|Home | About | Journals | Submit | Contact Us | Français|
Protein phosphorylation plays an important role in cell signaling. However, in the Archaea, little is known about which proteins are phosphorylated and which kinases are involved. In this study, we identified, for the first time, a typical eukaryote-like Ser/Thr protein kinase and its protein partner, a forkhead-associated (FHA)-domain-containing protein, from the archaeon Sulfolobus tokodaii strain 7. This protein kinase, ST1565, physically interacted with the FHA-domain-containing protein, ST0829, both in vivo and in vitro. ST1565 preferred Mn2+ as a cofactor for autophosphorylation and for substrate phosphorylation; the optimal temperature for this was 45°C, and the optimal pH was 5.5 to 7.5. The critical amino acid residues of the conserved FHA and kinase domain sites were identified by performing a series of mutation assays. Thr329 was part of a major activation site in the kinase, while Thr326 was a negative regulation site. Several mutants with amino acid substitutions in the conserved FHA domain sites of ST0829 did not physically interact with ST1565. A structural model for the FHA domain demonstrated that the mutation sites were located at the edge of the protein and thus were in the domain that potentially interacts with ST1565. This report describes pioneering work on the third domain of life, the Archaea, showing that a protein kinase interacts with and phosphorylates an FHA-domain-containing protein. Our data provide critical information on the structural or functional characteristics of archaeal proteins and could help increase our understanding of fundamental signaling mechanisms in all three domains of life.
Protein phosphorylation is used most commonly by cells for appropriate responses to various environmental cues (9). Although protein kinases have long been considered restricted to the Eucarya, homologues of eucaryal protein kinases have been reported to occur in Bacteria and more recently in Archaea (10). The importance of Ser/Thr/Tyr kinases in cell signaling in eukaryotes has been widely documented (20, 22). Protein phosphorylation has been studied less intensively in archaea, the so-called “third domain” life, whose members usually live in extreme environments, such as environments with a high salt content, high temperature, or extreme pH.
The first evidence of Ser/Thr/Tyr protein phosphorylation in Archaea was reported for the extreme halophile Halobacterium salinarum following 32P radiolabeling (29). Subsequently, protein phosphorylation of an isolated ribosomal fraction from the extremely acidothermophilic archaeon Sulfolobus acidocaldarius was characterized (26, 27). Several studies have also employed phosphoamino acid-directed antibodies to obtain direct evidence of the presence of phosphotyrosine in archaeal proteins from Sulfolobus solfataricus, Haloferax volcanii, and Methanosarcina thermophila TM-1 (1). Based on a comprehensive analysis of complete genome sequences, archaeal representatives of novel putative protein kinase families were reported by Leonard et al. (17). Several kinase activities have since been confirmed (18, 19). Furthermore, the recent elucidation of the crystal structure of a protein kinase of the archaeon Archaeoglobus fulgidus Rio2 suggested that this protein defines an entirely new family of protein kinases (12-15).
In the Archaea, the actual proteins that are phosphorylated and the kinases that are involved in the reaction remain largely unknown (32). Among the archaeal proteins, the CheA and CheY proteins from H. salinarum are two of the best-characterized sensor and response regulator proteins associated with phosphorylation (6, 31). A two-component system has been proposed for responses to various chemotactic and phototactic stimuli in the Archaea (23, 24). Recently, Aivaliotis et al. completed a genome-wide and site-specific phosphoproteome analysis of H. salinarum. They indicated that phosphoproteins are involved in a wide variety of cellular processes and are especially enriched in metabolic and translation processes (1). The study of these workers provided systematic evidence that protein phosphorylation is a general and fundamental regulatory process that is not restricted to eukaryotes and bacteria. Sequence evidence has suggested that archaea contain abundant protein kinases that are phosphorylated on serine, threonine, and tyrosine residues.
Some archaeal open reading frames (ORFs) have a number of the features characteristic of the eukaryotic protein kinase superfamily (1, 25). However, the overall level of sequence identity is not particularly high for these deduced protein kinases. Although catalytic capabilities have been inferred from the primary sequences, the structural or functional properties of archaeal protein kinases have not been characterized. For example, although the importance of Ser/Thr/Tyr kinases for cell signaling has been widely documented for eukaryotes and in some bacteria, very few target substrates for archaeal protein kinases have been identified.
Recently, a protein containing a forkhead-associated (FHA) domain has been proposed to interact with a protein partner in a process regulated by reversible protein phosphorylation (4). The FHA domain was determined to be a phosphoprotein recognition unit with a preference for phospho-threonine (pT) peptides (4, 5, 33). It usually is present in eukaryotic proteins (for example, in several forkhead-type transcription factors ) and has been characterized for some bacterial proteins (21). Such domains bind phospho-threonine peptides and mediate phosphorylation-dependent protein-protein interactions in a variety of cell signaling processes (21). However, the residues in the FHA domain are not well conserved, although there do appear to be conserved residues that are involved in recognition of the phosphopeptide backbone or a pT residue (4, 5). There have been no reports yet of any binding partner for any archaeal FHA domain (21).
In this study, we identified a typical eukaryote-like Ser/Thr protein kinase and its protein partner, an FHA-domain-containing protein, obtained from the archaeon Sulfolobus tokodaii strain 7. This protein kinase, ST1565, physically interacts with the FHA-domain-containing protein, ST0829, both in vivo and in vitro. ST1565 has clear autophosphorylation and substrate phosphorylation activities. Conserved FHA and kinase domain sites were identified by performing a series of mutation assays. This paper describes, for the first time, a protein kinase that interacts with and phosphorylates an FHA-domain-containing protein in the third domain of life, the Archaea. Our findings include essential information concerning the structural or functional characteristics of archaeal proteins mentioned above.
Escherichia coli BL21 [F− ompT hsdSB(rB− mB−) gal dcm (DE3)] cells and pET28a containing the T7 RNA polymerase promoter were purchased from Novagen and were used to express archaeal proteins. The pBT and pTRG vectors, E. coli XR host strains, and the reagents for the two-hybrid assay were purchased from Stratagene. Restriction enzymes, T4 ligase, DNA polymerase, deoxynucleoside triphosphates (dNTPs), and all antibiotics were obtained from TaKaRa Biotech. PCR primers were synthesized by Invitrogen (see Table S1 in the supplemental material). Ni2+-nitrilotriacetate (Ni-NTA) and glutathione S-transferase (GST) agarose were obtained from Qiagen.
Prokaryotic recombinant vectors expressing the genes for archaeal proteins and mutant proteins were constructed. E. coli BL21 CodonPlus(DE3)-RIL (Novagen) was used as the host strain to express archaeal proteins as described previously (34). Protein concentrations were determined by spectrophotometric absorbance at 260 nm, as described by Gill and Hippel (7).
A BacterioMatch II two-hybrid system library construction kit (Stratagene) was used to detect protein-protein interactions between a protein kinase and an FHA-domain-containing protein. The bacterial two-hybrid system detects protein-protein interactions based on transcriptional activation, and the analysis was carried out by using the procedure supplied with the commercial kit and our previously described procedures (34). The archaeal genes were amplified by PCR using specific primer pairs (see Table S1 in the supplemental material) from genomic DNA of S. tokodaii. pBT and pTRG vectors containing archaeal genes encoding protein kinase and FHA protein were generated. Positive-growth cotransformants were selected on screening medium plates containing 5 mM 3-amino-1,2,4-triazole (3-AT) (Stratagene), 8 μg/ml streptomycin, 15 μg/ml tetracycline, 34 μg/ml chloramphenicol, and 50 μg/ml kanamycin.
Equimolar amounts of normalized GST or GST-ST0829 proteins were combined with equimolar amounts of normalized His-tagged ST1565 proteins in 1.5-ml tubes containing 500 μl of phosphate-buffered saline (PBS). The protein mixtures were gently rocked at 4°C for 4 to 15 h. Before further purification, 60 μl of each mixture was removed and saved as a loading control. The remaining mixtures were then purified using the GST affinity assay as described above. All samples were subjected to SDS-PAGE. The protein bands were transferred to a nitrocellulose membrane. Western blot analysis was conducted using primary anti-ST1565 antibody (1:1,000) and secondary IgG-horseradish peroxidase (goat anti-rabbit) antibody (1:10,000). To quantify the protein, the signal was developed using diaminobenizidine (DAB) detection reagents, and the blot was photographed.
The in vivo interactions between the protein kinase and the FHA-domain-containing protein were analyzed by a coimmunoprecipitation (Co-IP) assay performed using our previously described modified procedures (34). Exponentially growing cells of S. tokodaii were harvested, resuspended, and lysed in 4 ml of buffer (50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 1 mM EDTA, 0.5% Nonidet P-40). Co-IP assays were performed by incubating with shaking 10 μg of archaeal cell extract with 3 μl of ST0829 antiserum in 100 μl of buffer for 3 h at 4°C. A 20-μl slurry of protein A Sepharose was added, and incubation was continued for another hour. Immune complexes were collected, and the beads were washed with buffer. Finally, the beads were resuspended in SDS-PAGE sample buffer. After boiling, the samples were analyzed by Western blotting using anti-ST1565 antibody.
Protein kinase activity was routinely assayed in solution. Briefly, in vitro phosphorylation was carried out by incubating 0.25 nM S. tokodaii protein kinase and 2.5 nM FHA-domain-containing protein in buffer (20 mM HEPES [pH 7.4], 10 mM MgCl2, 10 mM MnCl2) containing 300 μCi [γ-32P]ATP for 30 min at 55°C. The reaction was stopped with excess sample buffer, and proteins were separated by 10% SDS-PAGE and analyzed by autoradiography. Mg2+, Mn2+, and other divalent cations, including Ca2+, Cu2+, and Zn2+, at various concentrations were added to the reaction mixture to study their effects on the phosphorylation reactions.
The biochemical and genetic functions of archaeal FHA-domain-containing proteins have not been identified experimentally yet. The structure of the FHA domain of ST0829 was modeled computationally using the SWISS-MODEL automated comparative protein modeling web server (2), our previously described procedures (3), and the structure of Rad53 (16).
Using a BLAST assay, a group of proteins from the archaeon S. tokodaii strain 7 was found to contain conserved amino acid residues. As shown in Fig. Fig.1A,1A, these proteins, including ST1565, contain several conserved domains of a typical Ser/Thr protein kinase. For example, they contain a DVKPSN catalytic loop, a DFG motif, and conserved K166, D287, and D314 residues, indicating that they could be typical Ser/Thr protein kinases. On the other hand, as shown in Fig. Fig.1B,1B, an S. tokodaii protein, ST0829, contains a typical FHA domain in its C terminus and several conserved domain residues similar to those found in the proteins of the pathogen Mycobacterium tuberculosis. A Zn finger-RanBP domain is located in the N terminus of ST0829 (Fig. (Fig.1B1B).
To determine if the archaeal FHA-domain-containing protein, ST0829, is a substrate of the protein kinase ST1565, we examined the physical interaction between these two proteins. As shown in Fig. Fig.2A,2A, in our bacterial two-hybrid experiment, a positive cotransformant (CK+) grew on a selective screening medium, but a negative cotransformant (CK−) did not grow. Moreover, the ST1565-ST0829 cotransformant grew well on the selective screening medium, proving that ST1565 interacted with ST0829. No growth was observed for the ST0829 self-activation controls (Fig. (Fig.2A).2A). In addition, an unrelated ABC family kinase, ST1652, was unable to interact with ST0829 because the cotransformant strain did not grow (Fig. (Fig.1A).1A). To ascertain the interaction, a GST pull-down-Western blot assay was conducted. As shown in Fig. Fig.2B,2B, His-tagged ST0829 protein could be readily pulled down by the GST-tagged ST1565 kinase protein. GST coincubated with His-tagged ST1565 did not produce any specific bands (Fig. (Fig.2B2B).
The physiological significance of these in vitro interactions was studied by performing a coimmunoprecipitation (Co-IP) experiment. The in vivo physical interaction between ST1565 and ST0829 was examined using protein A beads that were first conjugated with antibody raised against ST1565. As shown in Fig. Fig.2C,2C, in our Co-IP assay ST0829 clearly associated with ST1565 as an additional specific hybridization signal was detected compared with a negative control (no beads added). Therefore, ST1565 kinase physically interacted with the ST0829 FHA-domain-containing protein under archaeal physiological conditions.
To establish which metal ion is essential for optimal activity of ST1565, the activating effects of several divalent metal ions on the autophosphorylation activity of this protein were tested. As shown in Fig. Fig.3A,3A, for the five metal ions examined, the protein kinase exhibited the best activity when Mn2+ was added to the reaction mixture. In comparison, very low levels of activity were observed when several other metal ions, such as Ni2+, Zn2+, Mg2+, and Ca2+, were added (Fig. (Fig.3A3A).
As shown in Fig. Fig.1A,1A, using a BLAST assay, several conserved residues, such as K166, D287, D314, T326, and T329, associated with potential protein kinases in S. tokodaii were discovered. These residues were also located in or close to the major functional domains of the Ser/Thr protein kinase (for example, the catalytic loop and DFG motif), as shown in Fig. Fig.1A1A (lower panel). When these mutant proteins were purified and their autophosphorylation activities were compared with that of the wild-type protein, several amino acid substitution mutants, including ST1565(K166A), ST1565(D287A), ST1565(D314A), and ST1565(T329A), exhibited very weak activity, as shown in Fig. Fig.3B.3B. One mutant, ST1565(T326A), exhibited a higher level of autophosphorylation activity than the wild-type protein (Fig. (Fig.3B).3B). As shown in Fig. Fig.3C,3C, in another time course experiment with ST1565(T326A) there was a stepwise increase in autophosphorylation activity as the reaction time increased, and the final rate was 2.0 nmol 32P/min/mg. In contrast, the highest rate for the wild-type protein was 1.0 nmol 32P/min/mg, although this level of activity was reached in a shorter time (20 min).
Therefore, several residues of ST1565 protein kinase were found to be essential for autophosphorylation activity, but the T326 residue of ST1565 was not essential.
The physical interaction of ST1565 with ST0829 suggests that there is a functional correlation between these proteins. To determine if ST0829 could be phosphorylated by ST1565, we assayed the activity with different metal ions. As shown in Fig. Fig.4A,4A, no obvious phosphorylation activity was observed if no metal ion was added to the reaction mixture. Nor was activity of the unrelated kinase ST1652 observed (see Fig. Fig.1S1S in the supplemental material). Mn2+ clearly stimulated the phosphorylation of ST0829 by ST1565, which was consistent with the autophosphorylation of ST1565. Additionally, Mg2+ also stimulated the activity (Fig. (Fig.4A),4A), although to a lesser extent than Mn2+.
To examine the optimal conditions for the kinase activity of the protein from the extremely thermoacidophilic archaeon S. tokodaii, we examined the phosphorylation activities under different temperature and pH conditions. As shown in Fig. Fig.4,4, when ST0829 was used as the substrate, the optimal temperature for the kinase activity of ST1565 was approximately 45°C (Fig. (Fig.4B),4B), and the optimal pH ranged from 5.5 to 7.5 (Fig. (Fig.4C);4C); these values were consistent with the physiological environment of this archaeon.
To characterize the effects of the conserved residues of ST1565 on the protein kinase, several amino acid substitution mutants, including ST1565(K166A), ST1565(D287A), ST1565(D314A), ST1565(T326A), and ST1565(T329A), were purified and their kinase activities were assayed. As shown in Fig. Fig.5A,5A, four amino acid substitution mutant proteins, namely, ST1565(K166A), ST1565(D287A), ST1565(D314A), and ST1565(T329A), did not exhibit kinase activity, as no phosphorylated ST0829 was observed. Unexpectedly, approximately 5.5-fold-higher activity was observed with the ST1565(T326A) mutant, indicating that residue 326 negatively regulated the kinase activity of ST1565 with the substrate protein ST0829 (Fig. (Fig.5A).5A). This most likely is because the ST1565 Thr326 residue could compete with the phosphate group for the active center of ST1565 and thus partially inhibit both autophosphorylation and transphosphorylation activities. This was consistent with the previous observation that the autophosphorylation activity of ST1565(T326A) was higher than that of wild-type ST1565 protein (Fig. (Fig.3B).3B). On the other hand, when the effects of several conserved ST0829 residues on the kinase activity of ST1565 were examined, smaller amounts of the ST0829 amino acid substitution mutant proteins than of the ST0829 wild-type protein were phosphorylated (Fig. (Fig.5B).5B). Therefore, all of the residues of ST0829 mentioned above had a negative effect on the kinase activity of the ST1565 protein.
As shown in the Fig. Fig.1B,1B, the BLAST assay results indicated that the FHA-domain-containing protein in S. tokodaii contains several conserved residues, such as R164, S178, T199, and N200. These residues were also located in the potential FHA domain of ST0829, as shown in Fig. Fig.1B1B (lower panel). To examine if these ST0829 conserved residues have an important role in the interaction with ST1565, we conducted a bacterial two-hybrid experiment. As shown in Fig. Fig.6A,6A, cotransformants of several ST0829 amino acid substitution mutants with ST1565 exhibited only very weak growth on the selective screening medium, while the wild-type ST1565-ST0829 cotransformant grew well. Additionally, a positive cotransformant (CK+) grew on the selective screening medium, but a negative cotransformant (CK−) did not grow. No growth was observed for the self-activation controls for the ST0829 mutants (Fig. (Fig.6A).6A). This result indicated that several ST0829 mutants could not interact with ST1565.
Using the SWISS-MODEL automated comparative protein modeling web server (23) and the Rad53 FHA domain (PDB ID: 2JQI) as a template, a structural model of the FHA domain of ST0829 (Fig. (Fig.6B)6B) was obtained. As shown in Fig. Fig.6B,6B, several mutant residues were located at the edge of the FHA domain, which indicated that these residues could be involved in the interaction between ST0829 and ST1565.
Protein phosphorylation on serine, threonine, and tyrosine is one of the most important posttranslational modifications in eukaryotes and bacteria (30). However, previously, there was no specific information concerning a Ser/Thr protein kinase and its partner substrate in Archaea, the third domain of life. In this study, we successfully characterized an archaeal Ser/Thr protein kinase and its partner substrate. In particular, for the first time, an archaeal FHA-domain-containing protein was found to be the substrate phosphorylated by a typical Ser/Thr protein kinase. Moreover, the conserved sites for both the kinase and the FHA-domain-containing protein were identified using a series of mutation assays. Several residues essential for the kinase activity were characterized, and a negative-regulation Thr326 site was found. A number of conserved residues in ST0829 were found to be important for interaction of this protein with ST1565 and for kinase activity. These data provide important clues that increase our understanding not only of the structural and functional characteristics of the archaeal proteins but also of a fundamental signaling mechanism in all three domains of life (eukaryote, bacteria, and archaea).
The thermoacidophilic archaeon S. tokodaii strain 7 grows under high-temperature and low-pH conditions. We discovered that the archaeal Ser/Thr protein kinase ST1565 has an optimal temperature of 45°C and a pH range of 5.5 to 7.5, which is consistent with its physiological environment. ST1565 preferred divalent cations as a cofactor for autophosphorylation and substrate phosphorylation, which is a feature shared with some bacterial and eukaryotic Ser/Thr protein kinases (11). This indicates that the third domain of life, Archaea, has a general catalytic mechanism for protein phosphorylation, like in the other two domains, Eucarya and Bacteria.
Several ORFs potentially encoding eukaryote-like protein kinases have been identified in members of Archaea (17, 28) based on computer assays. However, at present, only a few proteins have actually been demonstrated to possess the catalytic activity implied by their sequences (18). There also has been no report showing that the genome of any archaeon encodes a typical eukaryote-like Ser/Thr protein kinase. In the present study, we characterized a group of this kind of archaeal kinases, although the level of overall sequence identity was extremely low. As shown in Fig. Fig.1,1, conserved residues that correspond to those of eukaryotic Ser/Thr kinases (11) and are responsible for ATP binding, phosphotransfer, metal ion binding, and autophosphorylation were found in the archaeon examined. The importance of these residues for the function of the kinase was clear, since mutations at these sites resulted in loss of kinase activity.
Little is known about the structural or functional characteristics of archaeal eukaryote-like Ser/Thr protein kinases. Our results demonstrated that the archaeal kinase exhibits general and fundamental conservation with corresponding eukaryotic proteins. However, an unexpected finding was the totally different effects of two mutations in the activation loop of the enzyme on the protein kinase function (see Fig. S1 in the supplemental material). In contrast to the loss of activity due to mutation of Thr329, mutation of Thr326 obviously stimulated the autophosphorylation activity (Fig. (Fig.3)3) and improved the phosphorylation activity with the FHA-domain-containing substrate tremendously (about 5.5-fold) (Fig. (Fig.5).5). Thus, these two Thr residues contribute in totally different ways to the activity of the ST1565 protein kinase. In contrast to the negative regulation of Thr326, Thr329 was shown to be essential for both autophosphorylation and transphosphorylation activities of ST1565 (Fig. (Fig.33 and Fig. Fig.5).5). This implies that Thr329 is an important activation site for the kinase function. This finding provides important clues not only to the structural characteristics of archaeal protein kinases but also to the origins and evolution of a fundamentally important regulatory mechanism in eukaryotes.
Recently, an FHA-domain-containing protein has been proposed to interact with a protein partner in a process regulated by reversible protein phosphorylation in both eukaryotes and bacteria (4). However, the presence of FHA-domain-containing proteins in members of the Archaea has not been described previously, and there have been no reports of experimental identification of any binding partner of any archaeal FHA domain (21). In this study, we found several conserved FHA domain-like genes in the extreme acidothermophilic archaeon S. tokodaii (Fig. (Fig.1B).1B). Based on the structure of Rad53 (16), using homology structure modeling, we obtained a structure for the archaeal FHA domain (Fig. (Fig.6B).6B). The conserved sites of the FHA-domain-containing protein were also confirmed to be essential for interaction with and phosphorylation of the ST1565 kinase. All of the FHA-domain-containing mutant proteins in our study exhibited a partial loss of phosphorylation activity with the protein kinase (Fig. (Fig.5).5). Interestingly, in contrast to the wild-type protein, these mutant proteins also were not able to interact with protein kinase, based on the results of our bacterial two-hybrid experiment (Fig. (Fig.6A).6A). This demonstrated that the extensively conserved FHA domain sites might be important for recognition and phosphorylation of an FHA-domain-containing protein by its corresponding Ser/Thr protein kinase in all three domains of life (Eukaryota, Bacteria, and Archaea).
Numerous phosphorylated archaeal proteins have been reported previously. However, the cellular impact of the phosphorylations is largely unknown. The FHA-domain-containing protein ST0829, which we characterized as a partner substrate of the protein kinase ST1565, is a potential transcriptional regulator (Fig. (Fig.1B).1B). It contains an N-terminal Zn finger-RanBP domain, which is usually responsible for DNA-binding protein. Our results suggest that the phosphorylation signal might participate in transcriptional regulation in Archaea. This may be vital for the extreme acidothermophilic archaeon to exhibit appropriate adaptive responses to environmental cues. Our data are also the first data showing that the Ser/Thr protein kinase signaling may be coupled with transcriptional regulation in an archaeon. However, further detailed study is necessary.
In conclusion, here we present primary data showing that a protein kinase interacts with and phosphorylates an FHA-domain-containing protein in a member of the third domain of life, the Archaea. The information obtained for the structure, function, and protein-protein interaction of the archaeal proteins provides important clues for understanding the mechanisms by which these unique organisms adapt to their extreme environments and also provides a way to trace the origins and evolution of a fundamental biological signaling transduction mechanism.
We thank Yulong Shen (Shandong University, Shandong, China) for providing the archaeal strain.
This work was supported by the National Natural Science Foundation of China, the 973 Program (grant 2006CB504402), the New Century Excellent Talents Fund of the Ministry of Education of China (grant NECT-06-0664), the Doctoral Fund of the Ministry of Education of China (grant 200805040004), and the China National Fundamental Fund of Personnel Training (grant J0730649).
We have no conflict of interest.
Published ahead of print on 29 January 2010.
†Supplemental material for this article may be found at http://jb.asm.org/.