|Home | About | Journals | Submit | Contact Us | Français|
In the current study, we examined the regulatory interactions of a serine/threonine phosphatase (BA-Stp1), serine/threonine kinase (BA-Stk1) pair in Bacillus anthracis. B. anthracis STPK101, a null mutant lacking BA-Stp1 and BA-Stk1, was impaired in its ability to survive within macrophages, and this correlated with an observed reduction in virulence in a mouse model of pulmonary anthrax. Biochemical analyses confirmed that BA-Stp1 is a PP2C phosphatase and dephosphorylates phosphoserine and phosphothreonine residues. Treatment of BA-Stk1 with BA-Stp1 altered BA-Stk1 kinase activity, indicating that the enzymatic function of BA-Stk1 can be influenced by BA-Stp1 dephosphorylation. Using a combination of mass spectrometry and mutagenesis approaches, three phosphorylated residues, T165, S173, and S214, in BA-Stk1 were identified as putative regulatory targets of BA-Stp1. Further analysis found that T165 and S173 were necessary for optimal substrate phosphorylation, while S214 was necessary for complete ATP hydrolysis, autophosphorylation, and substrate phosphorylation. These findings provide insight into a previously undescribed Stp/Stk pair in B. anthracis.
A profile of the intracellular signaling proteins that regulate transition of Bacillus anthracis from dormancy to expression of virulence factors is emerging. Like many prokaryotes, B. anthracis utilizes two-component histidine kinase systems to regulate physiological changes and the expression of virulence factors. These systems include the Spo0 histidine kinase-based phosphorelay pathway (32, 37) and the Bacillus respiratory response A and B system involved in regulating toxin expression (36). Unlike for histidine kinase systems, little is known about reversible serine/threonine phosphorylation events in B. anthracis. These systems are common to eukaryotic cells (3, 14, 25, 40) but were only recently found in prokaryotes to modulate a variety of metabolic and physiological processes (1, 2, 7, 11, 12, 15, 17, 24, 28, 35, 38). Whether reversible serine/threonine phosphorylation contributes to similar events in B. anthracis is not known.
The current paradigm for prokaryotic serine/threonine kinases (STK) is based in part on the structure of PknB, a serine/threonine kinase from Mycobacterium tuberculosis that is structurally related to eukaryotic Hanks-type kinases (39). PknB autophosphorylates and is dephosphorylated by an M. tuberculosis phosphatase, PstP, in order to alter kinase activity (4). Similar to the findings for PnkB, Madec et al. identified critical autophosphorylated residues and autophosphorylated domains of PrkC, an STK from Bacillus subtilis (22), which suggested that the phosphorylation state of these residues impacts the activation of PrkC (22). These studies suggested that prokaryotic STKs exhibited activities similar to those of their eukaryotic homologs and were regulated by cognate phosphatases. Hence, studies of serine/threonine phosphatase (STP)/STK pairs may help define a core regulatory module in bacterial physiology and virulence, wherein the kinase autophosphorylates following interaction with stimuli and is subsequently downregulated by a cognate phosphatase when stimulus levels decline.
Analysis of the B. anthracis genome indicates that this organism has a single phosphatase-kinase pair encoded within a putative operon. This operon, between nucleotides 3588319 and 3678099 in the genome of B. anthracis Sterne, contains eight candidate open reading frames (ORFs). Six of the potential ORFs encode proteins involved in translation and DNA metabolism, while the phosphatase-encoding ORF (stp1) and the kinase-encoding ORF (stk1) are paired at the 3′ end of this operon. Examination of the genome sequences of several other Gram-positive bacteria indicates that this putative operon and the general orientation of stp1 and stk1 are conserved among members of the Firmicutes group of bacteria. B. anthracis Stp1 (BA-Stp1) and BA-Stk1 homologs influence a variety of bacterial processes. For example, homologs of BA-Stp1 and BA-Stk1 regulate growth in Bacillus subtilis (12), cell viability and segregation in Streptococcus agalactiae (28), competence in Streptococcus pneumoniae (26), and virulence in both Streptococcus pyogenes (17) and Staphylococcus aureus (9). Although kinases homologous to BA-Stk1 influence several bacterial processes in different species, the tandem association of this kinase with a phosphatase does not vary. This observation led us to hypothesize that the phosphatase (BA-Stp1) influences Ba-Stk1 activity by dephosphorylation.
In the current study, we analyzed the importance of BA-Stp1 and BA-Stk1 in the virulence of B. anthracis and assessed the biochemical interactions between these two proteins. Results from these studies indicate that this phosphatase-kinase pair contributes to the virulence of B. anthracis, as mutants lacking BA-Stp1 and BA-Stk1 exhibit decreased lethality in a mouse model of pulmonary anthrax. Furthermore, a series of biochemical analyses reveal an interaction between BA-Stk1 and BA-Stp1 where BA-Stk1 autophosphorylates in order to enhance kinase activity and is dephosphorylated by BA-Stp1 as a putative step in downregulating kinase activity as the levels of stimuli subside. Moreover, we have identified candidate serine and threonine residues that appear to modulate kinase activity. These findings provide insight into a previously undescribed serine/threonine phosphatase-kinase system in B. anthracis.
B. anthracis Sterne strain 7702, as described by Cataldi et al. (6), and Escherichia coli were grown in brain heart infusion broth (BHI) and Luria-Bertani (LB) medium, respectively, at 37°C.
Abelson murine leukemia virus-transformed murine macrophages derived from ascites of BALB/c mice (termed RAW 264.7) were obtained from the American Type Culture Collection (ATCC, Manassas, VA). The RAW 264.7 cell line was maintained in tissue culture-grade T-75 flasks grown in the presence of Dulbecco's modified essential medium (DMEM) (ATCC) supplemented with 10% fetal bovine serum (FBS) (ATCC) at 37°C with 5% CO2 in a humidified incubator. Cells were used between passages 10 and 20 for the infection experiments.
Allelic replacement was used to construct a double mutant lacking expression of BA-Stp1 and BA-Stk1. Briefly, a 3.75-kb fragment comprising 1,500-bp flanking regions sharing sequence similarity to the upstream and downstream region of the ba-stp1 gene was PCR amplified from B. anthracis Sterne 7702 genomic DNA using forward primer 5′-CCGGAATTCGGCATTAGATGCTAGAAAAGTGC-3′ and reverse primer 5′-CCGGAATTCGTCTAATTTGATAAGGGCCTTTAC-3′ incorporating EcoRI restriction sites. The PCR product was subsequently ligated to a pGEM-T-easy vector (Promega) using TA overhangs to generate the plasmid construct pGEM-PK101, which was transformed into E. coli JM109. Transformed clones were screened for ampicillin resistance, and DNA sequencing of the isolated plasmid confirmed clones positive for the insert. Inverse PCR was then performed using pGEM-PK101 as a template with right primer 5′-CGCGGATCCAGTGCAACGTGCTGATTGGAAAAC-3′ and left primer 5′-CGCGGATCCCTCGTCACCTCGTCTCTTCTTTGC-3′, which amplify a gene sequence inverse to ba-stp1 incorporating BamHI restriction sites. The inverse PCR product comprising the homologous flanking regions of ba-stp1 (inserted in pGEM-T-easy) was then ligated to the BamHI-digested aphA-3 gene (Ω-km2 cassette) isolated from pUTE618 (31) conferring kanamycin resistance. The resulting clone was then digested with EcoRI to release ΔBA-stp1-Kanr, which was then subcloned into pUTE568 (31), a generous gift from Theresa Koehler, producing pPK101. E. coli strain JM110 (Δdam Δdcm) was then transformed with pPK101 to obtain unmethylated DNA that was subsequently electroporated into B. anthracis Sterne 7702. Transformants were selected on solid LB agar plates containing kanamycin (100 μg/ml) and erythromycin (5 μg/ml). To select for the double-recombination event, isolated clones were serially passaged in antibiotic-free medium and clones exhibiting a Kanr Erms phenotype were identified. The mutant STKP101 was confirmed by PCR and maintained in a medium containing kanamycin (100 μg/ml).
Vegetative bacilli of the B. anthracis Sterne 7702 parent strain and the B. anthracis STPK101 mutant strain, grown in the presence of kanamycin (100 μg/ml), were induced to sporulate by growth in sporulation medium (PA broth) containing 0.6 mM CaCl2·2H2O, 0.8 mM MgSO4·7H2O, 0.3 mM MnSO4·H2O, 85.5 mM NaCl, and 8 g/liter nutrient broth (pH 6.0) as described by McKevitt et al. (23). Spore suspensions were diluted to working concentrations necessary for the corresponding experiment and heated to 70°C for 30 min to kill any residual vegetative cells as well as to heat activate the spores. RAW 264.7 macrophages were seeded into 24-well tissue culture-grade plates at a concentration of 7.5 × 105 cells/well in DMEM-10% FBS and incubated overnight at 37°C in 5% CO2. Macrophages were then infected with spores of the B. anthracis Sterne 7702 parent or the B. anthracis STPK101 mutant strain at a multiplicity of infection (MOI) of 10. Subsequently, the plates were centrifuged at 250 × g for 10 min. The plates were then incubated at 37°C, 5.0% CO2, for 30 min, after which gentamicin (10 μg/ml) was added to kill extracellular vegetative bacilli for an additional 30 min. Next, the infected macrophages were washed 10 times with DMEM-10% FBS and then incubated for up to 10 h. At 2-h intervals, the culture media were removed and the macrophages were scraped from the wells with phosphate-buffered saline (PBS), briefly vortexed, and subsequently serially diluted and plated on solid media. The total number of CFU was enumerated following an overnight incubation. The average and standard deviation were calculated from two independent experiments performed in triplicate.
In vivo analysis of B. anthracis STPK101 was performed using a mouse model of pulmonary anthrax similar to that previously described (20, 23). Briefly, female DBA/2 mice (housed in accordance with approved IACUC guidelines, University of New Mexico Health Sciences Center) were anesthetized and the trachea was exposed by a small incision under the skin. Spores were injected in a 50-ml volume using a bent 30-gauge needle. To determine the number of spores deposited per group, two mice were sacrificed and the number of total CFU was determined from homogenized lung tissue. The animals were monitored twice daily for survival and clinical signs.
Spore germination was determined using the protocol described by McKevitt et al. (23). Briefly, spore preparations (3 × 107) of B. anthracis Sterne 7702 parent and B. anthracis STPK101 mutant strains were heat activated (20 min at 65°C) and resuspended in 200 μl of medium (PBS, l-alanine, inosine, or 25% BHI) in a well of a 100-well Honeycomb 2 (Thermo Electron Corp) microplate. Germination was monitored at 37°C as loss of optical density at 600 nm caused by changes in the refractivity of the spore. Germination was analyzed as percent decrease in optical density measured every 3 min for 30 to 60 min under shaking conditions. Experiments were carried out in triplicate, and mean absorbance was determined.
To analyze sporulation, B. anthracis Sterne parent and B. anthracis STPK101 mutant vegetative cells were cultured in sporulation medium containing 0.6 mM CaCl2·2H2O, 0.8 mM MgSO4·7H2O, 0.3 mM MnSO4·H2O, 85.5 mM NaCl, and 8 g/liter nutrient broth (pH 6.0) for 3 days at 30°C. The samples were heated at 70°C for 15 min to kill vegetative bacilli, and the number of heat-resistant CFU was determined by serial dilution and plating. The percentage of heat-resistant CFU from the total number of CFU was calculated to determine the sporulation efficiency.
Genomic DNA was isolated from an overnight culture of B. anthracis Sterne 7702 in BHI. To construct His-BA-Stp1, a 753-bp fragment was PCR amplified with forward primer 5′-GGAATTCCATATGGAAGGAGCGCAAAGAAGAGAC-3′ and reverse primer 5′-GGCGCGGATCCGCTTATAACGGTCATTTAAGA-3′ from isolated genomic DNA using High Fidelity polymerase (Roche). Similarly, the putative catalytic domain of BA-Stk1cat was PCR amplified from genomic DNA using the following oligonucleotides: 5′-GGGATTCATATGGTGCTGATTGGAAAACGCTTAAATG-3′ and 5′-CGGATCCTTGTCGCTTCCATATCTTCCGG-3′. The primers were designed to incorporate NdeI and BamHI restriction enzyme sites, respectively. The PCR product was cleaned with a QIAquick PCR purification kit (Qiagen) and subsequently digested with NdeI and BamHI (New England Biolabs) overnight at 37°C. The digested PCR products were then ligated using T4 DNA ligase (Roche) into the expression vector, pET-15b, and transformed into E. coli XL1-Blue to obtain the recombinant plasmids pET-15b-BA-Stp1 and pET-15b-BA-Stk1cat. DNA sequence analysis of the BA-Stp1 insert at the Laboratory for Genomics & Bioinformatics Sequencing Core (OUHSC) demonstrated that it was identical to the locus BAS3714 of B. anthracis Sterne 7702.
The recombinant plasmids pET-15b-BA-Stp1 and pET-15b-BA-Stk1cat were transformed into E. coli BL21(DE3) cells (Novagen), and transformed clones were selected on LB agar plates containing ampicillin (50 μg/ml) (American Bioanalytical). Single colonies were grown in cultures to an optical density at 600 nm of 0.7 before being induced for 4 h at 37°C with 1 mM isopropyl β-d-1-thiogalactopyranoside (Denville Scientific). Cells were then harvested by centrifugation at 9,000 × g for 20 min, and the pellets were resuspended in 20 mM Tris-HCl, 500 mM NaCl, and 10 mM imidazole, pH 7.9, and lysed using a cell disruptor. Recombinant BA-Stp1 and BA-Stk1cat were purified using the His/Bind kit (Novagen) according to the manufacturer's instructions and desalted in 20 mM Tris-HCl using PD-10 columns (GE Healthcare). The purified proteins were visualized on a 12.5% SDS-PAGE gel stained with Coomassie brilliant blue. When necessary, the His tag of purified BA-Stp1 was cleaved using thrombin (Novagen), following the manufacturer's instructions.
All reactions were performed in 50 μl reaction buffer containing 50 mM Tris-HCl, pH 8.0, 5 mM MnCl2, 5 mM MgCl2, and 0.5 mM dithiothreitol (DTT) (Invitrogen). Single tablets of p-nitrophenyl phosphate (p-NPP) (Sigma) were dissolved in 5 ml of reaction buffer immediately before use to give a working stock of 10 mM. Fifty microliters of this working stock was added to each well in triplicate, and the reaction was initiated by adding 3 μg of purified BA-Stp1 to the appropriate wells. A405 was measured every 5 min for 90 min in a 96-well plate reader at 37°C.
Dephosphorylation of phosphopeptides was examined using a malachite green/molybdate phosphatase detection system (10). Briefly, 5 mM phosphopeptides [RRA(pT)VA (Promega), RRA(pS)VA, and RRLIEDAE(pY)AARG (Upstate)] were added to the reaction buffer containing 50 mM Tris-HCl, pH 8.0, 5 mM MnCl2, and 0.02% (vol/vol) β-mercaptoethanol. To initiate the reaction, BA-Stp1 was added to the reaction mixture across a range of concentrations from 0 to 600 mM, in a total final volume of 50 μl. The reaction mixtures were incubated at 37°C for 15 min. Subsequently, 50 μl of the molybdate dye/additive mixture (Promega) was added to stop the reaction and the plate was incubated for 15 min at room temperature to allow for color development. The A600 of the malachite green/molybdate complex was measured, and the amount of free phosphate released was determined using a standard curve. For enzymatic analysis, 1.3 μM BA-Stp1 and phosphopeptide concentrations ranging from 5 to 400 μM were used to determine the initial reaction rates. The Km and Vmax values were calculated by using nonlinear regression analysis (Prism software).
Mutagenesis of aspartic acid residues was performed using a QuikChange site-directed mutagenesis kit (Stratagene) according to the manufacturer's instructions. Briefly, 50 ng of pET-15b-BA-Stp1 was used as a template. Point mutations were inserted using synthetic oligonucleotide pairs as follows: D18A, forward primer 5′-AAAAGTTCGTCAACATAATGAAGCTAGTGCAGGAGTTTTTCACAAT-3′, reverse primer 5′-AATTGTGAAAAACTCCTGCACTAGCTTCATTATGTTGACGAACTTTT-3′; D36A, forward primer 5′-TTTAGCGGTAGTAGCAGCTGGAATGGGAGGTCATC-3′, reverse primer 5′-GATGACCTCCCATTCCAGCTGCTACTACCGCTAAA-3′; D194A, forward primer 5′-GATCAGCTACTTCTTTGCTCTGCTGGATTATCTAACAAAGTTTCT-3′, reverse primer 5′-AGAAACTTTGTTAGATAATCCAGCAGAGCAAAGAAGTAGCTGATC-3′; D233A, forward primer 5′-TGATCGTGGCGGGGAAGCTAATATCACCCTTGTTC-3′, reverse primer 5′-GAACAAGGGTGATATTGACTTCCCCGCCACGATCA-3′. Mutations were confirmed by DNA sequencing.
Phosphate release was measured by malachite green/molybdate reactivity after dephosphorylation of recombinant BA-Stk1cat by BA-Stp1 wild-type or aspartic acid mutant cells. Enzymatic analysis of BA-Stp1 with BA-Stk1cat as a substrate was performed. Briefly, BA-Stk1cat concentrations ranging from 2 to 50 μM were used to determine the initial reaction rates. The Km and Vmax values were calculated by using nonlinear regression analysis (Prism software). kcat (s−1) was calculated using the formula Vmax/Et (where E is the concentration of enzyme sites). The catalytic efficiency of the reaction (s−1 mM−1) was determined by the formula kcat/Km.
Dephosphorylation reactions of 2 μM BA-Stk1cat were initiated by the addition of equimolar amounts (5 μM) of either purified BA-Stp1, the D18A mutant, or a control of shrimp alkaline phosphatase (Roche) diluted in reaction buffer. Reaction mixtures were incubated at 37°C, stopped at 5, 15, and 30 min with the addition of 4× SDS-Laemmli buffer, and boiled for 5 min. Subsequently, 10-μl portions of the reaction products were separated on a 10% SDS-PAGE gel and transferred to a polyvinylidene difluoride (PVDF) membrane, which was probed with a 1:1,000 concentration of Phospho-(Ser) 14-3-3 binding motif antibody (Cell Signaling). The blot was then washed with Tris-buffered saline (TBS), 0.1% Tween-20 five times for 5 min each and incubated with goat anti-rabbit immunoglobulin G (heavy plus light chains)-horseradish peroxidase (HRP) conjugate antibody (Bio-Rad) for 1 h on a shaker at room temperature. The bands were visualized using enhanced chemiluminescence (Amersham) according to the manufacturer's instructions, followed by exposure to film.
Autophosphorylation was performed by the addition of 1 μCi of [γ-32P]ATP (Perkin Elmer) to 10 μg of purified BA-Stk1cat in 100 μl kinase reaction buffer (4 mM morpholinepropanesulfonic acid [MOPS], pH 7.2, 5 mM MnCl2, 0.05 mM DTT, and 10 μM ATP). The reaction mixture was incubated at 37°C, and unincorporated [γ-32P]ATP was removed using a G-50 Sephadex spin column (Amersham). One microgram of phosphorylated BA-Stk1cat was incubated with 3 μg of purified BA-Stp1 in phosphatase reaction buffer containing 50 mM Tris-HCl, pH 8.0, 5 mM MgCl2, and 0.5 mM DTT. Reactions were stopped at 0, 30, 60, and 90 min using 4× SDS-Laemmli buffer and boiled for 5 min. Samples were resolved by SDS-PAGE, and the dried gel was exposed and visualized by autoradiography. To test BA-Stk1 kinase activity after dephosphorylation with BA-Stp1, ATP and [γ-32P]ATP were replaced by thio-γATP and [35S]ATP-γS, respectively.
Kinase activities of 1.25 μM BA-Stk1cat wild-type, alanine mutant (T165A, S173A, S214A, T290A), or aspartic acid mutant (T165D, S173D, S214D, T290D) cells were assayed in reaction buffer (4 mM MOPS, pH 7.2, 5 mM MnCl2, 0.05 mM DTT) with 10 μM ATP (Cell Signaling Technology). After incubation for 90 min at 37°C, kinase activity was assayed using the Kinase-Glo luminescent kinase assay platform (Promega) and the luminescent signal was measured in a Victor 3 (Perkin Elmer) plate reader. The luminescent signal inversely correlates to the amount of kinase activity. The percent relative activity of BA-Stk1cat site-directed mutants was calculated as an inverse of units of luminescence, compared to that of the wild-type BA-Stk1cat. All reactions were performed in triplicate.
To analyze the effect of alanine or aspartic acid substitutions on the kinase activity of BA-Stk1cat, 2 μM purified wild-type BA-Stk1cat or site-directed mutant protein was incubated with 20 μM myelin basic protein (Upstate) in kinase buffer containing 1 μCi of [γ-32P]ATP. The reaction mixtures were incubated for 30 min at 37°C, stopped using 4× SDS-Laemmli buffer, and boiled for 5 min. Samples were resolved by SDS-PAGE, and the dried gel was exposed and visualized by autoradiography. The level of 32P incorporation was determined by phosphorimage analysis.
Mass spectrometry analysis was performed in the Molecular Biology Proteomics Facility (University of Oklahoma Health Sciences Center). Following SDS-PAGE and staining of autophosphorylated and dephosphorylated BA-Stk1cat, protein bands were excised and then digested with trypsin within the gel utilizing the Pierce in-gel tryptic digestion kit and then extracted from 1.0% trifluoroacetic acid-water and combined with the digestion solution. The samples were subjected to liquid chromatography-mass spectrometry (LC-MS) and LC-tandem MS (LC-MS-MS) on a Dionex Ultimate 3000 nanoscale high-performance liquid chromatograph (HPLC) connected to an Applied Biosystems QSTAR Elite mass spectrometer operated in the positive ion mode at a voltage of 2.3 kV. The peptide separation was performed on a PepMap C18 column (75-μm inner diameter, 150-mm length, packed with 3-μm particles) at a flow rate of 250 nl/min. For LC-MS-MS, data were collected in the Information Dependent Acquisition mode over a mass range of 300 to 2,000 m/z. Peptides with a +2 to +5 charge status were selected for MS-MS fragmentation. A survey scan in the MS mode was performed, followed by MS-MS fragmentations of the three most abundant ions in the mass range of 300 to 1,400 m/z for 5 s for each ion selected. The MS-MS data were collected over a mass range of 50 to 2,400 m/z. The MS data file was used to search the NCBI nonredundant protein database (release date, 16 February 2007) utilizing the MASCOT search engine (software version 2.2), both of which are installed on an on-site MASCOT server. The search parameters consisted of the database search of all Eubacteria (Firmicutes group), with trypsin selected as the enzyme, allowing one missed cleavage. Fixed modifications were set to carboxamidomethylation of cysteine residues, while oxidation of methionine, deamidation of Asn/Gln, and phosphorylation of serine and threonine were chosen for variable modifications. Protein identifications were made based upon MASCOT Mowse scores greater than the significance threshold value at P < 0.05. The elution point and relative ionization intensity of phosphorylated peptides and their nonphosphorylated versions were determined by LC-MS analysis over a mass range of 300 to 2,000 m/z. The differentially phosphorylated versions were observed by plotting the extracted-ion current for each peptide utilizing a 1.0-atomic-mass-unit window spanning the three most abundant isotopic forms of the +3 charged peptide.
A search of the B. anthracis Sterne genome revealed putative orthologs of eukaryotic STKs and phosphatases. In all, four putative STKs and six putative serine/threonine phosphatases (STPs) were identified. Of particular interest to our studies was a pair of ORFs juxtaposed at the 3′ end of a putative operon, as this was the only closely related STP/STK pair identified in B. anthracis. The STP (termed BA-Stp1) and STK (termed BA-Stk1) are encoded by BAS3714 and BAS3713 (GenBank accession numbers AAT56015 and AAT56016), respectively. No other closely associated STP/STK pair was identified within the genome of B. anthracis.
To characterize BA-Stp1/BA-Stk1 in B. anthracis Sterne, a double mutant (Δstp1 Δstk1) was constructed by allelic replacement using a Ω Kanr cassette consisting of the aphA-3 gene conferring kanamycin resistance, and the mutant was termed B. anthracis STPK101. In vitro analysis of B. anthracis STPK101 found that the mutant grew in liquid medium, sporulated, and germinated with efficiency similar to that of the parent strain (data not shown). Thus, we next determined whether the virulence of B. anthracis STPK101 was similar to that of the parent strain.
Macrophages are considered the primary site of B. anthracis spore germination following infection (13); hence, B. anthracis STPK101 was assessed for growth and survival in these cells. When RAW 264.7 macrophage-like cells were infected with spores of the parent B. anthracis Sterne strain or B. anthracis STPK101, the total number of CFU of B. anthracis STPK101 recovered at 10 h postinfection were significantly reduced compared to results for the parent strain (Fig. (Fig.1A).1A). A comparison between STPK101 and the parent strain found no differences in the rate of growth in tissue culture medium alone (data not shown), suggesting that the observed defect was due to attenuated germination, growth, or survival of the mutant within the macrophage.
Next, to assess the importance of the phosphatase-kinase pair in virulence of B. anthracis, the 50% lethal doses (LD50) of the parent B. anthracis Sterne strain and B. anthracis STPK101 were compared using a mouse model of pulmonary anthrax. As shown in Fig. 1B and C, the LD50 of B. anthracis STPK101 (4.5 × 105 spores) was 10-fold higher than the LD50 of the B. anthracis parent strain (4.5 × 104 spores). Collectively, these data indicate that the phosphatase-kinase pair contributes to the survival of B. anthracis during infection of cultured macrophages and following infection in a mouse model of pulmonary anthrax.
Because BA-Stp1 and BA-Stk1 appeared to be important in B. anthracis virulence, we began to analyze these proteins in more detail. We predicted that BA-Stp1 could dephosphorylate BA-Stk1, which made it critical to first establish that BA-Stp1 is indeed a phosphatase. Experiments using p-NPP as a synthetic substrate found that BA-Stp1 exhibits phosphatase activity and requires divalent metal ions, Mn2+ or Mg2+, to hydrolyze p-NPP (data not shown). Such a dependence on divalent cations suggested that BA-Stp1 could be a PP2C family phosphatase (1, 3, 5). This was confirmed by further biochemical analysis which revealed that BA-Stp1 is inactivated by nonspecific inhibitors such as sodium pyrophosphate but is unaffected by specific PPP and PP2B phosphatase inhibitors such as okadaic acid and calyculin A (data not shown).
Next, we performed experiments designed to identify the types of residues dephosphorylated by BA-Stp1. Synthetic phosphopeptides with phosphate groups at threonine [RRA(pT)VA], serine [RRA(pS)VA], or tyrosine [RRLIEDAE(pY)AARG] residues were used as substrates in a malachite green/phosphate detection assay. Phosphate release increased when phosphothreonine or phosphoserine peptides were used as substrates but not with the phosphotyrosine peptide (Fig. (Fig.2A).2A). Using these synthetic substrates, we also determined the kinetic parameters of the dephosphorylation reaction. The Vmax and Km values for the phosphothreonine peptide were calculated to be 115.3 pmol min−1 μg−1 and 121.5 μM, respectively. Similarly, the Vmax and Km values for the phosphoserine peptide were found to be 172.7 pmol min−1 μg−1 and 248.6 μM, respectively. The Km value of BA-Stp1 for the phosphotyrosine peptide was above the range for accurate determination (data not shown), which is consistent with low levels of Pi detected in the phosphate release assay.
Examination of BA-Stp1 identified residues (D18, D36, D194, and D233) which corresponded to metal-complexed sites in human PP2Cα, leading us to predict that BA-Stp1 might share a enzymatic mechanism with human PP2Cα (8, 16). To address this possibility, single aspartic acid mutants D18A, D36A, D194A, and D233A were generated in BA-Stp1 and each mutant was assayed for phosphatase activity using RRA(pS)VA or RRA(pT)VA as substrates. As shown in Fig. Fig.2B,2B, substituting alanine for these critical aspartic acid residues eliminated phosphatase activity of BA-Stp1, supporting our prediction that BA-Stp1 is similar mechanistically to the human PP2Cα.
Having confirmed that BA-Stp1 is a phosphatase, we next sought to determine the interaction of this phosphatase with BA-Stk1. To test whether BA-Stp1 dephosphorylates phosphoresidues on BA-Stk1, the catalytic domain of BA-Stk1, termed BA-Stk1cat, was cloned, expressed, and purified from E. coli. Western blot analysis with antibodies to phosphoserine and phosphothreonine residues demonstrated that BA-Stk1cat was autophosphorylated (data not shown), which is similar to previous findings for B. subtilis PrkC (21, 22). Phosphorylated BA-Stk1cat was then used as a substrate for BA-Stp1. As shown in Fig. Fig.3A,3A, when BA-Stk1cat was added to the reaction there was an increase in detectable levels of phosphate relative to the negative control, suggesting that BA-Stp1 hydrolyzes the phosphate groups on BA-Stk1cat. Kinetic analysis of dephosphorylation of BA-Stk1cat suggested that BA-Stp1 dephosphorylates phosphoresidues on BA-Stk1cat with a high affinity (Km, 9.65 μM) and a catalytic efficiency, kcat/Km, of 113.9 s−1 mM−1. Moreover, phosphate release was significantly reduced when each of the aspartic acid mutants D18A, D36A, D194A, and D233A was substituted for wild-type BA-Stp1 in the reaction (Fig. (Fig.3B3B).
Because BA-Stk1cat was isolated from E. coli in a recombinant form, it was necessary to show that the residues dephosphorylated by BA-Stp1 arose from autophosphorylation and not through cross-phosphorylation by E. coli kinases during isolation of the protein. To accomplish this, we incubated BA-Stk1cat with [γ-32P]ATP, thereby allowing the predicted autophosphorylation to occur. The 32P-labeled BA-Stk1cat was then incubated with BA-Stp1, or buffer control, and the level of incorporated 32P was determined by autoradiography. As shown in Fig. Fig.3C,3C, when labeled BA-Stk1cat was incubated with BA-Stp1 there was a precipitous decrease in the level of detectable phosphorylated kinase. Within 30 min following the incubation start, there was a greater than 90% reduction in the level of detectable phosphorylated kinase, indicating that BA-Stp1 dephosphorylates autophosphorylated BA-Stk1.
To confirm our prediction that BA-Stk1cat serine residues were dephosphorylated by BA-Stp1, we examined the dephosphorylated kinase using anti-phosphoserine antibodies (Fig. (Fig.4A).4A). As shown in the immunoblot, upon incubation with shrimp alkaline phosphatase, there was no detectable decrease in the phosphoserine signal. However, when BA-Stp1 was added to BA-Stk1cat, a precipitous reduction in the levels of phosphoserine was observed. Prolonged incubation (30 min) resulted in hydrolysis of the phosphate group from serine residues on BA-Stk1cat to levels that were no longer detectable by the anti-phosphoserine antibodies. Finally, when a similar analysis was performed using an inactive phosphatase mutant, BA-Stp1D18A, there was no detectable change in the level of phosphorylated BA-Stk1cat, further suggesting that this reaction requires catalytically active phosphatase.
Previous work by Madec et al. demonstrated that autophosphorylation is critical to the kinase activity of B. subtilis PrkC (22), suggesting that BA-Stp1-mediated dephosphorylation could alter the kinase activity of BA-Stk1. Hence, we next analyzed the impact of BA-Stp1 dephosphorylation on the kinase activity of BA-Stk1. To accomplish this, His-tagged BA-Stk1cat was incubated with recombinant BA-Stp1 or BA-Stp1D36A, both of which had the His tags removed. This approach was taken in order to separate BA-Stk1cat and BA-Stp1 prior to performing the kinase assay. After separation of the two proteins using nickel affinity chromatography, BA-Stk1cat was incubated with [γ-32P]ATP, and levels of autophosphorylated kinase were determined across a time course. As shown in Fig. Fig.4B,4B, pretreatment of the kinase with BA-Stp1 resulted in a reduction in the amount of 32P incorporated during autophosphorylation. In contrast, pretreatment of BA-Stk1cat with the inactive phosphatase mutant, BA-Stp1D36A, did not alter levels of kinase activity.
Because it was difficult to absolutely exclude the effects of trace amounts of phosphatase in these experiments, a second experimental approach was utilized to confirm that dephosphorylation reduced the kinase activity of BA-Stk1. After pretreatment of BA-Stk1 with BA-Stp1, BA-Stk1 autophosphorylation and kinase activity were measured using thio-γATP instead of ATP. The rationale for this approach was that thiophosphorylated substrates are resistant to dephosphorylation by protein phosphatases (4, 19). BA-Stk1 was preincubated with BA-Stp1 for 30 min, and the kinase activity was then measured in a time course using [35S]ATP-γS and myelin basic protein as substrates. As shown in Fig. Fig.4C,4C, pretreatment of BA-Stk1 with BA-Stp1 reduced its autophosphorylation and kinase activity toward myelin basic protein. Collectively, these data suggest that phosphorylation of BA-Stk1cat renders it active and that BA-Stp1 modulates this activity by dephosphorylation.
Phosphorylation and dephosphorylation appeared to be important for regulating the kinase activity of BA-Stk1cat, and this heightened our interest in identifying residues phosphorylated in BA-Stk1cat. Indeed, we predicted that such residues are involved in important aspects of BA-Stk1's enzymatic activity and might be subjected to regulation via BA-Stp1 dephosphorylation. Thus, in the next series of experiments we mapped phosphorylated regions in BA-Stk1. Autophosphorylated BA-Stk1cat was excised from an SDS-PAGE gel and trypsin digested, and the phosphopeptides were isolated using immobilized gallium. The enriched phosphopeptide pool was then subjected to LC-MS-MS analysis. The MS-MS spectra identified four phosphorylated peptides of BA-Stk1cat based on sequences from the NCBI nonredundant protein database bank (Table (Table1).1). Of these peptides, Stk1: 150-183, with the sequence VTDFGIATATSATTITHTNSVLGSVHYLSPEQAR, was predicted to have a cluster of seven phosphoresidues. When BA-Stk1cat was subjected to dephosphorylation by BA-Stp1, we observed a loss in the level of phosphorylated amino acids within this peptide fragment, indicating that BA-Stp1 hydrolyzes phosphate groups of this phosphopeptide (data not shown). A similar observation was made for the peptide Stk1: 208-221, with the sequence QPFSGESAVAIALK, indicating that serine residues on this peptide are dephosphorylated by BA-Stp1. Similar peptides were found by Madec et al. to be phosphorylated in PrkC (22), suggesting that PrkC and BA-Stk1 share autophosphorylation domains.
Madec et al. performed an extensive mutational analysis of candidate phosphorylated residues in the activation loop of PrkC by substituting alanine for threonines and serines in this domain of the kinase. PrkC mutants that could no longer phosphorylate substrates were then identified (22). Yet, whether these mutants were defective in ATP hydrolysis and autophosphorylation or whether sustained phosphorylation of these residues influenced activity was not determined. Moreover, whether BA-Stk1 utilized similar residues for autophosphorylation was not known. Thus, we examined representative serine/threonine residues for their contribution to regulation of BA-Stk1cat activity. The first two residues, T165 and S173, were selected as representatives of targets phosphorylated within the activation loop (T165) or outside the activation loop (S173) of the enzymatic domain of BA-Stk1cat. S214 was selected from the identified phosphopeptide Stk1: 208-221. Finally, T290 was selected as a representative of targets phosphorylated in the juxtamembrane region of BA-Stk1 (spanning 55 residues) that is implicated in the regulation of eukaryotic kinases (34). Alanine substitutions of these residues in BA-Stk1cat were made, and the impact of these mutations on kinase activity and autophosphorylation was determined. Kinase activity of the mutants was measured in terms of ATP hydrolysis by analyzing the consumption of ATP in a luminescence-based assay. As shown in Fig. Fig.5A,5A, the kinase activity of mutants S173A and S214A was reduced compared to that of wild-type BA-Stk1cat, suggesting that phosphorylation at residues S173 and S214 is important for BA-Stk1cat kinase activity. Mutants T165A and T290A, however, did not show reduced kinase activity, as levels of ATP hydrolysis were comparable to those of wild-type BA-Stk1cat.
To further test the importance of phosphorylation on these residues, aspartic acid substitutions were used to mimic a phosphorylated state of BA-Stk1 and the mutated proteins were tested for kinase activity by measuring levels of ATP hydrolysis. As shown in Fig. Fig.5B,5B, the kinase activity of S173D and S214D was similar to that of wild-type BA-Stk1cat, suggesting that the phosphorylated state of residues S173 and S214 is critical for kinase activity.
To confirm the role of these residues in the BA-Stk1 kinase activity, both alanine and aspartic acid mutants of T165, S173, S214, and T290 were analyzed in a myelin basic protein (MBP) phosphorylation assay using radiolabeled ATP. Analysis of autophosphorylation activity of the BA-Stk1 mutants demonstrated that the mutants undergo autophosphorylation to levels slightly reduced in comparison to, or similar to, that of wild-type BA-Stk1cat (Fig. (Fig.5C).5C). However, comparison of MBP phosphorylation by the mutants and wild-type BA-Stk1cat shows that kinase activities of both alanine and aspartic acid mutants of T165 and S173 are significantly reduced (Fig. (Fig.5C).5C). Calculation of the percent phosphorylation of MBP shows that mutants of T165 and S173 exhibit only 20% and <2% phosphorylation, respectively, compared to wild-type BA-Stk1cat (Fig. (Fig.5D).5D). These results suggest that the presence of the phosphate moiety on these residues is required for binding and/or phosphorylation of substrates. Though T165A and T165D had no detectable defect in ATP hydrolysis as demonstrated in Fig. Fig.5A5A and Fig. Fig.5B,5B, both mutants were attenuated in substrate phosphorylation (Fig. (Fig.5C5C and Fig. Fig.5D).5D). More importantly, the native phosphoserine at position 173 is essential both for ATP hydrolysis (Fig. (Fig.5A)5A) and for substrate phosphorylation (Fig. (Fig.5C5C and Fig. Fig.5D5D).
Analysis of MBP phosphorylation by mutant S214A also shows reduced activity; however, this defect is rescued by the aspartic acid substitution on the same site (Fig. (Fig.5D),5D), suggesting that the phosphorylation state of residue S214 is important not only for hydrolysis of ATP (Fig. (Fig.5A)5A) but for phosphorylation of substrates as well (Fig. (Fig.5D).5D). Though residue T290 was predicted as being phosphorylated in kinases homologous to BA-Stk1, alanine or aspartic acid substitutions at this site had no apparent effect on autophosphorylation or substrate phosphorylation (Fig. (Fig.5D5D).
The current findings provide insight into a paired serine/threonine phosphatase and kinase in B. anthracis. Several recent studies have shown that Stp/Stk pairs contribute to bacterial virulence and physiology, but the underlying molecular mechanisms and the regulation of these proteins remain poorly understood. To this end, the data provided herein address important gaps in our understanding of how the activity of a bacterial Stk is regulated by a partner phosphatase through dephosphorylation. The results of this study indicate that BA-Stp1 regulation of BA-Stk1 is a complex system of multiple residues targeted for dephosphorylation, with some phosphoresidues important only for autophosphorylation, while other residues are necessary for both ATP hydrolysis and substrate phosphorylation.
Analysis of the enzyme kinetic parameters (Km and Vmax) indicated that BA-Stp1 dephosphorylates phosphothreonine and phosphoserine residues with Km values similar to those reported for phosphatases, such as PphA from Synechocystis sp. strain PCC 6803 (30) and PrpZ from Salmonella enterica (18). Thus, these biochemical and enzymatic characterizations of BA-Stp1 all correspond to the expected activities of a PP2C phosphatase. The data shown in Fig. Fig.33 and and44 indicate that BA-Stp1 utilizes this PP2C phosphatase activity to modulate BA-Stk1, wherein dephosphorylation leads to a decline in kinase activity. Importantly, this reaction appears to be more specific, as BA-Stp1 exhibited a higher affinity toward BA-Stk1 than generic substrates (Km, 9.65 μM versus 248.6 μM).
A graphical summary of our data is presented in Fig. Fig.6.6. Unlike histidine kinases, whose phosphorylation status can either turn on or turn off a single phosphotransferase mechanism, BA-Stk1 exhibits a complex assembly of critical autophosphorylated residues, some of which contribute to ATP hydrolysis and autophosphorylation and others of which contribute only to substrate phosphorylation. For example, in the absence of phosphorylation of S173 and S214, substrate phosphorylation is attenuated but autophosphorylation increases. The phosphorylation of S173 and S214 may, therefore, shift the kinase from an autophosphorylating state to substrate phosphorylation. While it is difficult to absolutely exclude the influence of subtle conformational changes occurring from the alanine substitution, the aspartic acid substitutions further support the idea that the balance between these two biochemical activities is influenced by the phosphorylation of these residues. Indeed, when phosphorylation is simulated at S214 by aspartic acid substitution, autophosphorylation is reduced and substrate phosphorylation is enhanced. In a corresponding manner, when the residue cannot be phosphorylated due to an alanine substitution, Stk1 increases autophosphorylation and decreases substrate phosphorylation. Conversely, mimicking phosphorylation of S173 by aspartic acid substitution reduced autophosphorylation and substrate phosphorylation, indicating that the addition of a phosphate group to this residue has an overall negative effect on kinase activity. Finally, aspartic acid substitution at T165 slightly increased autophosphorylation, suggesting that phosphorylation of this residue enhances kinase activity. We predict that T165 is one of the key residues dephosphorylated by BA-Stp1 leading to downregulation of kinase activity and that the presence of the threonine at position 165 is necessary to allow for binding and/or phosphorylation of target substrates. Collectively, we observed that phosphorylation of some residues in BA-Stk1 promotes kinase activity, while phosphorylation of other residues represses kinase activity. These results reveal a sophisticated mechanism of control for BA-Stk1 kinase activity, wherein a network of phosphorylated residues are altered to either activate or inactivate the enzymatic activity of this signaling protein in B. anthracis.
A comparison of our findings for BA-Stp1 relative to those for B. subtilis PrpC is instructive. Like BA-Stp1, PrpC is believed to be functionally coupled with its partner kinase, PrkC. Moreover, PrpC was found to dephosphorylate PrkC in earlier work by Obuchowski and colleagues (27) and this is similar to our current results on Ba-Stp1 dephosphorylation of Ba-Stk1. However, the impact of dephosphorylation on kinase activity has not been determined for the PrpC/PrkC pair. PrpC/PrkC also differs from BA-Stp1/BA-Stk1 in that the phosphatase-kinase pair appears to regulate late-stationary-growth-phase events (e.g., sporulation and biofilm formation) in B. subtilis, an effect we did not observe in B. anthracis. Despite several attempts, we could not detect or demonstrate any difference in sporulation efficiency between the parent B. anthracis Sterne strain and B. anthracis STPK101 (data not shown). In fact, extensive analysis of B. anthracis STPK101 did not find any notable growth defect under in vitro conditions, and a prominent phenotype became apparent only when infection models were used. These data suggest that BA-Stp1/BA-Stk1 may have been refined by B. anthracis for growth and virulence in vivo, while PrpC/PrkC have evolved to regulate events more important to survival of B. subtilis in the environment. Very recently, Shah et al. (33) reported a role for B. subtilis PrkC in sensing muropeptides released during germination, and B. anthracis was among the list of other bacteria for which this group defined this system. The study by Shah et al. found that null mutants lacking the gene for PrkC and BA-Stk1 were unable to sense muropeptide and peptidoglycan as a germinant. Thus, although we did not observe a defect in germination of the mutant in vitro, it is reasonable to suspect that the attenuation of virulence could be due to defects in the ability of B. anthracis to regulate peptidoglycan-related germination during the establishment of disease.
Our findings report the importance of a critical serine/threonine phosphatase-kinase pair in the virulence of B. anthracis and highlight a mechanism by which a bacterial phosphatase modulates the activity of its partner kinase at the molecular level. We hypothesize that BA-Stk1 acts as a sensor for regulating virulence in response to stimuli and does so by autophosphorylation, and as the stimuli subsides BA-Stp1 serves to inactive the kinase by dephosphorylation. Further studies of the phosphatase-kinase association between BA-Stp1 and BA-Stk1 will provide new insights into identification of novel target substrates and the importance of reversible phosphorylation as a regulatory mechanism in B. anthracis.
We thank Ken Jackson and Stephen Snow at the Molecular Biology-Proteomics Facility (OUHSC) for their help with the mass spectrometric analysis of BA-Stk1.
This work was supported by Public Health Service grant U54A1057156, Western Region Center for Biodefense and Emerging Infectious Diseases (to J.D.B. and C.R.L.).
Published ahead of print on 13 November 2009.