32. Hutter, M. C., and Helms, V. (1999) Protein Sci. 8, 2728-2733[Abstract]
33. van der Wolk, J. P., Klose, M., de Wit, J. G., den Blaauwen, T., Freudl, R., and Driessen, A. J. (1995) J. Biol. Chem. 270, 18975-18982[Abstract/Free Full Text]
34. Knighton, D. R., Zheng, J. H., Ten Eyck, L. F., Ashford, V. A., Xuong, N. H., Taylor, S. S., and Sowadski, J. M. (1991) Science 253, 407-414[Abstract/Free Full Text]
35. Runquist, J. A., Rios, S. E., Vinarov, D. A., and Miziorko, H. M. (2001) Biochemistry 40, 14530-14537[CrossRef][Medline] [Order article via Infotrieve]
36. Deyrup, A. T., Krishnan, S., Cockburn, B. N., and Schwartz, N. B. (1998) J. Biol. Chem. 273, 9450-9456[Abstract/Free Full Text]
37. Bilwes, A. M., Quezada, C. M., Croal, L. R., Crane, B. R., and Simon, M. I. (2001) Nat. Struct. Biol. 8, 353-360[CrossRef][Medline] [Order article via Infotrieve]
38. Thompson, P. R., Boehr, D. D., Berghuis, A. M., and Wright, G. D. (2002) Biochemistry 41, 7001-7007[CrossRef][Medline] [Order article via Infotrieve]
39. McKay, G. A., Robinson, R. A., Lane, W. S., and Wright, G. D. (1994) Biochemistry 33, 14115-14120[CrossRef][Medline] [Order article via Infotrieve]
40. Hishida, T., Iwasaki, H., Yagi, T., and Shinagawa, H. (1999) J. Biol. Chem. 274, 25335-25342[Abstract/Free Full Text]
41. Spitaler, M., Villunger, A., Grunicke, H., and Uberall, F. (2000) J. Biol. Chem. 275, 33289-33296[Abstract/Free Full Text]
42. Fernandez-Tornero, C., Garcia, E., Lopez, R., Gimenez-Gallego, G., and Romero, A. (2002) J. Mol. Biol. 321, 163-173[CrossRef][Medline] [Order article via Infotrieve]
43. Brown, M., Schumacher, M. A., Wiens, G. D., Brennan, R. G., and Rittenberg, M. B. (2000) J. Exp. Med. 191, 2101-2112[Abstract/Free Full Text]
44. Sicheri, F., Moarefi, I., and Kuriyan, J. (1997) Nature 385, 602-609[CrossRef][Medline] [Order article via Infotrieve]
45. Hosaka, K., Kodaki, T., and Yamashita, S. (1989) J. Biol. Chem. 264, 2053-2059[Abstract/Free Full Text]
46. Kim, K., Kim, K. H., Storey, M. K., Voelker, D. R., and Carman, G. M. (1999) J. Biol. Chem. 274, 14857-14866[Abstract/Free Full Text]
47. Lykidis, A., Wang, J., Karim, M. A., and Jackowski, S. (2001) J. Biol. Chem. 276, 2174-2179[Abstract/Free Full Text]
48. Pavlidis, P., Ramaswami, M., and Tanouye, M. A. (1994) Cell 79, 23-33[CrossRef][Medline] [Order article via Infotrieve]
49. Burk, D. L., Hon, W. C., Leung, A. K., and Berghuis, A. M. (2001) Biochemistry 40, 8756-8764[CrossRef][Medline] [Order article via Infotrieve]
50. Madhusudan, Trafny, E. A., Xuong, N. H., Adams, J. A., Ten Eyck, L. F., Taylor, S. S., and Sowadski, J. M. (1994) Protein Sci. 3, 176-187[Abstract]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

M.-G. Choi, V. Kurnov, M. C. Kersti