A novel candidate for the true fructose-1,6-bisphosphatase in archaea.

Fructose-1,6-bisphosphatase (FBPase) is one of the key enzymes of the gluconeogenic pathway. Although enzyme activity had been detected in Archaea, the corresponding gene had not been identified until a presumable inositol monophosphatase gene from Methanococcus jannaschii was found to encode a protein with both inositol monophosphatase and FBPase activities. Here we display that a gene from the hyperthermophilic archaeon, Thermococcus kodakaraensis KOD1, which does not correspond to the inositol monophosphatase gene from M. jannaschii, displays high FBPase activity. The FBPase from strain KOD1 was partially purified, its N-terminal amino acid sequence was determined, and the gene (Tk-fbp) was cloned. Tk-fbp encoded a protein of 375 amino acid residues with a molecular mass of 41,658 Da. The recombinant Tk-Fbp was purified and characterized. Tk-Fbp catalyzed the conversion of fructose 1,6-bisphosphate to fructose 6-phosphate following Michaelis-Menten kinetics with a K(m) value of 100 microm toward fructose 1,6-bisphosphate, and a k(cat) value of 17 s(-1) subunit(-1) at 95 degrees C. Unlike the inositol monophosphatase from M. jannaschii, Tk-Fbp displayed strict substrate specificity for fructose 1,6-bisphosphate. Activity was enhanced by Mg(2+) and dithioerythritol, and was slightly inhibited by fructose 2,6-bisphosphate. AMP did not inhibit the enzyme activity. We examined whether expression of Tk-fbp was regulated at the transcription level. High levels of Tk-fbp transcripts were detected in cells grown on pyruvate or amino acids, whereas no transcription was detected when starch was present in the medium. Orthologue genes corresponding to Tk-fbp with high similarity are present in all the complete genome sequences of thermophilic Archaea, including M. jannaschii, Pyrococcus furiosus, Sulfolobus solfataricus, and Archaeoglobus fulgidus, but are yet to be assigned any function. Taking into account the high FBPase activity of the protein, the strict substrate specificity, and its sugar-repressed gene expression, we propose that Tk-Fbp may represent the bona fide FBPase in Archaea.

FBPases have been characterized from bacteria (3)(4)(5), yeast (6,7), as well as higher eukaryotes (8 -11). The regulation of FBPase gene expression has been extensively studied in yeast, and is regarded as a typical example of glucose repression/ derepression (12). In Saccharomyces cerevisiae, gene transcription is repressed by the MIG1 repressor in the presence of glucose, and is derepressed in the absence of glucose via the SNF1/CAT8 regulation pathway (13). Besides the regulation at the transcriptional level, FBPase is also known to be an allosteric enzyme (14). FBPases from bacteria, yeast, and mammals have been reported to be inhibited allosterically by AMP (15). Addition of AMP causes the enzyme to cooperatively shift from the fully active R-state (relaxed) to the inactive T-state (tense) (16). Fructose 2,6-bisphosphate has also been found to inhibit FBPase activity as a substrate analog (17).
The amino acid sequences of mammalian enzymes are 85% identical to each other, and are similar to the Class I FBPases in bacteria (3). In Escherichia coli, two FBPases have been identified. The Class I enzyme encoded by the fbp gene (Ec-Fbp) has been long recognized as the sole FBPase in E. coli (18). However, a second FBPase encoded by the glpX gene (Ec-GlpX) has recently been identified (4). Ec-GlpX does not display structural similarity with Ec-Fbp, and has been classified as a Class II enzyme. A very divergent Class III FBPase has also been identified in Bacillus subtilis (5). In E. coli, disruption of the fbp gene led to a mutant that grew as well as the wild type strain on glucose or fructose, but could not grow on glycerol or other gluconeogenic substrates. In contrast, the glpX disruptant strain did not display a particular phenotype. In the case of E. coli, the Class I Ec-Fbp is presumed to be the major enzyme involved in gluconeogenesis (4).
In Archaea, the identification of FBPase has attracted much attention. This is because of the fact that although FBPase activity had been detected in cell extracts of several Archaea (19 -22), orthologous genes with structural similarity to previously reported FBPases are not present on their genomes. This was resolved to some extent by the finding that the MJ0109 * 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.
The nucleotide sequence for the Tk-fbp gene reported in this paper has been submitted to the DDBJ/GenBank TM  gene product from the archaeon Methanococcus jannaschii displayed FBPase activity (21). The gene had been assigned as an inositol monophosphatase gene, and the protein product was found to harbor both inositol monophosphatase and FBPase activities. No other protein with FBPase activity has been reported from Archaea.
Thermococcus kodakaraensis KOD1 is a hyperthermophilic archaeon isolated from Kodakara Island, Kagoshima, Japan (23). The strain can grow with amino acids as a carbon and energy source and sulfur as the terminal electron acceptor. The cells can also assimilate starch or pyruvate, providing a good tool for studying gluconeogenesis and its regulation in Archaea. Here we report the FBPase from T. kodakaraensis KOD1, a structurally distinct enzyme from previously identified FB-Pases, including the MJ0109 gene product from M. jannaschii.

EXPERIMENTAL PROCEDURES
Bacterial Strains, Plasmids, and Bacteriophages-T. kodakaraensis KOD1 was isolated from a solfataric hot spring at a wharf in Kodakara Island, Kagoshima, Japan (23). E. coli strain DH5␣ was used for subcloning of the gene fragments and DNA manipulations. E. coli strain BL21(DE3) (Novagen, Madison, WI) was used as a host and pET-8c vector (Novagen) was used for gene expression.
DNA Manipulation-Restriction enzymes and DNA polymerase were purchased from Toyobo (Osaka, Japan) and Takara Shuzo (Kyoto, Japan). Genomic, plasmid and phage DNAs were isolated using Qiagen genomic, plasmid, and phage DNA isolation kits, respectively (Qiagen, Hilden, Germany). DNA ligations were performed using the DNA ligation kit (Toyobo). The QIAEX gel extraction kit (Qiagen) was used to recover DNA fragments from agarose gels.
Partial Purification of FBPase from KOD1 Cells-T. kodakaraensis KOD1 cells were cultivated in a medium containing pyruvate as a carbon source. After overnight cultivation, cells were harvested, resuspended in 50 mM potassium phosphate buffer (pH 7.0), and disrupted by sonication in ice water. All purification steps were performed at room temperature unless mentioned otherwise. Membrane and cytosolic fractions from the cell lysate were separated by ultracentrifugation at 110,000 ϫ g for 70 min at 4°C. The cytosolic fraction exhibiting the FBPase activity was loaded on an anion exchange column (RESOURCE Q, Amersham Biosciences AB, Uppsala, Sweden) equilibrated with 50 mM potassium phosphate buffer (pH 7.0). The column was washed with 3 bed volumes of 50 mM potassium phosphate buffer (pH 7.0), and the bound proteins were eluted with a linear gradient of 0 -1000 mM potassium chloride (pH 7.0). Fractions with FBPase activity were pooled and dialyzed against 50 mM potassium phosphate buffer (pH 7.0). The dialyzed sample was loaded onto another anion exchange column (Mono Q HR 5/5, Amersham Biosciences) equilibrated with 50 mM potassium phosphate buffer (pH 7.0). The column was washed with 5 bed volumes of equilibrated buffer, and the bound proteins were eluted with a linear gradient of 0 -1000 mM potassium chloride (pH 7.0). Fractions displaying FBPase activity were dialyzed against 2 M ammonium sulfate. The dialyzed sample was applied onto a hydrophobic column (RESOURCE ISO, Amersham Biosciences) that was equilibrated with 2 M ammonium sulfate in 20 mM potassium phosphate buffer (pH 7.0). The column was washed with 4 bed volumes of 2 M ammonium sulfate in 20 mM potassium phosphate buffer (pH 7.0), and the bound proteins were eluted with a linear gradient of 2-0 M ammonium sulfate (pH 7.0). Active fractions were pooled and dialyzed against 50 mM sodium phosphate buffer containing 150 mM sodium chloride (pH 7.0). The dialyzed sample was concentrated using Centricon YM-30 (Millipore, Bedford, MA). The concentrated sample was further purified by gel filtration (Superdex 200 HR 10/30, Amersham Biosciences) equilibrated with 50 mM sodium phosphate buffer containing 150 mM sodium chloride (pH 7.0). Protein concentration was determined with a bicinchoninic acid (BCA) protein assay kit (Pierce) according to the manufacturer's instructions using bovine serum albumin as a standard.
Determination of N-terminal Amino Acid Sequences-To determine the N-terminal amino acid sequence, proteins after SDS-PAGE were electroblotted onto a polyvinylidene difluoride membrane (Millipore). The membrane was stained with Amido Black solution (5% Amido Black, 40% methanol, 10% acetic acid) for 5 min. The membrane was de-stained and the N-terminal amino acid residues were determined with a protein sequencer (Model 491 cLC, PerkinElmer Life Sciences).
Cloning and Characterization of the Tk-fbp Gene-Probes were constructed using the digoxigenin DNA labeling kit (Roche Molecular Biochemicals). DNA sequencing was performed using ABI PRISM Big-Dye TM Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems). Nucleotide and the deduced amino acid sequence analyses and data base homology search were performed using the Basic Local Alignment Search Tool (BLAST) program. Open reading frame search and molecular mass calculations were performed using DNASIS software (Hitachi Software, Yokohama, Japan). Multiple alignment and phylogenetic analysis was performed using the Clustal W program provided by the DNA Data Bank of Japan (DDBJ).
Expression of Tk-fbp Gene in E. coli-The Tk-fbp gene was amplified by PCR and a BamHI site was introduced in the 3Ј-flanking region of the gene. The DNA fragment was inserted into the pET-8c expression vector (Novagen) at NcoI and BamHI sites and designated pET-fbp. E. coli strain BL21(DE3) cells carrying pET-fbp were grown overnight at 37°C in LB medium (1% tryptone, 0.5% yeast extract, 0.5% NaCl, pH7.0) containing ampicillin (50 g/ml). The culture was inoculated (1%) into fresh LB medium containing ampicillin and the cultivation was continued until A 660 reached 0.5. The culture was then supplemented with 1 mM (final concentration) isopropyl-1-thio-␤-D-galactopyranoside and incubated for another 4 h at 37°C.
Purification of Recombinant Tk-Fbp-Cells were harvested by centrifugation at 6,000 ϫ g for 10 min at 4°C and washed with 50 mM potassium phosphate buffer (pH 7.0). The cell pellet was resuspended in the same buffer and the cells were then disrupted by sonication in ice water. Soluble and insoluble fractions were separated by centrifugation (15,000 ϫ g for 30 min at 4°C). The soluble fraction containing the recombinant Tk-Fbp was incubated at 85°C for 20 min and centrifuged at 15,000 ϫ g for 30 min at 4°C to remove heat-labile proteins from the host E. coli. The supernatant carrying Tk-Fbp was purified to homogeneity with the same methodology described above for the native Tk-Fbp. The purity of the protein was examined by SDS-PAGE. Apparent molecular mass of the purified protein was calculated by gel filtration on a Superdex 200 HR 10/30 column. The void volume was determined with blue dextran, and a standard calibration curve was obtained using thyroglobulin (669 kDa), ferritin (440 kDa), catalase (232 kDa), aldolase (158 kDa), and albumin (67 kDa) (Amersham Biosciences).
Enzyme Activity Assay-To detect FBPase activity during protein purification, a spectrophotometric coupled enzyme assay was employed to measure the enzyme activity. FBPase activity was coupled with phosphoglucose isomerase and NADP-dependent glucose-6-phosphate dehydrogenase, and NADPH formation was measured. Assay mixture (1 ml) contained: 100 mM Tris-HCl buffer (pH 8.0), 0.4 mM NADP ϩ , 20 mM MgCl 2 , 20 mM dithioerythritol, 0.5 units of phosphoglucose isomerase (Sigma), 0.5 units of glucose-6-phosphate dehydrogenase (Sigma), and 50 l of protein sample. The reaction was initiated by adding 2 mM fructose 1,6-bisphosphate. The values obtained when glucose-6-phosphate dehydrogenase was omitted from the reaction mixture were subtracted in each measurement.
To determine the effects of temperature on enzyme activity, a reaction mixture containing 100 mM Tris-HCl buffer (pH 8.0), 20 mM MgCl 2 , and 20 mM dithioerythritol was incubated at the desired temperature (37-100°C) for 5 min. The FBPase reaction was initiated with the addition of enzyme and fructose 1,6-bisphosphate and was incubated for 1 min. Product formation was proportional to incubation time under these conditions. A control experiment without Tk-Fbp was performed at each temperature. After the incubation, the assay mixture was cooled in ice water for 5 min and then the exogenous enzymes and cofactor were added to initiate the coupling reaction. Incubation was carried out at 25°C for 3 min. Generation of NADPH was monitored at 340 nm. The reaction mixture without Tk-Fbp at each temperature was used as a blank value. When the effect of pH on enzyme activity was examined, the reaction was carried out in a reaction volume of 100 l containing 20 mM of the following buffers: citrate buffer (pH 4.5 to 6.5), MES buffer (pH 6.0 to 7.5), and Tris buffer (pH 7.5 to 9.0). All buffers were prepared so that their pH would reflect accurate values at 95°C. After the first reaction, 100 l of 1 M Tris-HCl (pH 8.0) was added to the reaction mixture to bring the pH of the reaction mixture to 8.0. For examination of the various metal ions effect on the enzyme activity, the first reaction mixture was incubated with the respective metal cations. After incubation the mixture was cooled in ice water and as glucose-6-phosphate dehydrogenase is a Mg 2ϩ -dependent enzyme, the coupling reaction was initiated by adding 20 mM Mg 2ϩ , NADP ϩ , and the exogenous enzymes.
Analysis of the Reaction Substrate and Product with High Performance Liquid Chromatography-Reaction mixtures (100 l) containing 100 mM Tris-HCl buffer (pH 8.0), 20 mM MgCl 2 , 20 mM dithioerythritol, 2 mM fructose 1,6-bisphosphate, and 12 g of purified Tk-Fbp were incubated at 50°C. Substrates other than fructose 1,6-bisphosphate, such as fructose 2,6-bisphosphate, fructose 6-phosphate, fructose 1-phosphate, glucose 6-phosphate, and glucose 1-phosphate were also incubated under the same conditions. Samples, after incubation for the desired period of time, were kept on ice for 10 min. The reaction mixture was centrifuged at 15,000 ϫ g and the supernatant was analyzed by high performance liquid chromatography with a Shodex Asahipak NH2P-50 4E column (Shodex, Tokyo, Japan). Sodium phosphate buffer (300 mM) at pH 4.4 was used as an eluent at a flow rate of 1 ml/min. Column temperature was set at 40°C and the product was detected with a refractive index detector.
RNA Isolation and Northern Blot Analysis-For isolation of RNA from strain KOD1, cells were harvested at the early log phase when A 660 was ϳ0.1. RNA was isolated using the RNeasy Midi Kit (Qiagen). For Northern blot analysis, 15 g of total RNA was denatured by heat treatment at 65°C for 15 min, separated by 1% agarose gel electrophoresis, and transferred to a nylon membrane (Hybond TM -N ϩ ; Amersham Biosciences) by capillary blotting. Digoxigenin labeling of DNA fragments, hybridization, and washing of the membranes were performed according to the instructions of the manufacturer (Roche Molecular Biochemicals). A DNA fragment corresponding to the entire Tk-fbp coding region was used as a probe. A 1.5-kilobase pair region within the coding region of the DNA ligase gene from strain KOD1 (24) was also used as a probe.

RESULTS
FBPase Activity in KOD1 Cells-T. kodakaraensis KOD1 cells were grown on pyruvate (1%) in the presence of 0.5% yeast extract and 0.5% tryptone. Under these growth conditions, gluconeogenesis was required to supply the necessary sugars for cell proliferation. FBPase activity was investigated in these cells. We detected FBPase activity in the cell extracts with a specific activity of 0.4 units/mg at 95°C.
Partial Purification and N-terminal Amino Acid Sequence of FBPase-We partially purified the FBPase from the cell extracts of pyruvate-grown cells. FBPase was purified 13-fold by anion exchange, hydrophobic, and gel-filtration column chromatography. SDS-PAGE analysis of the active fractions after partial purification displayed two major protein bands with molecular masses of 25 and 42 kDa (Fig. 1). During the purification procedure, intensities of the 42-kDa protein on SDS-PAGE corresponded well to the levels of FBPase activity in each fraction. The 25-and 42-kDa proteins were both subjected to N-terminal amino acid sequencing. The sequence of the 25-kDa protein (VVIGEKFPEVEVKTT) showed high similarity to probable peroxiredoxin proteins from various Archaea species. On the other hand, the N-terminal amino acid se-  Table I are shown by asterisks at the top of the alignment. Organism names are shown on the left, whereas residue numbers are shown on the right. The accession numbers for each sequence are listed in Table I. Cloning of the Tk-fbp Gene-Among the 15 N-terminal amino acid residues of the 42-kDa protein, 13 residues were identical to those of the hypothetical proteins from P. furiosus, P. abyssi, and P. horikoshii. Therefore, two oligonucleotides were designed: one from the N-terminal amino acid sequence of the 42-kDa protein, and the other based on a conserved C-terminal region of the hypothetical proteins from the Pyrococcus strains mentioned above. PCR with the two primers and genomic DNA of KOD1 as a template led to specific amplification of a DNA fragment with the expected length of ϳ1-kilobase pair. The entire gene was then isolated from the genomic library of strain KOD1 using the 1-kilobase pair DNA fragment as a probe. DNA sequence analysis identified an open reading frame consisting of 1125 bp encoding a protein of 375 amino acids with a calculated molecular mass of 41,658 Da. The N-terminal sequence deduced from the open reading frame was identical to the N-terminal sequence of the 42-kDa protein from strain KOD1. No other open reading frames were found in the immediate flanking regions of the open reading frame. A putative ribosomal binding site (5Ј-GGTGG) was identified 6 nucleotides upstream from the initiation codon along with a putative TATA-like element (TATAA, A-Box) 24 nucleotides upstream of the ribosomal binding site. A transcriptional termination signal (poly-(TC)) was also found downstream of the stop codon TGA. As we found that the gene encoded a protein with FBPase activity (see below), we named the gene Tk-fbp.
Amino Acid Sequence of Tk-Fbp-The deduced amino acid sequence of Tk-Fbp displayed high similarity to hypothetical proteins of unknown function from various Archaea strains and the bacterium Aquifex aeolicus. These included the hypothetical proteins from the Pyrococcus species with similar N-terminal amino acid sequences. The Tk-Fbp sequence did not show similarity with previously reported Class I, Class II, or Class III FBPases. All orthologue gene products that were found in the genome data bases are listed in Table I and a representative alignment of the Archaea sequences is shown in Fig. 2. Six regions relatively conserved among Class I FBPases from mammals, plants, fungi, and bacteria have been identified (6). However, sequences with notable similarity to these regions were not found in the Tk-Fbp orthologues.
Production and Purification of Recombinant Tk-Fbp-To characterize the protein product of the Tk-fbp gene, and to   Cell-free extract  40  190  361  1.9  100  1  Heat-treatment  39  26  336  13  93  7  Resource Q  18  16  288  18  79  9  Mono Q  15  11  220  20  60  10  Resource ISO  13  9  207  23  57  12  Superdex 200  11  8  192  24  53  13 determine whether the enzyme was a bona fide FBPase, we expressed the gene in E. coli. When E. coli BL21(DE3) cells harboring the expression plasmids were grown at 37°C and induced with 1 mM isopropyl-1-thio-␤-D-galactopyranoside for gene expression, the gene product was produced in a soluble form. The recombinant protein was purified to apparent homogeneity by heat treatment at 85°C for 20 min followed by anion exchange and hydrophobic and gel filtration chromatography (Table II, Fig. 3). The molecular mass of recombinant Tk-Fbp estimated by SDS-PAGE agreed with that deduced from the amino acid sequence. Furthermore, the N-terminal 14 amino acid residues of the purified protein were identical to the deduced amino acid sequence of the gene, confirming that we had obtained purified Tk-Fbp. Molecular Mass Determination-The molecular mass of the purified Tk-Fbp was determined by gel filtration chromatography. Tk-Fbp eluted at a retention volume of 10.75 ml equivalent to a molecular mass of 340 kDa calculated from the standard curve obtained from the retention volume of the standard markers. Taking into account the molecular mass of the subunit (42 kDa), this result indicates that Tk-Fbp exists in an octameric form.

Effect of pH, Temperature, and Metal Cations on the Enzyme
Activity-Purified Tk-Fbp was dialyzed against 50 mM sodium phosphate buffer (pH 7.0) containing 10 mM EDTA and used for further analysis. Activity measurements were performed in a linked assay coupled with phosphoglucose isomerase and glucose-6-phosphate dehydrogenase. Tk-Fbp displayed significant FBPase activity in the linked assay, and no activity was observed when any component of the reaction mixture was omitted. Particularly, we could not detect activity when phosphoglucose isomerase was removed from the reaction mixture, indicating that Tk-Fbp harbored only FBPase activity and not an additional phosphoglucose isomerase activity. We examined the effects of divalent metal cations on the enzyme activity. Addition of EDTA did not lead to a decrease in enzyme activity. Addition of Mg 2ϩ significantly enhanced enzyme activity up to concentrations of 20 mM (Ͼ5-fold) (Fig. 4A). Addition of 1 mM Mn 2ϩ led to a 4-fold increase in activity, but higher concentrations led to a decrease in this effect. Zn 2ϩ also enhanced activity at 1 mM, whereas higher concentrations displayed an inhibitory effect. Ca 2ϩ and Ni 2ϩ had no notable effects on activity (Fig. 4A). We observed that dithioerythritol and, to a lesser extent, other reducing agents such as dithiothreitol and 2-mercaptoethanol stimulated enzyme activity at concentrations of 20 mM (Fig. 4B). One possibility may be that the reducing agents prevent inactivation of the enzyme by oxygen. We did observe a slight oxygen sensitivity in the enzyme, as we found a small decrease in activity when the purified enzyme solution was subjected to bubbling with air (ϳ6% decrease/min bubbling). A combination of Mg 2ϩ and dithioerythritol resulted in a 9-fold increase in enzymatic activity (Fig. 4B).
We examined the effect of pH and temperature on the FB-Pase activity of Tk-Fbp in the presence of 20 mM Mg 2ϩ and 20 mM dithioerythritol. Tk-Fbp displayed maximal activity at pH 8.0 (Fig. 5A). The enzyme showed a nearly linear increase in activity between 37 and 95°C, with a ϳ6-fold increase between these temperatures (Fig. 5B). At 95°C and pH 8.0, Tk-Fbp displayed a specific activity of 24 units/mg. Kinetic analysis was also carried out, and Tk-Fbp catalyzed the reaction following Michaelis-Menten kinetics with a K m value of 100 M toward fructose 1,6-bisphosphate, and a k cat value of 17 s Ϫ1 subunit Ϫ1 at 95°C. In a linked assay with fructose-1,6-bisphosphate aldolase and glycerol-3-phosphate dehydrogenase, Tk-Fbp did not exhibit activity for the reverse reaction. Thermostability of the recombinant protein was monitored in the presence of Mg 2ϩ and the protein was found highly stable even Detection of Substrate and Product with High Performance Liquid Chromatography-To further confirm the FBPase activity of Tk-Fbp, we examined the production of Fru-6-P from Fru-1,6-P 2 . Analysis with high performance liquid chromatography was carried out using D-Fru-6-P and D-Fru-1,6-P 2 as standards. Under our measurement conditions, D-Fru-6-P eluted at a retention volume of 4.22 ml, whereas the retention volume for D-Fru-1,6-P 2 was 17.63 ml. When purified Tk-Fbp was added in the reaction mixture we could detect the specific production of D-Fru-6-P from Fru-1,6-P 2 (data not shown).
The peak corresponding to D-Fru-6-P increased with longer incubation periods of the reaction mixture, and the substrate peak decreased. An important observation was that when fructose 2,6-bisphosphate, fructose 6-phosphate, fructose 1-phosphate, glucose 6-phosphate, and glucose 1-phosphate were incubated with Tk-Fbp, substrate levels did not change. The result indicates that the enzyme was specific for fructose 1,6-bisphosphate.
Transcriptional Regulation of Tk-fbp Gene-As mentioned above, FBPase is necessary when cells require the synthesis of sugars from gluconeogenic substrates such as pyruvate (2). To examine the regulation of gene expression of Tk-fbp, KOD1 cells were grown independently on pyruvate and starch. Two probes were constructed for Northern blot analysis, one corresponding to the Tk-fbp gene, and the other corresponding to the DNA ligase gene from strain KOD1 (24, 25) as a control. The mRNA of the Tk-fbp gene was clearly detected from the RNA of cells grown on pyruvate (Fig. 6). In contrast, a positive signal could not be detected with the RNA of cells grown on starch. When cells were grown on amino acids, a condition that requires gluconeogenesis, Tk-fbp transcripts could also be detected. When pyruvate and starch were both present in the medium, only a very faint signal could be observed. Under all conditions, the signals of mRNA for the DNA ligase gene were visible irrespective of the carbon source. This result provides direct evidence that the transcription of the Tk-fbp gene was regulated at the transcription level and under the control of glucose repression. We also measured FBPase activity in cells grown on pyruvate or starch. We found that starch-grown cells displayed a specific activity of 0.03 units/mg, ϳ8% of the activity detected in pyruvate-grown cells (0.4 units/mg).

DISCUSSION
Complete genome sequences have contributed enormously in identifying the presence or absence of particular genes in various microorganisms (26,27). However, this advantage heavily relies on the assumption that proteins with similar function will display similarity in primary structure. The archaeal FBPases have represented an intriguing exception to this assumption. Although FBPase activity had been detected in P. furiosus (20) and Methanobacterium thermoautotrophicum (19), no orthologue genes of previously identified FBPases were present on Archaea genomes. A most interesting and valuable finding was that the MJ0109 gene product from M. jannaschii harbored an unexpected FBPase activity in addition to its expected inositol monophosphatase activity. The report also mentions that orthologue genes from A. fulgidus and Thermotoga maritima also encode a protein with FBPase activity (21).
In this study, we have identified and characterized a novel FBPase, Tk-Fbp, from the hyperthermophilic archaeon, T. kodakaraensis KOD1. The structure is distinct to the structures of all previously identified FBPases, including the MJ0109 gene product from M. jannaschii. The FBPase activity has been detected by both an enzyme linked assay and direct observation of substrate and product, leaving no doubt that Tk-Fbp harbors FBPase activity. Tk-Fbp displayed a K m value of 100 M toward fructose 1,6-bisphosphate, and a k cat value of 17 s Ϫ1 subunit Ϫ1 at 95°C. The K m value was slightly higher than that observed for the MJ0109 gene product (38 M). The k cat value of Tk-Fbp at 85°C was ϳ2-fold higher than the M. jannaschii enzyme (7 s Ϫ1 subunit Ϫ1 at 85°C). The k cat value of Tk-Fbp at 37°C (2.9 s Ϫ1 subunit Ϫ1 ) was ϳ20% of that of the FBPase from E. coli (14.6 s Ϫ1 subunit Ϫ1 ). Tk-Fbp did not display catalytic activity for the reverse reaction, indicating that it is not the protein responsible for 6-phosphofructokinase activity in strain KOD1. In support, phosphofructokinases have been identified and characterized in various Archaea including P. furiosus (28), M. jannaschii (29), and Aeropyrum pernix (30,31), and these proteins do not correspond to the Tk-Fbp orthologues mentioned in this study.
The results of this study strongly indicate that Tk-Fbp is the major FBPase in T. kodakaraensis KOD1. Besides the high activity of the purified enzyme mentioned above, we could not detect FBPase activity in fractions other than those containing Tk-Fbp during partial purification from pyruvate-grown KOD1 cells. Furthermore, the enzyme displayed high substrate specificity toward fructose 1,6-bisphosphate, unlike the MJ0109 gene product from M. jannaschii (21). No activity could be observed with fructose 2,6-bisphosphate, fructose 6-phosphate, fructose 1-phosphate, glucose 6-phosphate, and glucose 1-phosphate. Finally, the gene expression was strictly regulated in a manner that perfectly agreed with its presumed physiological role. Gene transcription was repressed in the presence of starch, regardless of the presence or absence of pyruvate. Derepression was also observed in cells grown on either amino acids or pyruvate, indicating that the regulation was not because of induction by a certain carbon source, but a typical example of glucose repression/derepression. These findings lead us to conclude that, at least in the case of T. kodakaraensis KOD1, Tk-Fbp is the "missing" archaeal FBPase.
At present, it is difficult to determine whether the Tk-Fbp orthologues represent the missing FBPases in other Archaea strains. However, it should be noted that among the complete microbial genome sequences, all thermophilic Archaea harbored a highly similar gene on their chromosomes (Table I). In addition, a Tk-fbp orthologue was also present in the hyperthermophilic bacterium A. aeolicus. This at least denies that Tk-fbp itself is not a unique gene for T. kodakaraensis KOD1. Additionally, as most of these genes display 75-97% similarity to Tk-fbp, it is highly likely that their gene products harbor FBPase activity. Gene disruption studies and biochemical characterization of Tk-Fbp orthologues in other Archaea strains will be an attractive subject of future research on gluconeogenesis in Archaea.