Autoregulatory Characteristics of a Bacillus anthracis Serine/Threonine Kinase

Abstract

BA-Stk1 is a serine/threonine kinase (STK) expressed by Bacillus anthracis. In previous studies, we found that BA-Stk1 activity is modulated through dephosphorylation by a partner phosphatase, BA-Stp1. In this study, we identified critical phosphorylation regions of BA-Stk1 and determined the contributions of these phosphodomains to autophosphorylation and substrate phosphorylation. The data indicate that BA-Stk1 undergoes trans-autophosphorylation within a regulatory domain, referred to as the activation loop, which carries eight putative regulatory serine and threonine residues. We identified activation loop mutants that impacted kinase activity in three different manners: regulation of autophosphorylation (T162), regulation of substrate phosphorylation (T159 and S169), and regulation of overall kinase activity (T163). Tandem mass spectrometry (MS/MS) analysis of the phosphorylation profile of each mutant revealed a second site of phosphorylation on the kinase that was influenced by the phosphorylation status of the activation loop. This second region of the kinase contained a single phosphorylation residue, S214. Previous work has shown S214 to be necessary for downstream substrate phosphorylation, and we have shown that this residue is subject to dephosphorylation by BA-Stp1. These findings indicate a connection between the phosphorylation status of the activation loop and phosphorylation of S214, and this suggests a previously undescribed model for how a bacterial STK shifts from a state of autophosphorylation to targeting downstream substrates.

INTRODUCTION

The lifestyle of Bacillus anthracis is complex. B. anthracis resides in the soil but also grows to large numbers in the bloodstreams of infected animals and can invade solid tissues (9, 27). B. anthracis infects its host as a spore, germinates to cause disease, and sporulates following death of the host. Intertwined within these events, B. anthracis exists in intracellular and extracellular environments and coordinates expression of virulence factors necessary to cripple host immune responses (8, 16, 30). B. anthracis is among the most complicated bacterial pathogens in regards to the number of environmental conditions to which the organism must respond and the transitions the organism goes through during disease. For these reasons, B. anthracis is a useful model to help better understand the intracellular signaling networks that bacterial pathogens use to sense and adapt to changes in their environment. The current understanding of B. anthracis signaling pathways that regulate these transitions and adaptations to changes in environmental conditions is limited to studies on histidine kinases. For example, the Spo0 histidine kinase phosphorelay signal transduction pathway regulates transitions from vegetative bacilli to spores (3, 35). Other histidine kinase signaling pathways include the stress response pathway, mediated by σ^B (13), and the ResDE two-component respiration pathway (38, 41). However, little is known about the role of reversible serine/threonine phosphorylation systems in B. anthracis.

The current work examines a B. anthracis eukaryote-like serine/threonine kinase (STK) which our group previously found to be subject to regulation by a cognate phosphatase (37). To date, STKs have been shown to regulate growth and/or virulence of bacterial pathogens such as Streptococcus agalactiae (31, 32), Streptococcus pneumoniae (10), Streptococcus pyogenes (20), Enterococcus faecalis (22), and Mycobacterium tuberculosis (6). Although the family of bacterial STKs is homologous across several genera of bacteria, each microbe has coopted its STK to regulate processes critical to the particular parent organism. Collectively, prokaryotic STKs have been reported to regulate growth and virulence, as well as stress responses (29), gene expression (15, 34), development (24, 28), biofilm formation (18), and metabolism (6, 33), in an organism-specific manner. Furthermore, a recent study by Shakir et al. characterized a serine/threonine kinase-phosphatase pair important for survival of B. anthracis within cultured macrophages (37). The cognate serine/threonine phosphatase (BA-Stp1) in this pair was found to modulate kinase activity by dephosphorylating phospho-residues on B. anthracis Stk1 (BA-Stk1).

STKs are temporally regulated by phosphorylation, occurring in trans (autophosphorylation) or through modification by upstream kinases (21, 23). Bacterial STKs contain motifs that are subject to phosphorylation, which can determine the activation state of the kinase. A well-studied example of such a regulatory domain is an activation loop found in kinases from both eukaryotic and prokaryotic species (2, 17, 25). This activation domain, defined as the region between the conserved motifs DFG and APE, is located within close proximity to the catalytic loop (17). Phosphorylation of target residues in the activation loop stabilizes a catalytic Asp residue and promotes the binding of ATP, divalent cations, and substrates (17). Phosphorylation of residues within the activation loop has been implicated in regulating kinase activity in M. tuberculosis PknB and Bacillus subtilis PrkC (2, 25). Amino acid substitutions that render the activation loops of PknB and PrkC resistant to phosphorylation decrease overall kinase activity.

Interestingly, a comparison of the putative activation loop of BA-Stk1 with those of STKs from other bacterial pathogens identified 8 serine and threonine residues within the proposed activation loop. For comparison, PknB has 4 such residues, and many eukaryotic kinases, such as cAPK, contain only 2 (21). This observation led us to hypothesize that BA-Stk1 utilizes the activation loop to tightly control transitions from autophosphorylation to substrate phosphorylation.

In the current work, we demonstrate that BA-Stk1 is critical to growth and survival of B. anthracis in a model of infection. We show that BA-Stk1 is autophosphorylated via an intermolecular interaction and that the putative phosphorylation status of serine and threonine residues within the activation loop determines if the kinase targets an exogenous substrate or promotes autophosphorylation. Finally, we report the first link between the phosphorylation status of the activation loop and phosphorylation of a distal residue (S214) on the protein, which appears to shift the kinase from autophosphorylation to phosphorylation of a substrate.

MATERIALS AND METHODS

Bacterial strains, cell lines, and reagents.

Standard reagents were purchased from Sigma, unless otherwise noted. B. anthracis Sterne strain 7702 (obtained from Theresa Koehler) (4) was used as the parent strain for mutant construction. Abelson murine leukemia virus-transformed murine macrophages derived from ascites of BALB/c mice (termed RAW 264.7 cells) were obtained from the American Type Culture Collection (ATCC). RAW 264.7 cells were maintained in tissue culture-grade T-75 flasks grown in the presence of Dulbecco's modified Eagle's medium (DMEM) (ATCC) supplemented with 10% fetal bovine serum (FBS; ATCC). Cells were maintained in a humidified incubator in the presence of 6.0% CO₂ at 37°C and passaged every 48 to 72 h at the point of confluence.

Generation of Δba-stk1 strain.

Overlap extension PCR was used to construct a DNA fragment consisting of apha-3 flanked by regions of DNA homologous to upstream and downstream regions of ba-stk1. apha-3, which provides antibiotic resistance to kanamycin, was amplified by a PCR using the Ω km-2 cassette as template DNA, with the following oligonucleotides: 5′-TGCTCTAGAGAAGAGGATGAGGAGGCAGATTGCC-3′ and 5′-TGCTCTAGAGCTCGGGACCCCTATCTAGCGA-3′. A 1,519-bp region upstream of ba-stk1 (nucleotides 3680075 to 3681594 of B. anthracis Sterne [GenBank accession no. AE017225]) was amplified from genomic DNA by a PCR using the following oligonucleotides: 5′-TGAATATGGAAATTCCATTTGTG-3′ (primer 1) and 5′-TCCTCATCCTCTTCTCTAGAGCATGCACTTCACCTACTTTCGTTTG-3′. A 1,500-bp region downstream of ba-stk1 (nucleotides 3676597 to 3678095 of B. anthracis Sterne [GenBank accession no. AE017225]) was amplified by PCR with the following oligonucleotides: 5′-TAGGGGTCCCGAGCTCTAGAGCATACATTCCACTTTCTTTCTAGAG-3′ and 5′-TGTGGATTTAATACAACTCCTGC-3′ (primer 6). Thermal cycling conditions consisted of an initial denaturation at 95°C for 5 min and 30 cycles of 94°C for 30 s, 53 to 55°C for 45 s, and 68°C for 1.5 min, with a final extension at 68°C for 10 min. Using primer 1, primer 6, and the aforementioned PCR products, a DNA fragment was generated using a PCR that combined all three products together. Thermal cycling conditions were as follows: initial denaturation at 94°C for 5 min; 3 cycles of 94°C for 30 s, 40°C for 30 s, and 68°C for 4.25 min; 10 cycles of 94°C for 45 s, 45°C for 45 s, and 68°C for 4.25 min; 15 cycles of 94°C for 45 s, 50°C for 45 s, and 68°C for 4.25 min; 10 cycles of 94°C for 45 s, 55°C for 45 s, and 68°C for 4.25 min; and a final extension at 68°C for 10 min. A 4.2-kb DNA fragment produced by overlap extension PCR was cloned into pGEM-T Easy (Promega). The fragment was subcloned into pUTE-568 (5), which has origins of replication for both Escherichia coli and B. anthracis. pUTE-568 also confers resistance to erythromycin (5 μg/ml), which allows for selection of transformants that have undergone a double homologous recombination event. E. coli JM110 was transformed with the pUTE-568-derived construct (pK110) to obtain unmethylated DNA. Unmethylated pK110 was electroporated into B. anthracis Sterne, and transformants were selected on solid medium containing kanamycin (100 μg/ml). Following antibiotic selection, colonies were inoculated into brain heart infusion (BHI) broth without antibiotics, aerated at 37°C, and transferred to fresh antibiotic-free medium four times a day for 3 days. The cultures were screened for clones demonstrating kanamycin resistance and erythromycin sensitivity. Isolates in which a double-crossover event had occurred were confirmed by PCR, using oligonucleotides specific for apha-3 and upstream of ba-stk1. The resulting PCR product was sequenced and analyzed to verify that ba-stk1 had been replaced with apha-3.

Growth analysis of B. anthracis.

Stationary-phase (optical density at 600 nm [OD₆₀₀] of >2.0) cultures of B. anthracis were diluted into approximately 500 μl of BHI broth and adjusted to an OD₆₀₀ of 0.01. Next, 1 μl of diluted culture was used to inoculate 200 μl of BHI broth in the wells of a 100-well Honeycomb 2 (Thermo Electron Corp) microplate. Growth was analyzed by monitoring the OD₆₀₀ every 30 min for 12 h in a Bioscreen plate reader (Bioscreen C) with constant shaking at 37°C. Samples were analyzed in triplicate, and the mean absorbance was determined using the statistical module of Excel.

Preparation of B. anthracis spores.

Spores were prepared as described by McKevitt et al. (26). Briefly, cultures of B. anthracis Sterne 7702 and B. anthracis Sterne 7702 Δba-stk1 (grown in the presence of 100 μg/ml kanamycin) in BHI broth were grown to mid-log phase (OD₆₀₀, 0.8). Next, 500 μl of each culture was inoculated into 50 ml of 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). Sporulation cultures were grown at 30°C with constant shaking for 48 h. Subsequently, cultures were centrifuged at 3,000 × g, the supernatant was removed, and the spore pellet was resuspended in 10 ml of sterile double-distilled water (ddH₂O). Cultures were incubated at 30°C with constant shaking for 24 h. The cultures were passed through two inline glass microfiber syringe filters (3.1 μm and 1.2 μm) (VWR) to remove vegetative organisms from the spore suspension. Spores were pelleted by centrifugation at 3,000 × g, resuspended in 1 ml of sterile ddH₂O, and stored at −20°C. Prior to use, all spore suspensions were diluted to working concentrations necessary for corresponding experiments and heated to 65°C for 25 min in order to kill any residual vegetative organisms as well as to heat activate the spores. CFU were determined by dilution plating on BHI agar.

Monitoring germination by using changes in the refractive index of the spore.

Germination was determined by observing the drop in refractive index (OD₆₀₀) that occurs during the transition from spores to vegetative bacilli. Spores (3 × 10⁷) were resuspended in 200 μl of medium (phosphate-buffered saline [PBS] with 100 mM l-alanine, 0.5 mM l-alanine and 1.0 mM inosine, 0.5 mM l-alanine and 5 mM l-tryptophan, 1 mM inosine and 5 mM l-tryptophan, or 0.5 mM l-alanine and 50 mM l-tryptophan) in the wells of a 100-well Honeycomb 2 (Thermo Electron Corp) microplate. Absorbance was monitored every 90 s for 30 min in a Bioscreen plate reader (Bioscreen C) with constant shaking at 37°C. Samples were analyzed in duplicate, and the mean absorbance was determined using Excel software.

Spore infection of RAW 264.7 cells.

RAW 264.7 cells were cultured in 24-well plates (5 × 10⁵ cells/well) overnight in 0.5 ml of medium (DMEM and 10% FBS) prior to infection, resulting in a confluent monolayer of cells. Macrophages were infected at a multiplicity of infection (MOI) of 10 for 30 min, followed by treatment with gentamicin (10 μg/ml) for an additional 30 min. Macrophages were subsequently washed 10 times with medium. At the indicated time points, macrophages were harvested, serially diluted in PBS, and plated on Luria-Bertani (LB) agar. Numbers of CFU were determined for quantification of B. anthracis survival and growth within the macrophages. Student's t test was performed using the statistical module of Prism.

RT-PCR.

Total RNAs were isolated from B. anthracis lag-phase (OD₆₀₀, 0.1), log-phase (OD₆₀₀, 0.6), and early-stationary-phase (OD₆₀₀, 1.5) cultures by use of an RNeasy minikit (Qiagen) following the manufacturer's protocol. This included on-column DNase digestion with RNase-free DNase (Qiagen). Next, the RNA was treated with another DNase digestion with RQ DNase (Promega) for 1 h at 37°C. RNA was further purified using the RNeasy minikit RNA cleanup protocol. To verify that RNA was not contaminated with DNA, PCR was performed using 16S rRNA universal primers on the isolated RNA fraction. For quality control, RNA samples were glyoxylated and analyzed by agarose gel electrophoresis, using a NorthernMax-Gly system (Ambion), and the quality of representative 16S and 23S rRNA bands was determined by visualization. RNA was converted into cDNA by use of a SuperScript II reverse transcriptase (RT) kit (Invitrogen). Briefly, 100 ng of RNA was incubated with 250 ng of random oligonucleotide primers (Invitrogen) and 1 μl deoxynucleoside triphosphate (dNTP) mix (10 mM [each] dATP, dGTP, dCTP, and dTTP) at 65°C for 5 min and then chilled quickly on ice. Next, 4 μl of 5× first-strand buffer (250 mM Tris-HCl, 375 mM KCl, 15 mM MgCl₂), 2 μl 0.1 M dithiothreitol (DTT), and 1 μl RNaseOUT were added to the RNA mixture. The sample was then incubated at 42°C for 2 min, at which point 200 units of SuperScript II RT was added. The sample was next incubated at 25°C for 10 min, 42°C for 50 min, and then 70°C for 15 min. Two units of RNase H was added, followed by incubation at 37°C for 20 min to remove any RNA complementary to the cDNA. Enzymes and buffers were removed using a MiniElute PCR purification kit (Qiagen). cDNA fragments were amplified by PCR, using the following primers specific to the genes of interest: for ba-stk1, 5′-GTACCAGGACAAACATTGAAG-3′ and 5′-CTGCTCATTTATCCTCTCTGG-3′; for the cotranscript of ba-stk1 and ba-stp1, 5′-GAAGACGCAGAATATCACCCG-3′ and 5′-GATGCGACAACGTTGTAACAG-3′; and for the cotranscript of ba-stk1 and bas3712, 5′-GCGACATTGTGTAATCCCTTC-3′ and 5′-GAACAATTTCAGAGTCGGCAAA-3′. Thermal cycling conditions consisted of an initial denaturation at 95°C for 5 min and 30 cycles of 94°C for 30 s, 49 to 59°C for 45 s, and 68°C for 30 s, with a final extension at 68°C for 10 min. PCR products were analyzed by agarose gel electrophoresis. Controls included a positive control with genomic DNA and negative controls with either water or RNA as a template.

Expression and isolation of recombinant BA-Stk1_cat and BAS3712.

The gene fragment encoding the putative catalytic domain of BA-Stk1 (BA-Stk1_cat) was amplified by PCR from B. anthracis Sterne genomic DNA, using the following oligonucleotides: 5′-GGGATTCATATGGTGCTGATTGGAAAAGCGTTAAATG-3′ and 5′-CGGATCCTTGTCGCTTCCATATCTTCCGG-3′. The gene encoding BAS3712 was also amplified by PCR from B. anthracis Sterne genomic DNA, using the following oligonucleotides: 5′-GGGATTCATATGCCAGAAGGAAAAATTG-3′ and 5′-CGGATCCCTAATACCTCGGCTTTCTC-3′. The primers were designed to incorporate NdeI and BamHI restriction enzyme sites, respectively. The resulting PCR product and pET-15b (Promega) were digested with the NdeI and BamHI restriction enzymes (New England BioLabs). The PCR product was ligated with pET-15b, using T4 DNA ligase (Roche). The pET-15b-derived construct (BA-Stk1_cat-pET15b) was transformed into Escherichia coli BL21(DE3) cells (Novagen) for protein expression. E. coli cultures were incubated at 37°C until an OD₆₀₀ of 0.8 was reached, at which point expression was induced with 0.1 mM isopropyl-β-d-thiogalactopyranoside (Denville Scientific) for 4 h at 37°C. Protein was purified using nickel affinity chromatography according to the manufacturer's protocol (Novagen).

Measurement of BA-Stk1_cat autophosphorylation and substrate phosphorylation.

Kinase autophosphorylation activities of 1.25 μM wild-type BA-Stk1_cat and site-directed mutants were assayed in reaction buffer (4 mM MOPS [morpholinepropanesulfonic acid], pH 7.2, 5 mM MnCl₂, 0.05 mM DTT) containing 1 μCi of [γ-³²P]ATP. The reaction mixtures were incubated for 30 min at 37°C, reactions were stopped using 4× SDS-Laemmli buffer, and the samples were 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 phosphorimager analysis. Kinase phosphorylation of myelin basic protein (MBP) or BAS3712 (putative GTPase) was assayed by the addition of 20 μM substrate to the kinase reaction mixture containing reaction buffer, ATP, and kinase. Samples were analyzed in the same way as that for the autophosphorylation experiments.

Measurement of BA-Stk1_cat intermolecular autophosphorylation.

Serial dilutions of BA-Stk1_cat were incubated in reaction buffer (4 mM MOPS, pH 7.2, 5 mM MnCl₂, 0.05 mM DTT) containing 1 μCi of [γ-³²P]ATP. The reaction mixtures were incubated for 2 h at 37°C. Following incubation, equal amounts (moles) of BA-Stk1_cat from each reaction mixture were spotted onto phosphocellulose paper and immersed in 75 mM phosphoric acid to terminate the reaction. The phosphocellulose paper was washed gently with phosphoric acid five times. Radioactivity was measured by liquid scintillation spectrometry.

Modeling of BA-Stk1 activation loop based on that of a homologous protein.

Three-dimensional (3D) modeling was performed using SWISS-MODEL software. The N-terminal 290-amino-acid fragment of BA-Stk1 was used as the target and was modeled on the A chain of PknB (Protein Data Bank [PDB] accession number 1o6yA) from M. tuberculosis (42).

Generation of BA-Stk1 activation loop mutants.

Mutagenesis of serine and threonine residues was performed using a QuikChange site-directed mutagenesis kit (Stratagene) according to the manufacturer's instructions. Briefly, 50 ng of pET-15b-BA-Stk1 was used as a template. Point mutations were inserted using synthetic oligonucleotide pairs designed according to the manufacturer's instructions. Mutations were confirmed by DNA sequencing.

MS analysis of BA-Stk1 activation loop mutants.

Mass spectrometry (MS) analysis was performed in the Molecular Biology Proteomics Facility of the University of Oklahoma Health Sciences Center. Following SDS-PAGE and staining of BA-Stk1_cat and activation loop mutants, protein bands were excised, digested with trypsin within the gel, utilizing a 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-MS (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 and 150-mm length; packed with 3-μm particles) at a flow rate of 200 nl/min. For LC-MS-MS, data were collected in the information-dependent acquisition mode over a mass range of 300 to 2,500 m/z. Peptides with a +2 to +4 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 2,500 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, 5 May 2010) by utilizing the MASCOT search engine (version 2.2), both of which are installed on an on-site MASCOT server. The search parameters consisted of a 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, N-terminal Q-to-pyro-Glu modification, and phosphorylation of serine and threonine were chosen as variable modifications. Protein identifications were made based upon MASCOT Mowse scores above the significance threshold value (P < 0.05). The elution points and relative ionization intensities of phosphorylated peptides and their nonphosphorylated versions were determined by LC-MS analysis over a mass range of 300 to 2,500 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 peptides.

Statistical analysis.

The statistical modules of Excel and Prism were used to perform Student's t test and analysis of variance (ANOVA) on the appropriate data sets.

RESULTS

Characterization of ba-stk1.

An inspection of the GenBank-accessible B. anthracis Sterne genome sequence revealed four genes encoding putative STKs. Of particular interest was the BAS3713 gene (GenBank accession number AAT5601), termed ba-stk1 herein, which shares identity with previously studied genes encoding STKs found in Gram-positive bacteria such as S. agalactiae, B. subtilis, and M. tuberculosis (7, 11, 14, 20, 24, 31, 33).

As summarized in Fig. 1, BA-Stk1 is homologous to PrkC (B. subtilis), STK1 (S. agalactiae), and PknB (M. tuberculosis), all of which are Hanks-like kinases. BA-Stk1 shares 68% sequence similarity with PrkC from B. subtilis, including residue K40, which is critical for PrkC activity (24). Other conserved regions of BA-Stk1 which are indicative of STKs are residues 132 to 139 and 152 to 154, which are designated the catalytic loop and the divalent cation binding region, respectively (17). Furthermore, each of the proteins shown in Fig. 1 contains a series of PASTA (penicillin binding protein-associated serine/threonine kinase-associated) domains located within the putative extracellular C terminus.

Fig. 1.

Sequence alignment of BA-Stk1 and other known STKs. The amino acid sequence of the enzymatic domain of BA-Stk1 is aligned with those of PrkC (B. subtilis), Stk1 (S. agalactiae), and PknB (M. tuberculosis). The putative catalytic domain is highlighted in gray, and the putative activation domain is underlined. Conserved residues characteristic of Hanks-like kinases are shown in bold. Stars indicate identity, and single and double dots indicate lower and higher degrees of similarity.

Figure 2 provides an overview of the genomic location of ba-stk1. As shown in Fig. 2A, ba-stk1 is the 3′ open reading frame in a region predicted to be an eight-gene operon based on the absence of upstream transcriptional terminators. Upstream genes in this region include a PP2c phosphatase gene (ba-stp1), a hypothetical protein gene (bas3715), a Sun protein gene (sunL), a methionyl-tRNA formyltransferase gene (fmt), a polypeptide deformylase gene (priA), a primosomal protein N′ gene (bas3719), and a phosphopantothenoylcysteine decarboxylase/phosphopantothenate-cysteine ligase gene (bas3720).

Fig. 2.

Expression analysis of ba-stk1. (A) The putative operon is 10.22 kb long and carries eight open reading frames, including ba-stk1 as the 3′-terminal gene in the operon. (B) RT-PCR analysis of ba-stk1 (primers 1 and 2) expression during lag-phase (lanes 2 and 3), log-phase (lanes 4 and 5), and early-stationary-phase (lanes 6 and 7) growth of Bacillus anthracis Sterne. Reactions were performed using cDNA (lanes 3, 5, and 7) or RNA (lanes 2, 4, and 6) as a template. (C) Cotranscription of ba-stk1 and ba-stp1. PCR products were amplified using primers specific to ba-stk1 and ba-stp1 (primers 3 and 4). Reactions were performed using cDNA (lanes 2, 3, and 4), no DNA (lane 5), or genomic DNA (lane 6) as a template. (D) Cotranscription of ba-stk1 and bas3712. Reactions were performed using primers specific to ba-stk1 and bas3712 (primers 5 and 6), using cDNA (lane 2), no DNA (lane 3), or genomic DNA (lane 4) as a template. Lanes 1 (B, C, and D), 1-kb DNA ladder.

RT-PCR analysis was performed on ba-stk1 and proximal genes in order to confirm that the kinase is expressed in B. anthracis and to demonstrate the polycistronic expression of ba-stk1 with other genes within this region of the chromosome. RNAs were isolated from the lag, log, and early stationary phases of growth, and cDNA synthesis was carried out as described in Materials and Methods. As shown in Fig. 2B, products were amplified using primers specific to ba-stk1 (primers 1 and 2 in Fig. 2A), indicating that ba-stk1 was expressed during growth of B. anthracis. To investigate whether ba-stk1 was cotranscribed with ba-stp1, PCR was performed using primers specific to ba-stk1 and ba-stp1 (primers 3 and 4 in Fig. 2A). As shown in the image of the DNA-containing agarose gel in Fig. 2C, cDNA representing the cotranscribed message was detected at all stages of growth. Analysis of the sequence located between ba-stk1 and bas3712 revealed a putative transcriptional terminator. Therefore, to determine if ba-stk1 was the terminal gene of the putative operon, PCR was performed using primers specific to ba-stk1 and the downstream gene, annotated bas3712 (primers 5 and 6 in Fig. 2A), using cDNA from log-phase growth. As shown in the image of the DNA-containing agarose gel in Fig. 2D, a cotranscribed product derived from ba-stk1 and bas3712 could not be detected. These data indicate that ba-stk1 is expressed in B. anthracis and is likely the 3′ gene in the putative operon shown in Fig. 2A. Interestingly, this organization differs from that in B. subtilis, where the phosphatase/kinase module prpC-prkC is cotranscribed with the downstream gene yloQ, encoding an essential GTPase (19). Unlike the intergenic region between ba-stk1 and bas3712 in B. anthracis, this intergenic region between prkC and yloQ does not carry a transcriptional terminator in B. subtilis.

Contribution of BA-Stk1 to B. anthracis physiology and pathogenesis.

In a previous study, our lab generated a ba-stp1/ba-stk1 double-knockout mutant and investigated the contribution of this signaling module to B. anthracis physiology and pathogenesis (37). However, the impact of a single mutation resulting in only the loss of ba-stk1 was not determined. To analyze the contribution of BA-Stk1 to B. anthracis physiology and pathogenesis, a ba-stk1 null mutant was generated in the current study. Verification of the knockout strain was determined via Western blotting and RT-PCR (data not shown). Furthermore, RT-PCR revealed no disruption of bas3712, the gene downstream of ba-stk1 (data not shown).

Experiments comparing the growth, germination, and sporulation of B. anthracis Δba-stk1 with those of the parent strain were conducted. The data revealed no discernible difference in growth rates or sporulation efficiencies between B. anthracis Sterne and B. anthracis Δba-stk1 under nutrient-rich conditions (data not shown). For B. anthracis, five distinct germination pathways have been characterized (alanine, alanine and proline, aromatic amino acid-enhanced alanine, AAID-1, and AAID-2 pathways) (12). Experiments were performed using germinants from each of the known pathways. Similar to the case for growth and sporulation effects, there was no discernible difference in the germination rates of B. anthracis Sterne and B. anthracis Δba-stk1 in vitro (data not shown).

B. anthracis spore-macrophage interactions are thought to be critical to the establishment of disease (16). Therefore, we analyzed the germination and survival of B. anthracis Δba-stk1 in cultured macrophages. RAW 264.7 macrophage-like cells were infected with either B. anthracis Δba-stk1 spores or B. anthracis Sterne 7702 spores at an MOI of 10. At the indicated time points (1.5, 7.5, and 13.5 h postinfection), macrophage lysates were collected, and the CFU from the parent strain and mutant infections were determined. As shown in Fig. 3, the survival of B. anthracis Δba-stk1 was significantly reduced compared to that of the parent strain. These data suggest that BA-Stk1, while not critical for in vitro growth, sporulation, or germination, may function as an infection-specific kinase that contributes to growth and survival within the host.

Fig. 3.

Analysis of growth and survival of B. anthracis Δba-stk1 within cultured macrophages. RAW 264.7 cells were infected at an MOI of 10. At the indicated time points postinfection, macrophages were harvested, serially diluted in PBS, and plated on LB agar. Numbers of CFU were determined for quantification of B. anthracis survival and growth within the macrophages. Black bars, B. anthracis Sterne; gray bars, B. anthracis Δba-stk1. The P value was 0.0016 for the 7.5-h time point and <0.0001 for the 13.5-h time point, as determined by Student's t test. The data are from a representative experiment of three separate spore preparations. Experiments were performed in triplicate, and error bars represent the standard errors of the means.

Analysis of BA-Stk1 kinase activity.

Our previous studies have shown that BA-Stk1 is a functional kinase capable of phosphorylating MBP and undergoing autophosphorylation (37). Whether this autophosphorylation event occurs in cis or in trans has not been addressed for this kinase. In the current study, an experiment was carried out to determine whether BA-Stk1 autophosphorylation occurs via an intermolecular or intramolecular mechanism. To accomplish this, a range of BA-Stk1 concentrations were analyzed for autophosphorylation, with the concentrations of [γ-³²P]ATP and buffer kept constant and only the concentration of kinase changed in the reaction mixtures. Following incubation, equal amounts (moles) of kinase from each reaction mix were spotted on phosphocellulose strips and washed with phosphoric acid to remove reaction buffer and unincorporated ³²P. Radioactivity of the membrane was then measured by liquid scintillation spectrometry. If autophosphorylation occurred in cis, we expected the amount of incorporated ³²P per mole of protein to remain the same, even when the kinase was diluted. However, as shown in Fig. 4, dilution of BA-Stk1 resulted in a corresponding decrease in kinase activity. Our analysis of kinase autophosphorylation suggests that BA-Stk1 is autophosphorylated primarily through an intermolecular interaction and does not rely solely on cis phosphorylation. However, it should be noted that we did not detect a statistically significant decrease in autophosphorylation for each individual BA-Stk1 concentration. Thus, while the overall data support a mechanism of trans-autophosphorylation, it is possible that cis-autophosphorylation occurs within a narrow concentration range of the kinase.

Fig. 4.

Intermolecular kinase activity. Serial dilutions of BA-Stk1_cat were incubated in kinase reaction buffer with [γ-³²P]ATP for 120 min. Following incubation, equal amounts (moles) of kinase from each reaction mix were spotted onto phosphocellulose paper and washed with phosphoric acid. Radioactivity was measured by liquid scintillation spectrometry. Experiments were performed in duplicate. Error bars represent standard errors of the means. Statistical analysis was performed by ANOVA (P < 0.05) and unpaired two-tailed Student's t test (P < 0.05), as indicated by asterisks.

Kinase activities of BA-Stk1 activation loop mutants.

Previous studies found that BA-Stk1's kinase activity could be modulated through dephosphorylation by a partner phosphatase, BA-Stp1 (37). These earlier findings, along with the observation that BA-Stk1 can autophosphorylate, suggested that the kinase modulates its activity by differential phosphorylation of critical residues. To investigate this possibility, we examined BA-Stk1 for phosphorylation sites in putative regulatory regions of the protein.

Similarities between BA-Stk1 and other bacterial STKs suggest that BA-Stk1 is a transmembrane protein with PASTA domains located outside the cell membrane and an enzymatic domain located within the cell. This is analogous to the case for the well-studied eukaryotic receptor protein tyrosine kinases. Dimerization of receptor protein tyrosine kinases has been shown to stimulate intermolecular phosphorylation of residues within the activation loop, stimulating kinase activity (40). The activation segment, located in the center of the kinase domain, is defined as the region between conserved sequences DFG and APE and contains residues whose phosphorylation state influences kinase activity (21). A similar activation segment has also been described for some representative bacterial STKs (2, 25, 42). Thus, we examined the sequence of BA-Stk1 and found a candidate activation segment which lies between residues 152 and 180. We also developed a model of the BA-Stk1 three-dimensional structure, using the previously reported structure of PknB as a scaffold (42; data not shown). This theoretical structure of BA-Stk1 suggested that the activation loop was an alpha helix located between residues 156 and 169. Importantly, a previous study by our lab identified, via LC-MS-MS, a phosphopeptide spanning this region of BA-Stk1 (37). This phosphopeptide was predicted to contain seven phosphate groups, as determined by a search of the NCBI nonredundant protein database (Firmicutes group). These observations suggested that the 156–169 region of BA-Stk1 is highly phosphorylated and, as such, could play an important role in regulation of the kinase.

To determine the contributions of activation loop phosphorylation sites to BA-Stk1 kinase activity, alanine mutants of T157, T159, S160, T162, T163, T165, T167, and S169 were analyzed for autophosphorylation. Also, as a control, an alanine mutant of the critical residue K40 was generated. K40 is an invariant residue found in all STKs and provides the proper orientation of ATP for hydrolysis by interacting with the α and β phosphates of the molecule (17). As shown in Fig. 5A, the T157A, T159A, T165A, T167A, and S169A mutants exhibited decreased autophosphorylation activity, whereas the T162A and T163A mutants exhibited increased activity.

Fig. 5.

Kinase activities of BA-Stk1 activation loop mutants. (A) Autophosphorylation activities. (B) MBP phosphorylation activities. Purified wild-type BA-Stk1 and alanine mutants of BA-Stk1 were incubated with MBP in the presence of [γ-³²P]ATP for 30 min in kinase buffer. The reactions were stopped with SDS loading buffer, and the reaction products were analyzed by SDS-PAGE. Levels of ³²P incorporation were obtained by phosphorimager analysis, and the relative values are depicted as percentages of wild-type phosphorylation. Error bars represent standard deviations of the means. Statistical analysis was performed by ANOVA (P < 0.0001).

The activation loop mutants were also analyzed for substrate phosphorylation activity. As shown in Fig. 5B, the T157A, T162A, T165A, and T167A mutants exhibited decreased substrate phosphorylation compared to the wild type. However, the T159A, T163A, and S169A mutants exhibited increases in kinase activity. These data suggest that phosphorylation of the activation loop of BA-Stk1 regulates the extent of differential autophosphorylation and substrate phosphorylation.

Contributions of activation loop residues to the phosphorylation status of BA-Stk1 S214.

Since the phosphorylation status of specific residues within the activation loop influences the kinase activity of BA-Stk1, we predicted that mutation of these residues might also impact the phosphorylation status of other regulatory residues in BA-Stk1. Thus, to address this possibility, we mapped the phosphorylated regions of each activation loop mutant. Phosphorylated BA-Stk1 and alanine mutants were excised from an SDS-PAGE gel and trypsin digested. The phosphopeptide pool was then subjected to LC-MS-MS analysis. The MS-MS spectra identified phosphopeptide BA-Stk1: 208–221, with the sequence QPFSGESAVAIALK. Phosphopeptide BA-Stk1: 208–221 was previously identified in an earlier study by our lab and is predicted to be phosphorylated on one serine residue, S214 (37). Our earlier analysis of BA-Stk1: 208–221 revealed that the peptide is a target for dephosphorylation by BA-Stp1 (37). Moreover, phosphorylation of S214 is critical for substrate phosphorylation (37). The T162A, T165A, and T167A activation loop mutants exhibited decreased levels of phosphorylated peptide BA-Stk1: 208–221, whereas the S169A activation loop mutant exhibited substantially increased levels (Fig. 6). The kinase activity of each activation loop mutant and the impact on S214 phosphorylation are summarized in Table 1. These data, along with our previous findings, suggest that the phosphorylation status of the activation loop regulates phosphorylation of downstream targets through S214.

Fig. 6.

Contributions of activation loop residues to the phosphorylation status of BA-Stk1 S214. BA-Stk1_cat and activation loop mutants were excised from an SDS-PAGE gel and trypsin digested. Isolated phosphopeptide BA-Stk1: 208–221 was subjected to LC-MS-MS analysis. The relative ion intensities of the differentially phosphorylated forms (either no phosphoamino acid or one phosphoamino acid) are indicated.

Table 1.

Summary of BA-Stk1 activation loop mutant activities and their impacts on S214 phosphorylation status

Activation loop mutant	Autophosphorylation (% of WT level)	MBP phosphorylation (% of WT level)	Change in S214 phosphorylation status
T157A	70	45	No change
T159A	42	124	No change
S160A	86	92	No change
T162A	129	56	Decrease
T163A	137	141	No change
T165A	69	32	Decrease
T167A	59	46	Decrease
S169A	46	156	Increase

Results from the previous experiments suggested that activation loop residue S169 may function in a novel mechanism for regulating substrate phosphorylation. When S169 could not be phosphorylated due to an alanine mutation, substrate phosphorylation and peptide BA-Stk1: 208-221 phosphorylation were enhanced (Fig. 5B and 6E). These data suggest that phosphorylation of S169 may inhibit the transition to substrate phosphorylation. Thus, to address this possibility, we simulated phosphorylation of S169 with an aspartic acid substitution. Mimicking phosphorylation of S169 reduced MBP phosphorylation (Fig. 7A). To further investigate the contribution of S169 to substrate phosphorylation, we chose to use BAS3712 as a substrate for kinase phosphorylation. BAS3712 is a putative GTPase whose gene is located downstream from the BA-Stk1 gene, and moreover, a B. subtilis homolog of this protein has been identified as a substrate for PrkC (1). As shown in Fig. 7B, mimicking phosphorylation of S169 also reduced phosphorylation of the putative GTPase, further suggesting that the addition of a phosphate group to this residue has a negative effect on substrate phosphorylation.

Fig. 7.

Kinase activities of BA-Stk1 S169 mutants. Purified BA-Stk1 and mutants were incubated with MBP (A) or GTPase (B) in the presence of [γ-³²P]ATP for 30 min in kinase buffer. The reactions were stopped with SDS loading buffer, and the reaction products were analyzed by SDS-PAGE. Levels of ³²P incorporation were obtained by phosphorimager analysis, and the relative values are depicted as percentages of wild-type phosphorylation. Error bars represent standard errors of the means.

DISCUSSION

Findings from the current study provide insight into a Hanks-like serine/threonine kinase (BA-Stk1) expressed in the spore-forming pathogen B. anthracis. The data indicate that BA-Stk1 is a serine/threonine kinase that exhibits intermolecular trans-phosphorylation and that the enzymatic activities of this kinase are determined by the phosphorylation state of several residues in a putative activation loop of the protein. Unlike other STKs described for prokaryotes, BA-Stk1 exhibits a higher-order level of autoregulation, where phosphorylation of specific residues increases activity and phosphorylation of other residues reduces kinase activity. To our knowledge, this is the first reported example of inhibitory phosphorylation in a bacterial STK, although similar systems have been described for eukaryotic STKs (39). The data also reveal a third level of regulation, where the phosphorylation status of S214, a residue critical for phosphorylation of exogenous substrate, is determined by the phosphorylation state of the activation loop. Collectively, the data support the model summarized in Fig. 8, where the extent of autophosphorylation within the activation loop influences the phosphorylation of S214 and the transition to modification of downstream substrates.

Fig. 8.

Contributions of activation loop residues to BA-Stk1 activities and S214 phosphorylation. This model was derived from a series of experiments using a combination of mutagenesis approaches and kinase assays to determine the role of each residue in regulating BA-Stk1 activity. Residues T162, T165, T167, and S169 influence the phosphorylation status of S214, which further impacts substrate phosphorylation. Arrows indicate the contributions of phosphorylated residues to autophosphorylation and substrate phosphorylation. Overall kinase activity is measured in levels of autophosphorylation and substrate phosphorylation.

For this study, we chose to evaluate the contribution of each putative phosphorylation site within the activation loop to the kinase activity of BA-Stk1. As shown in Fig. 8, the data from our study suggest a regulatory mechanism wherein phosphorylation of particular residues can shift the kinase from an autophosphorylating state to phosphorylation of exogenous substrate. For example, phosphorylation of T162 reduces autophosphorylation but increases substrate phosphorylation. In contrast, phosphorylation of T159 and S169 reduces phosphorylation of exogenous substrate but increases autophosphorylation. Finally, in a third class of residues, substituting alanine for T163 increases both autophosphorylation and substrate phosphorylation, suggesting that phosphorylation of T163 inhibits kinase activity. This finding differs from that for PrkC T163, for which an alanine substitution was found to inhibit kinase activity (25). Four residues (T162, T165, T167, and S169) within the activation loop impacted the phosphorylation state of S214, a residue we previously found to be crucial for phosphorylation of exogenous substrate (37). Alanine substitution at T162, T165, or T167 led to a reduction in S214 phosphorylation in BA-Stk1. Only alanine substitution at S169 increased phosphorylation of S214. This suggested to us that phosphorylation of S169 might be inhibitory for downstream phosphorylation and that this effect is relayed by preventing S214 phosphorylation. By substituting aspartic acid for serine at residue 169, we mimicked phosphorylation, and as predicted, this mutation inhibited phosphorylation of exogenous substrate (Fig. 7). These results support a model wherein autophosphorylation occurs until S169 becomes phosphorylated, and this then serves as a signal to attenuate overall kinase activity.

Based on our experiments characterizing B. anthracis Δba-stk1, BA-Stk1 is not required for in vitro growth under nutrient-rich conditions. We also did not detect significant differences in sporulation or amino acid-induced germination of this mutant. However, compared to the parent strain, B. anthracis Δba-stk1 had a reduced capacity to survive and grow within macrophages (Fig. 3). During the initial stages of macrophage interaction, B. anthracis Δba-stk1 exhibited a similar level of survival; however, during later stages, between 7.5 h and 13.5 h, the mutant did not recover and grow like the parent strain. One possible interpretation of these data is that B. anthracis Δba-stk1 is initially able to survive macrophage engulfment as a spore, but after germination has occurred, it is either more susceptible to macrophage killing or is unable to grow in that environment.

Similar to our earlier findings on B. anthracis Δba-stp1 Δba-stk1, a strain lacking both the kinase and the partner phosphatase, growth and survival of B. anthracis Δba-stk1 within cultured macrophages were significantly reduced (37). Moreover, this phenotype was even more pronounced in the single knockout, suggesting that the presence of the phosphatase without the kinase is more detrimental to bacterial survival within macrophages. Our observations have led us to believe that BA-Stk1 may be more important for modulation of growth and not necessarily for controlling expression of virulence factors. Indeed, Western blot analysis of protective antigen from B. anthracis Δba-stk1 and the parent strain revealed no differences (data not shown). This is in contrast to what is known about S. agalactiae, for which Stk1 and Stk1/Stp1 mutants exhibit a significantly increased 50% lethal dose (LD₅₀), possibly due to the dysregulation of β-hemolysin/cytolysin (31, 32). Similarly, S. pyogenes contains a eukaryote-like serine/threonine kinase-phosphatase pair which is also important for virulence factor expression (20). Further studies will be necessary to determine the host-specific conditions affecting BA-Stk1-dependent growth, but this process does appear to be different from that mediated by B. subtilis PrkC. This B. subtilis homolog of BA-Stk1 is necessary for B. subtilis stationary-phase growth, sporulation, and biofilm formation (14, 24). It is also possible that B. anthracis Δba-stk1 is marginally attenuated in germination under in vivo conditions and that this results in less growth. Previous studies have shown that BA-Stk1 is required for efficient peptidoglycan-induced germination (36), although we did not detect a difference in amino acid-induced germination in our mutant lacking this kinase.

In summary, a model for molecular regulation of BA-Stk1 is starting to emerge, based in part on our previous work with BA-Stp1 and on current results (37). We have identified activation loop residues whose phosphorylation is necessary for maximum autophosphorylation to occur (T157, T159, T165, T167, and S169). Conversely, we have also identified residues that, once phosphorylated, reduce autophosphorylation (T162 and T163). The transition to substrate phosphorylation and the extent of substrate phosphorylation are also dynamic processes. Transition from autophosphorylation to substrate phosphorylation may occur through the phosphorylation of T162, whereas the extent of substrate phosphorylation is determined by phosphorylation of S169. Subsequent phosphorylation at residues such as S169 could temper downstream phosphorylation events. Studies are ongoing to determine which of these residues are targeted for dephosphorylation by BA-Stp1, and such data should provide new insight into how the phosphatase influences the overall activation state of BA-Stk1 and expand our understanding of this critical kinase-phosphatase system in B. anthracis.