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

J. Biol. Chem., Vol. 278, Issue 2, 1174-1185, January 10, 2003

This Article Abstract Full Text (PDF) All Versions of this Article: 278/2/1174    most recent M209052200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ramström, H. Articles by Haiech, J. Search for Related Content PubMed PubMed Citation Articles by Ramström, H. Articles by Haiech, J. Social Bookmarking                      What's this?

Properties and Regulation of the Bifunctional Enzyme HPr Kinase/Phosphatase in Bacillus subtilis^*

Helena Ramström^§, Sarah Sanglier^¶, Emmanuelle Leize-Wagner^¶, Claude Philippe, Alain Van Dorsselaer^¶, and Jacques Haiech^**

From the  Pharmacologie et Physico-Chimie des Interactions Cellulaires et Moléculaires, UMR CNRS 7034, Université Louis Pasteur de Strasbourg, Faculté de Pharmacie, 74 route du Rhin, B.P. 24, F-67401 Illkirch, France and ^¶ Laboratoire de Spectrométrie de Masse Bio-Organique, UMR CNRS 7509, ECPM, Université Louis Pasteur de Strasbourg, 25 rue Becquerel, F-67087 Strasbourg, France

Received for publication, September 4, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION CONCLUSION REFERENCES

The bifunctional allosteric enzyme HPr kinase/phosphatase (HPrK/P) from Bacillus subtilis is a key enzyme in the main mechanism of carbon catabolite repression/activation (i.e. a means for the bacteria to adapt rapidly to environmental changes in carbon sources). In this regulation system, the enzyme can phosphorylate and dephosphorylate two proteins, HPr/HPr(Ser(P)) and Crh/Crh(Ser(P)), sensing the metabolic state of the cell. To acquire further insight into the properties of HPrK/P, electrospray ionization mass spectrometry, dynamic light scattering, and BIACORE were used to determine the oligomeric state of the protein under native conditions, revealing that the enzyme exists as a hexamer at pH 6.8 and as a monomer and dimer at pH 9.5. Using an in vitro radioactive assay, the influence of divalent cations, pH, temperature, and different glycolytic intermediates on the activity as well as kinetic parameters were investigated. The presence of divalent cations was found to be essential for both opposing activities of the enzyme. Furthermore, pH values equal to the internal pH of vegetative cells seem to favor the kinase activity, whereas lower pH values increased the phosphatase activity. Among the glycolytic intermediates evaluated, fructose 1,6-diphosphate and fructose 2,6-diphosphate were found to be allosteric activators in the kinase assay, whereas high concentrations inhibited the phosphatase activity, except for fructose 1,6-diphosphate in the case of HPr(Ser(P)). Phosphatase activity was induced by inorganic phosphate as well as acetyl phosphate and glyceraldehyde 3-phosphate. Kinetic parameters indicate a preference for binding of HPr compared with Crh to the enzyme and supported a strong positive cooperativity. This work suggests that the oligomeric state of the enzyme is influenced by several effectors and is correlated to the kinase or phosphatase activity. The phosphatase activity is mainly supported by the hexameric form.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION CONCLUSION REFERENCES

Protein kinases and phosphatases play a decisive role in many biological processes by phosphorylating and dephosphorylating target proteins. For humans, it was estimated in the mid-1990s that the genome may contain as many as 2000 protein kinases and 1000 protein phosphatases (1). Initial analysis, after decoding of the DNA that constitutes the human genome, indicates a somewhat lower number of predicted protein kinases and protein phosphatases (2, 3). Since protein kinases are not only involved in normal cell growth but also in malignant transformations, these enzymes have been in focus during recent years as new drug targets (4-6). Protein kinases and phosphatases also play a fundamental role in modulating signals in cellular processes in prokaryotic cells. In the Bacillus subtilis (B. subtilis strain 168) genome sequencing project (7) 4106 protein genes were identified encoding 52 known and 46 putative kinases and 26 known and 10 putative phosphatases (on the World Wide Web, see genolist.pasteur.fr/SubtiList/genome.cgi).

The bifunctional HPr kinase/phosphatase (HPrK/P)¹ in the low guanine and cytosine (low GC) Gram-positive bacteria, B. subtilis, is involved in the main regulatory mechanism for carbon catabolite repression/activation (CCR/CCA) (9-12). The enzyme possesses kinase activity in the presence of ATP and a favorable carbon source, such as glucose, which generates high concentrations of glycolytic intermediates (e.g. fructose 1,6-diphosphate (FBP)), and can phosphorylate two protein substrates (i.e. HPr (histidine-containing protein) and Crh (catabolite repression HPr)) on Ser-46 (10, 11, 13, 14). The presence of two substrates for HPrK/P seems to be unique among low GC Gram-positive bacteria, since Crh has only been detected, thus far, in species of Bacillus (15). High concentrations of inorganic phosphate and low concentrations of ATP, reflecting the intracellular state of cells at starvation (16), have been shown to trigger the phosphatase activity of the enzyme HPrK/P of B. subtilis and, thus, dephosphorylation of HPr(Ser(P)) and Crh(Ser(P)) (9). Furthermore, it has been demonstrated that HPrK/P is an allosteric homo-oligomeric enzyme yielding sigmoidal velocity curves and is strongly regulated by allosteric effector molecules such as FBP (17). In addition to being involved in the CCR/CCA regulatory mechanism, HPr is also a part of the bacterial phosphoenolpyruvate-sugar phosphotransferase system (PTS) and becomes phosphorylated on His-15 by enzyme I (EI) (18). The phosphoryl group can then be transferred to the sugar-specific enzymes II, which phosphorylate incoming PTS sugars (18). The other protein, Crh, which was discovered within the B. subtilis genome sequencing project (7), has a Gln instead of a His at position 15, and in vitro phosphorylation by phosphoenolpyruvate and EI could not be demonstrated (14).

When HPr and Crh are phosphorylated at Ser-46, an interaction has been demonstrated under in vitro conditions with the trans-acting protein CcpA (carbon catabolite control protein A) (13, 19). This complex has been shown to recognize a cis-acting palindromic sequence called cres (catabolite response elements), leading to repression or activation of target gene or operon transcription. There is evidence that CcpA and cre meditate CCR of many catabolic genes or operons such as acsA (acetyl coenzyme A synthetase) (20), xyl (encoding enzymes for xylose utilization) (21, 22), gnt (encoding enzymes for gluconate catabolism) (23, 24), hut (encoding enzymes involved in histidine catabolism) (25, 26), and lev (encoding a fructose-specific PTS and a levanase) (27). CCA involving CcpA as a positively acting regulator has been demonstrated for genes involved in the overflow metabolism for excreting excess glucose, such as ackA encoding acetate kinase (12, 28), pta encoding phosphotransacetylase (29), and alsS encoding acetolactate synthase (30). Both CCR and CCA mediated by CcpA appear to be a general mechanism for Gram-positive bacteria with low GC content (31, 32).

The aim of this study was to further characterize the bifunctional enzyme HPrK/P from B. subtilis and to propose a model for the regulation of the kinase/phosphatase "switch" induced by the metabolic state of the cell.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION CONCLUSION REFERENCES

Purification of HPr, Crh, and HPrK/P from B. subtilis

In order to purify HPr(His)₆, Escherichia coli strain M15 containing plasmid pREP4 (Qiagen, Courtaboeuf, France) was transformed with plasmid pAG2 (14). For Crh(His)₆, E. coli strain BL21(DE3) containing plasmid pLysS (Novagen, Madison, WI) was transformed with plasmid pAG1 (14). For the purification of HPrK/P, E. coli strain BL21(DE3) was transformed with plasmid pET32LIC (Novagen, Madison, WI) encoding HPrK/P with a thioredoxin-, His-, and S-tag (Trx-His₆-S-tag). Transformants were grown overnight at 37 °C in Bouillon Trypto-caseine soja medium (Bio-Rad) with 50 µg/ml ampicillin for HPrK/P, 50 µg/ml ampicillin and 25 µg/ml kanamycin for HPr, and 50 µg/ml ampicillin and 170 µg/ml chloramphenicol for Crh. The following morning, the cultures were transformed to 1 liter of medium containing the desired antibiotic drugs and were grown until A₆₀₀ was 0.6. Expression of the genes was induced by adding isopropyl -D-thiogalactopyranoside (Sigma) to a final concentration of 1 mM. Further incubation was performed at 37 °C for 2 h. Cells were harvested by centrifugation at 5000 rpm for 25 min at 4 °C. The supernatant was discarded, and the cell pellet was stored at 80 °C for at least 1 h before being resuspended in 5-10 volumes of 50 mM Tris buffer, pH 8, containing 0.3 M NaCl. A small amount of deoxyribonuclease I type I (Sigma) was added, and the cell mixture was homogenized using Ultra-TURRAX for 2-3 min. The cells were then ruptured by two passages through the French press (American Instrument Co.) at 500 p.s.i. Whole cells and cell debris were removed by centrifugation at 30,000 rpm for 30 min at 4 °C, and the resultant supernatant was then mixed during 30 min with Ni²⁺-NTA resin (Qiagen, Courtaboeuf, France) (1 ml of resin/g of cells) pre-equilibrated with 10 column volumes of 100 mM Tris, pH 7.4. For HPrK/P, this was performed at 4 °C as well as all subsequent purification steps, whereas for HPr and Crh purification was performed at room temperature. The Ni²⁺-NTA resin was then transferred to a column, and the resin was washed with 50 mM Tris, pH 7.4, containing 50 mM Na₂SO₄ and 15% glycerol. The proteins were eluted with imidazole in 50 mM Tris buffer, pH 7.4, containing 50 mM Na₂SO₄ and 15% glycerol. HPr was collected from the elution fractions with 30 and 300 mM imidazole. Crh and HPrK/P were eluted with 300 mM imidazole. The proteins were verified with SDS-PAGE separations using PhastGel (Amersham Biosciences) and Coomassie staining. After concentrating the fractions by an Ultrafree centrifugal filter unit with a molecular mass cut-off of 5000 Da, 4 ml (Millipore Corp., Bedford, MA) at 5000 rpm to about 2.5 ml, the protein solutions were desalted using a PD-10 column (Amersham Biosciences) pre-equilibrated with 25 ml of 10 mM Tris buffer, pH 8. The protein solutions were run into the column and eluted with 3.5 ml of the buffer solution.

The sequences of the proteins are as follows: HPr(His)₆; MRGSHHHHHH GSMAQKTFKV TADSGIHARP ATVLVQTASK YDADVNLEYN GKTVNLKSIM GVMSLGIAKG AEITISASGA DENDALNALE ETMKSERLGE; Crh(His)₆, MVQQKVEVRL KTGLQARPAA LFVQEANRFT SDVFLEKDGK KVNAKSIMGL MSLAVSTGTE VTLIAQGEDE QEALEKLAAY VQEEVLQHHHHHH; and HPrK/P(Trx-His₆-S-tag), MSDKIIHLTD DSFDTDVLKA DGAILVDFWA EWCGPCKMIA PILDEIADEY QGKLTVAKLN IDQNPGTAPK YGIRGIPTLL LFKNGEVAAT KVGALSKGQL KEFLDANLAG SGSGHMHHHH HHSSGLVPRG SGMKETAAAK FERQHMDSPD LGTDDDDKMG NVRTKDVMEQ FNLELISGEE GINRPITMSD LSRPGIEIAG YFTYYPRERV QLLGKTELSF FEQLPEEEKK QRMDSLCTDV TPAIILSRDM PIPQELIDAS EKNGVPVLRS PLKTTRLSSR LTNFLESRLA PTTAIHGVLV DIYGVGVLIT GKSGVGKSET ALELVKRGHRL VADDCVEIRQ EDQDTLVGNA PELIEHLLEI RGLGIINVMT LFGAGAVRSN KRITIVMNLE LWEQGKQYDR LGLEEETMKI IDTEITKLTI PVRPGRNLAV IIEVAAMNFR LKRMGLNAAE QFTNKLADVI EDREQEE. The concentration of HPrK/P(Trx-His₆-S-tag) was determined spectrophotometrically using the Bio-Rad protein assay (Bio-Rad) with Bio-Rad protein assay standard I lyophilized bovine plasma -globulin (Bio-Rad) as a standard, and the concentrations of Crh and HPr were determined by UV spectrophotometry using the extinction coefficient for one and two tyrosine residues, respectively (1500 and 2900 M¹ cm¹). Protein solutions were stored at 20 °C.

Electrospray Ionization Mass Spectrometry (ESI-MS) Measurements

Further Treatment of the Proteins-- For ESI-MS analysis of HPrK/P(Trx-His₆-S-tag), further purification was performed using a 1-ml HiTrap Q Sepharose high performance column (Amersham Biosciences). The HiTrap Q column was equilibrated with 5 column volumes of 25 mM Tris buffer (pH 8) followed by 5 column volumes of 25 mM Tris containing 1 M KCl (pH 8) and finally 10 column volumes of 25 mM Tris buffer (pH 8). The protein solution was applied to the column, and fractions were eluted with 25 mM Tris buffer (pH 8), containing increasing concentrations of KCl (100, 300, 600, and 1000 mM). The purity of the fractions, after concentration using an Ultrafree centrifugal filter unit with a molecular mass cut-off of 10,000 Da (Millipore), was confirmed with SDS-PAGE separation using PhastGel (Amersham Biosciences) and Coomassie staining. The pure fractions were then desalted through a PD-10 column (Amersham Biosciences) as described above.

Prior to mass spectrometry analysis, an additional desalting procedure was performed with Centricon (Millipore Corp.) using 10 mM ammonium acetate (pH 6.8) as reconstitution solution. Ammonium acetate was used, since this buffer preserves the native structure of proteins and is compatible with the ESI-MS analysis. Concentrators with a molecular mass cut-off of 10,000 Da were used for HPrK/P(Trx-His₆-S-tag), and devices with a molecular mass cut-off of 3000 Da were used for HPr(His)₆ and Crh(His)₆. The procedure for desalting was conducted following the protocol provided by the company and contained six dilution/concentration steps performed at 4 °C for 60 min. Finally, the concentration of the proteins was determined spectrophotometrically using the Bio-Rad protein assay. The samples were stored at 20 °C until ESI-MS measurements were performed.

Instrumentation-- Mass spectrometry experiments were conducted using a hybrid quadrupole time-of-flight mass spectrometer (Q-TOF II; Micromass, Altrincham, UK) equipped with a Z-Spray ESI source. All spectra were recorded in the positive ion mode.

ESI-MS Analysis under Denaturing Conditions-- Purity and homogeneity of HPr(His)₆, Crh(His)₆, and HPrK/P(Trx-His₆-S-tag) were verified by mass analysis under denaturing conditions with the proteins diluted to 10 µM in a 1:1 water/acetonitrile (v/v) mixture acidified with 1% formic acid. Under these conditions, noncovalent interactions between subunits are disrupted, thus allowing for the accurate measurement of the molecular weight of the constitutive enzyme subunits with a precision of better than 0.01%. Mass spectra were recorded in the positive ion mode in the mass range of 500-4000 m/z (mass to charge) after calibration of the instrument with horse heart myoglobin diluted to 2 µM in a 1:1 water/acetonitrile (v/v) mixture acidified with 1% formic acid.

ESI-MS Analysis under Native Conditions-- Under native conditions, the mass measurements of HPr(His)₆, Crh(His)₆, and HPrK/P(Trx-His₆-S-tag) were performed using 10 mM ammonium acetate buffer (pH 6.8). Samples were diluted to ~20 µM and continuously infused into the ESI ion source at a flow rate of 5 µl/min. To study the influence of pH variations on the stability of the oligomeric state of the enzyme, increasing amounts of NH₃ were added, up to pH 9.5, to the 10 mM ammonium acetate buffer. HPrK/P(Trx-His₆-S-tag) was then diluted to 20 µM with the pH-adjusted buffer. All of the measurements were performed at an accelerating voltage equal to 200 V and with the pressure in the interface region of the mass spectrometer equal to 6.5 millibars. Mass data were acquired in the positive ion mode in the 2500-12,000 m/z mass range. Clusters of CsI (separate injections of 1 mg/ml CsI in 50% aqueous isopropyl alcohol) were used for the calibration of an extended mass range in the high m/z region. The molecular weight of the noncovalent complex of the enzyme was calculated as the mean of five peak maximum values ± S.D.

Dynamic Light Scattering Measurements

Dynamic light scattering data were obtained with the DynaPro-801 instrument (Protein Solutions Inc.) using a 30-milliwatt, 833-nm wavelength argon laser at 20 °C and equipped with a solid-state avalanche photodiode. During the illumination, the photons scattered by proteins were collected at 90 °C on a 10-s acquisition time and were fit with the analysis software, Dynamics. Intensity fluctuations of the scattered light resulting from Brownian motion of particles were analyzed with an autocorrelator to fit an exponential decay function and then measuring a translational diffusion coefficient D. For polydisperse particles, the autocorrelation function was fit as the sum of contributions from the various size particles using the regularization analysis algorithm. D is converted to a hydrodynamic radius R_h through the Stokes-Einstein equation (R_h = k_bT/6D, where  represents the solvent viscosity, k_b is the Boltzmann's constant, and T is the temperature), and then to a molecular weight for a spherical particle. Apparent molecular weights were deduced from histograms of distribution of percentage mass versus R_h. HPrK/P was diluted in a 10 mM ammonium acetate buffer, similar to that used in mass spectrometry analysis under native conditions. All solutions were filtered with 0.22-µm Millex filters prior to dilution of the protein. Sample preparations were achieved either by diluting the protein in the pre-equilibrated buffer or by adding increasing amounts of NH₃ to a 10 µM HPrK/P(Trx-His₆-S-tag) solution in ammonium acetate until pH ranged from 6.8 up to 9.5. An aliquot was removed at the desired pH and kept at room temperature for at least 2 h before dynamic light scattering measurement.

BIAcore Surface Plasmon Resonance Analysis

HPrK/P(Trx-His₆-S-tag) oligomerization was examined on a BIAcore instrument (BIAcore J; Amersham Biosciences). The principle of this technology relies on a surface plasmon resonance phenomenon that transforms the specific incident angle of the light reflected from a metallic surface in response to the solute bound to the surface. The Biacore measurements are arbitrary expressed in resonance units that are proportional to the concentration of solute bound to the surface. The monitoring of the surface plasmon resonance response allows analysis such as kinetics in real time and stochiometries for the formation of multimolecular complexes. To determine the effect of pH on the quaternary structure of HPrK/P(Trx-His₆-S-tag), the protein was diluted in a solution of 40 mM Tris-HCl buffered either at pH 6.8 or at pH 9.5 and loaded at a flow rate of 35 µl/min on an NTA sensor chip in conditions of low density of nickel (50 resonance units) and saturation of protein over the nickel (200 nM). Typical experiments were carried out in the same conditions, at 25 °C, with running buffer 10 mM HEPES, 0.15 M NaCl, 50 µM EDTA, 0.005% surfactant P20, pH 7.4, and with a constant flow rate of 5 µl/min. Alternatively, to evaluate the dissociation constants, a running buffer of 10 mM Tris-HCl, 50 µM EDTA, 0.005% surfactant P20, pH 6.8 or 9.5, was used. To correct for nonspecific binding to dextran matrix, experiments were performed with two channels of the sensor chip simultaneously, one not being regenerated with nickel.

Chemicals

Pyruvic acid, D-fructose 2,6-diphosphate (F2,6BP), D-fructose 1,6-diphosphate (FBP), acetyl coenzyme A, -nicotinamide adenine dinucleotide (-NAD⁺), -nicotinamide adenine dinucleotide reduced form (-NADH), D-glucose 6-phosphate, DL-glyceraldehyde 3-phosphate, CoCl₂·6H₂O, CuCl₂·2H₂O, MnCl₂·4H₂O, and acetyl phosphate were purchased from Sigma. MgCl₂·6H₂O and CaCl₂·2H₂O were obtained from Merck.

Radioactive Kinase and Phosphatase Assay

The assay mixture for in vitro phosphorylation of HPr(His)₆/Crh(His)₆ contained 50 mM Tris buffer (pH 8), 5 mM MgCl₂, 0.5 mM ATP, 3.3-5 Bq (leading to 200-300 cpm of [-³²P]ATP/pmol of ATP) (Amersham Biosciences), 2 mM FBP, 0.1% bovine serum albumin, 10 µM HPr/Crh, and, to initiate the reaction, 100 nM HPrK/P(Trx-His₆-S-tag), in a final volume of 20 µl. The mixture was incubated at 37 °C for 10 min, and the phosphorylation reaction was then terminated by spotting samples onto 1 × 1-cm P81 phosphocellulose paper (Whatman, Maidstone, UK) and dropped immediately into a beaker containing 75 mM H₃PO₄. The total volume of phosphoric acid solution used was ~10 ml for each paper. Unreacted ATP was removed by washing three times with 75 mM H₃PO₄, for 15 min each, and once with ethanol, just covering the papers, for 5 min. The papers were dried and transferred to scintillation vials containing 6 ml of scintillation solution for water samples, Rotiszint ecoplus (Carl Roth, Karslruhe, Germany), and the radioactivity was determined in a scintillation counter, LKB 1211 Rackbeta (PerkinElmer Life Sciences). Typically, each condition was tested in triplicate.

For the phosphatase assay, HPr(His)₆/Crh(His)₆ was first phosphorylated in a kinase assay. The final volume was 20 µl and contained 50 mM Tris buffer (pH 8), 7 mM MgCl₂, 2 mM ATP, 3.3-5 Bq (leading to 200-300 cpm of [-³²P]ATP/pmol of ATP) (Amersham Biosciences), 5 mM FBP, 0.1% bovine serum albumin, 200 µM HPr/Crh, and, to initiate the reaction, 800 nM HPrK/P(Trx-His₆-S-tag). After incubation at 37 °C for 2 h, HPr(Ser(P))/Crh(Ser(P)) was purified on Ni²⁺-NTA resin using a column procedure. The resin was pre-equilibrated with 100 mM Tris buffer (pH 7.4) and then combined with the assay mixture for at least 30 min at room temperature. To remove ATP and FBP, the resin was washed extensively with 50 mM Tris buffer (pH 7.4) containing 50 mM Na₂SO₄ and 15% glycerol until no radioactivity was detected in the eluent. The enzyme and the phosphorylated protein were eluted with 50 mM Tris buffer (pH 7.4) containing 50 mM Na₂SO₄, 15% glycerol, and 300 mM imidazole. Desalting was performed using a PD-10 column (Amersham Biosciences). The column was pre-equilibrated with 25 ml of 10 mM Tris buffer (pH 8). After adding the sample (not more than 2.5 ml) to the column, the proteins were eluted with 3.5 ml of 10 mM Tris buffer (pH 8). Fractions containing radioactivity were collected and pooled. In most of the cases, remaining enzyme from the kinase reaction was removed by S-protein-agarose (Novagen, Madison, WI). After equilibration with 20 mM Tris buffer (pH 7.5), 150 mM NaCl and 0.1% Triton X-100 (Bio-Rad) the S-protein-agarose was gently shaken with the protein solution for 30 min at room temperature. The phosphorylated protein was eluted with 20 mM Tris buffer (pH 7.5), 150 mM NaCl, and 0.1% Triton X-100 as confirmed by radioactivity in the eluent. Desalting was then performed as described above. An Ultrafree centrifugal filter unit (molecular mass cut-off of 5000 Da) was used to concentrate the protein solution (4 °C) according to the operating procedure provided by the company. The amount of phosphorylated protein, pmol/µl, was determined on the basis of cpm/pmol ATP. Amounts of HPr(Ser(P))/Crh(Ser(P)) up to 10 µM were then included in a phosphatase assay containing, in general, 50 mM Tris buffer (pH 8), 5 mM MgCl₂, 0.1% bovine serum albumin, 1 mM inorganic phosphate (K₂HPO₄ and KH₂PO₄, pH 8.0), and, to initiate the reaction, 100 nM HPrK/P(Trx-His₆-S-tag). The reaction mixture was incubated at 37 °C for 2 h. Termination of the reaction and the following steps were the same as for the phosphorylation assay. Typically, each condition was tested in triplicate.

Divalent Cations for HPrK/P Activity-- Different divalent cations, Mg²⁺, Mn²⁺, Co²⁺, Ca²⁺, and Cu²⁺ as chloride salts, were tested at a final concentration of 5 mM. To exclude possible influence from the buffer, 50 mM HEPES (pH 8) was used, since this buffer is known to form only weak complexes with Mg²⁺, Mn²⁺, Ca²⁺, and Cu²⁺ (33, 34).

Effects of pH-- The activity of the enzyme was determined at 11 different pH values, ranging from 5.0 to 8.7, for the different substrates, HPr/HPr(Ser(P)) and Crh/Crh(Ser(P)). The buffer used, to cover the pH range, was a mixture of 0.2 M citric acid and 0.2 M Tris (final concentration of 50 mM). The pH was adjusted at 37 °C. For the kinase stability study, the enzyme was incubated for 10 min at 37 °C at the different pH values indicated above in 50 mM citric acid/Tris. After incubation, the residual activity was measured at pH 8.0 for HPr and pH 9.0 for Crh (final concentration of 250 mM Tris) at 37 °C for 10 min. In the phosphatase stability study, the procedure was the same except that the activity of the enzyme was measured at pH 6.2 for HPr(Ser(P)) and 7.0 for Crh(Ser(P)). In the former case, MES was used as a buffer.

Effects of Temperature-- The temperature effects on enzyme stability were monitored in an assay with 50 mM MOPS (pH 8.0) adjusted at room temperature over a temperature range of 4-70 °C, and the assay mixture was allowed to incubate for 10 min. MOPS was used as a buffer, since pK_a/°C is low, 0.006, compared with other buffers (33). For the phosphatase assay, 10 mM inorganic phosphate was used, and the reaction was allowed to proceed for 10 min before termination.

Effects of Different Glycolytic Intermediates-- Glycolytic intermediates were tested at the following concentrations: 1 mM acetyl phosphate, 2 mM FBP, 1 mM F2,6BP, 0.5 mM D-glucose 6-phosphate, 1 mM DL-glyceraldehyde 3-phosphate, 1 mM NAD⁺, 0.1 mM NADH, 0.5 mM acetyl coenzyme A, and 1 mM pyruvate. Up to 40 mM FBP and F2,6BP were tested in a phosphatase assay. Inorganic phosphate was omitted when evaluating the potential of acetyl phosphate and DL-glyceraldehyde 3-phosphate as possible candidates to induce phosphatase activity.

Kinetic Parameters for HPr/Crh and ATP-- To determine the kinetic parameters for HPr/Crh, the concentration of the protein was varied from 0.1 to 200 µM while keeping the concentration of the second substrate, ATP, at saturation (2 mM) with 7 mM MgCl₂. When the kinetic parameters for ATP were determined, the concentration of HPr/Crh was 200 and 169 µM, respectively, and the concentration of ATP was varied between 1 µM and 1 mM. The concentration of MgCl₂ was kept with an excess of 5 mM over the total ATP concentration to have a constant proportion of ATP existing as MgATP² and a constant concentration of free Mg²⁺ (35). The initial velocity was determined after 10 min of incubation.

Estimation of Kinetic Parameters-- The program GraphPad Prism was used to determine K_0.5 (the half-saturation constant in the Hill equation), the Hill coefficient (h), and V_max using the following equation: Y = V_maxS^h/(K_0.5 + S^h), where S is the substrate concentration.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION CONCLUSION REFERENCES

Characterization of the Oligomeric State of the Purified Recombinant Proteins

ESI mass spectra were first recorded under denaturing conditions of the different proteins, HPr(His)₆, Crh(His)₆, and HPrK/P(Trx-His₆-S tag), to verify purity and homogeneity, and to determine the molecular weight of constitutive subunits. The molecular mass determined for HPr(His)₆ was 10,688.5 ± 0.2 Da, which is in agreement with the theoretical value of 10,687 Da calculated from the amino acid sequence (data not shown). A minor compound (~10%) with a mass of 10,768.6 ± 0.2 Da was also detected, corresponding most probably to the phosphorylated form of the protein (M = 80 Da). For Crh(His)₆, the main species detected (~85%) had a mass of 10,392.5 ± 0.2 Da, which is in agreement with the theoretical value of 10,390 Da (data not shown). About 15% of the signals were converted to a molecular mass of 10,261.6 ± 0.1 Da, which is probably due to the loss of the N-terminal Met (M = 131 Da) from Crh. The different constructions of HPr and Crh with a His tag at the N terminus preceded by a Met, Arg, Gly, and Ser for HPr and a His tag on the C terminus and a Met followed by Val on the N terminus for Crh could be a reason for the differences observed in the mass spectra. A possible explanation for the findings that a part of HPr was phosphorylated, but not Crh, may be due to the fact that HPr was phosphorylated at His-15 by the EI during the purification procedure when the plasmid was expressed in Escherichia coli. In Crh the His in position 15 is exchanged by a Gln, and no phosphorylation has been demonstrated in vitro by phosphoenolpyruvate and EI (14). ESI-MS analysis of HPrK/P(Trx-His₆-S-tag) under denaturing conditions revealed a highly homogeneous sample with a single species of 51,700.7 ± 1.0 Da, which is in agreement with the theoretical value, 51,699 Da, taking into account the loss of the N-terminal Met (data not shown).

In addition to the use of ESI-MS for the evaluation of the purity and homogeneity of proteins, the technique has also become a useful method to study noncovalent complexes (for recent reviews, see Refs. 36-38). ESI-MS analysis under native conditions of the proteins revealed only monomers of HPr(His)₆ and Crh(His)₆ (10 mM ammonium acetate buffer (pH 6.8), data not shown). Using other biophysical techniques, circular dichroism and NMR, HPr was reported to be monomeric, whereas Crh was detected both as monomer and dimer (39). Substantially different experimental conditions presumably explain the different results obtained by ESI-MS and NMR.

ESI-MS analysis unambiguously revealed that HPrK/P(Trx-His₆-S-tag) from B. subtilis is a specific noncovalent hexamer at pH 6.8 with a measured molecular mass of 310,337 ± 22 Da (Fig. 1A). The molecular mass obtained under native conditions is 0.05% higher than the mass predicted from the denatured analysis of HPrK/P(Trx-His₆-S-tag). This mass difference is very low compared with other mass measurements analyzing noncovalent subassemblies (40-42) and is not significant given the uncertainty of the measurements in native conditions. However, this difference is presumably due to the inclusion of water molecules or small cations that are not present in the denatured monomer. Despite the discrepancy, there is no doubt that HPrK/P(Trx-His₆-S-tag) exists as a hexamer at pH 6.8. The ESI-MS analysis is not in agreement with earlier reported results performed by size exclusion chromatography and analytical ultracentrifugation, which suggest that the enzyme is an octamer or possibly a heptamer (17). However, a precise quantification and an accurate determination of the molecular weight of oligomeric proteins are not directly and precisely determined with these latter techniques. The results from ESI-MS measurements are more precise and consistent with the x-ray crystallography experiments of HPrK/P from Lactobacillus casei showing that the catalytic domain of the enzyme is a hexamer in the crystal, suggesting the same state in solution (43). The x-ray structure was solved at pH 5.2 with the truncated HPrK/P with an N-terminal fragment missing. Whereas the N-terminal fragment is poorly conserved in the enzyme from different bacteria, the remaining residues, including the putative ATP- and HPr-binding domains, are highly conserved (43). Recently, full-length HPrK/P from Staphylococcus xylosus was crystallized (pH 7.6), also supporting a hexameric structure of the enzyme (44). Furthermore, full-length Mycoplasma pneumoniae HPrK/P(His)₆ was also reported to be a hexamer as revealed by biophysical and crystallographic data (pH 7.5) (45).

View larger version (21K): [in this window] [in a new window]   Fig. 1.   ESI mass spectra of HPrK/P (20 µM) under native conditions at pH 6.8 (A) and pH 9.5 (B). At pH 6.8, the major signals observed in the mass spectrum are due to multiply charged ions of the hexamer. Minor signals reveal the coexistence of small amounts of monomer and dimer. Increasing the pH to 9.5 results in destabilization of the hexamer, and most of the enzyme was detected as a monomer. Traces of dimer and hexamer are also observed. The accelerating voltage was equal to 200 V, and the pressure in the interface region was 6.5 millibars.

Since the conditions for ESI mass measurements differed from the radioactive kinase assay (see below), due to different buffer solutions and an additional desalting procedure, the kinase activity of HPrK/P was also tested under the same conditions as those used for mass spectrometry. Using 10 mM ammonium acetate (pH 6.8) as buffer, the native enzyme was found to be fully active.

Divalent Cations Are Necessary for Both the Kinase and Phosphatase Activity

In the absence of divalent cations, neither kinase nor phosphatase activity was detected. The ion required to produce maximal kinase activity, when HPr was included in the assay, was Mg²⁺ (Fig. 2A), which is in agreement with previously published results for HPrK/P and HPr from B. subtilis (11). When Crh was used as a substrate, Mg²⁺ gave the same degree of phosphorylation as the divalent cation Mn²⁺. However, for HPr, a somewhat lower activity was observed with Mn²⁺ compared with Mg²⁺. Some activity was also observed with Co²⁺, more for HPr than when Crh was used as a substrate. For Ca²⁺ and Cu²⁺, no activity was observed for either HPr or Crh under the conditions used. The effect of divalent cations on HPr phosphorylation by HPr kinase from other Gram-positive bacteria with low GC content has also been investigated. For HPr kinase from Streptococcus salivarius, Mg²⁺ was the preferred cation, but activity was also observed in the presence of Mn²⁺ and Co²⁺ (46). No effect was seen for Ca²⁺ and Cu²⁺ (46). HPr kinase from Streptococcus pyogenes was maximally activated by Mg²⁺ and Mn²⁺, and no activity was observed for the cations Sn²⁺, Ni²⁺, and Cu²⁺ (47). For the different HPr kinases, Mg²⁺ seems to be the preferred divalent cation.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   The effect of different divalent cations on kinase activity (Graph A) and phosphatase activity (Graph B) of HPrK/P for HPr (black bars) and Crh (white bars). The relative activity of the phosphorylated proteins from the kinase assay (100%), before adding the phosphate, is illustrated by the gray bar in graph B. The concentration of the divalent cations was 5 mM. The data represent the mean ± S.E. (n = 3).

For the phosphatase assay, activity was observed in the presence of Mg²⁺ and Mn²⁺ (Fig. 2B). However, in this case, the same activity was obtained with either Mg²⁺ or Mn²⁺ when including HPr(Ser(P)) as the substrate, but in the case of Crh(Ser(P)) a more pronounced effect was observed with Mg²⁺ compared with Mn²⁺ on dephosphorylation of the proteins. For Co²⁺, Ca²⁺, and Cu²⁺, no major effect on the phosphatase activity was detected.

The HPrK/P was completely dependent on divalent cations for kinase activity. In general, protein kinases require metal ions, preferentially Mg²⁺, for catalysis (48). The metal ion neutralizes charge and orients and polarizes the -phosphoryl group, which facilitates the phosphorylation reaction (48, 49). Most protein kinases seem to bind two metal ions that surround the triphosphate of ATP (48). Although divalent metal ions other than Mg²⁺ can be involved, Mg²⁺ is considered the physiological activator in, at least, eukaryotes, due to its high concentration in the cell (48). For protein kinases in general, maximal activity is usually also observed with Mg²⁺ except for most tyrosine kinases, which are maximally activated by Mn²⁺ (50). The phosphatase activity of HPrK/P also required the addition of metal cations for catalysis, which is a common contributing factor for catalysis among Ser(P)/Thr(P) protein phosphatases (49).

Is pH Involved in the Switch between the Kinase and Phosphatase Activity?

Effect of pH on the Oligomeric State of the Enzyme Probed by Mass Spectrometry-- Under carefully controlled operating conditions (accelerating voltage set to 200 V and the pressure in the interface region of the mass spectrometer set to 6.5 millibars), HPrK/P was investigated under native conditions at different pH values. The major ion series observed in the mass spectrum at pH 6.8 was related to multiply charged ions of a hexamer (310,337 ± 22 Da) (Fig. 1a). Minor peaks were attributed to the monomer (51,699 Da) and dimer (103,404 Da) species of the enzyme. An estimation of the different oligomeric forms of the enzyme at pH 6.8 revealed that ~80% of the detected ions correspond to the hexamer, whereas ~10% correspond each to the monomer and dimer. Increasing the pH to 9.5, by the addition of ammonium hydroxide but in otherwise strictly identical conditions as at pH 6.8, resulted in a complete dissociation of the hexamer, and the most intense signals in the spectrum were those of the multiply charged monomer and dimer (Fig. 1b). Approximately 70% of the detected ions corresponded to the monomer, and ~20% of the signals were those of the multiply charged dimer. Only weak signals representative of the hexamer (~10%) were observed.

Effect of pH on Kinase and Phosphatase Activity by HPrK/P of HPr and Crh-- The effect of pH on substrate binding and catalysis was studied using a mixed citric acid/Tris buffer covering the pH range between 5 and 8.7 (Fig. 3). The kinase activity versus pH shows an optimal activity between pH 6.6 and 8.7 for the enzyme and HPr and a maximal activity at pH 8.7 for the enzyme and Crh. To obtain further information regarding enzyme characteristics and to assess the pH stability of HPrK/P, the enzyme was incubated for 10 min at 37 °C at the indicated pH values in 50 mM citric acid/Tris. After incubation, the residual activity was measured at pH 8.0 for HPr and pH 9.0 for Crh (final concentration of 250 mM Tris) at 37 °C and after incubation for 10 min. A decline in activity was observed at pH below 5.8, but above this value the enzyme retained its activity (data not shown). The decrease in activity between 5.8 and 6.6 when using HPr as substrate and between 5.8 and 8.7 when using Crh as substrate is probably due to improper ionic forms of the active site and/or substrate (51). The decline in activity below pH 5.8 can in part be related to irreversible denaturation of the enzyme (51). Thus, the activity loss between 6.6 and 8.7 for the enzyme and Crh seems to be related more to an improper ionic form of the substrate than to the enzyme, since the enzyme plus HPr does not show this reduction in activity. Regarding the influence of pH on HPr kinase activity from other species, the pH optima for HPr kinase from S. salivarius and Streptococcus mutans Ingbritt were reported to be 7.5 and 7.0, respectively (46, 52). The pH optimum for the stability of the enzyme from S. salivarius after preincubation for 30, 60, and 90 min prior to measuring its activity was 8.0 (46).

View larger version (13K): [in this window] [in a new window]   Fig. 3.   The effects of pH on enzyme activity using the substrates HPr/HPr(Ser(P)) (Graph A) and Crh/Crh(Ser(P)) (Graph B). The kinase activity () correlates to the left axis, whereas the phosphatase activity () correlates to the right axis. The activities were recorded using mixed citric acid/Tris buffer at a final concentration of 50 mM. The data represent the mean ± S.E. (n = 3). Graph C represents the oligomeric state of the enzyme (apparent molecular weigh) as a function of pH using dynamic light scattering () or mass spectrometry ().

The activity and stability of HPrK/P when acting as a phosphatase were also investigated. Whereas the activity, when investigated with HPr(Ser(P)) as substrate, was optimal at pH 6.2 and then decreased successively above and below this value, a different activity was observed when using Crh(Ser(P)) as substrate (Fig. 3). In the latter case, optimal activity was observed between pH 6.2 and 7.5. Stability experiments measuring the activity at pH 6.2 for HPr(Ser(P)) and pH 7.0 for Crh(Ser(P)) after preincubating the enzyme did not influence the activity (data not shown).

A pH-dependent switch between the kinase and the phosphatase activities of the bivalent enzyme was observed. The kinase activity was predominant at higher pH, and increasing phosphatase activity was recorded at lower pH. The reason for this switch may be connected to the starvation metabolism of the bacterium. In general, the internal pH of neutralophilic bacteria, including B. subtilis, is between 7.5 and 8.0 (53). It has also been demonstrated that B. subtilis possesses a rather high cytoplasmic buffering capacity (54). However, entering the sporulation phase in response to starvation for a variety of nutrients, the internal pH of spores from B. subtilis was found to be ~6 (55). Thus, the kinase activity is high at pH around 8 when growing in the presence of certain nutrients, whereas decreasing pH favors phosphatase activity when the bacteria is about to enter sporulation due to starvation before metabolic dormancy. Favoring the phosphatase activity leads to relief of CCR/CCA. For another bifunctional enzyme, the 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase from bovine liver, the pH optimum for the kinase activity was 8.5, whereas the phosphatase reaction was maximal at pH 6.5 (56). Thus, a similar pH dependence for the two divergent opposing activities was observed, as for HPrK/P.

The ESI-MS results and the activity data indicate that the state of oligomerization may be an important factor in the switch between the kinase and phosphatase activity. Thus, when the enzyme exists as a hexamer, the phosphatase activity is favored.

In order to analyze this pH switch, we followed the oligomerization state of the enzyme as a function of pH using mass spectrometry, dynamic light scattering, and BIAcore techniques. The pH titration by mass spectrometry was done using the previously described conditions.

We used dynamic light scattering to evaluate the changes in the quaternary structure of the HPrK/P as it has been observed from mass spectrometry analysis upon changes of pH. Interestingly, the analysis of histograms of distribution in percentage of mass shows that the apparent molecular masses were about 2-fold higher at pH 6.8 than at pH 9.5, respectively (343 and 143 kDa (data not shown). These average molecular weights calculated for a spherical particle and obtained from eight independent measurements are fully consistent with the existence of the HPrK/P as a hexamer at pH 6.8, whereas at pH 9.5 it more likely forms a trimer, although we cannot exclude the possibility that a dimer would be also compatible with the molecular weight of the particles at pH 9.5 with regard to the shape of the oligomers. We noted the existence of a mixture of particles in size as suggested by the high polydispersity value (around 40%) and the poor fit to a monomodal distribution. We assume that this polydispersity reflected a dynamic equilibrium of the HprK/P between different oligomeric states and a trend to form aggregates as well, since it persisted despite attempts to eliminate high size particles or possible dust particles by centrifugation, filtrating, and increasing the ionic strength of the protein samples. Therefore, the regularization in bimodal distribution of radii by percentage of mass as a measurement of lower bound particle size appears to be a suitable approach for surveying the pH effect on the structure of the multimeric HprK/P. In summary, our results suggest that HprK/P undergoes a transition in its oligomeric state upon the increase of basicity from pH 6.8 to pH 9.5. The trimer (and/or dimer) appears to be a predominant form of the HprK/P at pH 9.5, and the hexamer appears to be the main form at pH 6.8. Fig. 4C shows the pH titration experiments performed by mass spectrometry or by dynamic light scattering. There is a good agreement between the results obtained by those two techniques and a good correlation with the pH titration of the phosphatase activity using HPr as a substrate (Fig. 3A).

View larger version (14K): [in this window] [in a new window]   Fig. 4.   The temperature effect on enzyme activity in a kinase (Graph A) and phosphatase assay (Graph B) with HPr () and Crh (). The experiments were performed in 50 mM MOPS buffer. The data represent the mean ± S.E. (n = 3).

We used the BIAcore technology and NTA sensor chip to capture pH-dependent different quaternary structures of the HPrK/P. The experiments were designed to measure a different amount of protein depending on whether the nickel was bound to the hexameric, trimeric, dimeric, or monomeric form of the protein. Thus, we generated NTA chips with low density of nickel so that each nickel was bound to only one hexamer of HprK/P at saturation. Under these conditions, we observed that for the same amounts of HprK/P injected on an NTA chip (regenerated with the same density of nickel), the amount of protein bound was about 2 times less when it was loaded in Tris buffer, pH 9.5, than when it was at pH 6.8. Accordingly to the previous results from mass spectrometry and dynamic light scattering studies, this difference was consistent with the binding of hexamer to the NTA-Ni²⁺ chip when HprK/P is loaded at pH 6.8 and the binding of trimer when loaded at pH 9.5. Furthermore, the analysis of the sensorgrams obtained from the protein loaded at pH 6.8 showed that the level of resonance units decreased rapidly after it had reached a maximum during the phase of binding. This dissociation phase was initiated at the switch between the loading buffer and the running buffer. We explained this effect as a dissociation of the hexamer at the change of pH from 6.8 to 7.4 between both buffers. In order to investigate more thoroughly the kinetic of dissociation, we performed experiments using running buffers similar to the loading buffers, pH 6.8 and 9.5, following each injection of the protein. The analysis of the sensorgram (data not shown) showed that only the protein loaded at pH 6.8 was subject to a sharp decrease of resonance units following the switch to the running buffer, pH 9.5. The absence of such a fast dissociation with the protein loaded at pH 9.5 indicated that this effect was due to the dissociation of the bound oligomers itself rather than to its dissociation from the Ni²⁺-NTA chip. Both the stochiometry of the fixation of the HprK/P on the NTA-Ni²⁺ chip and the kinetics of its dissociation provided further evidence of a transition in the protein oligomeric state dependent on the pH that is consistent with the formation of a hexamer at pH 6.8 and a trimer at pH 9.5.

Altogether, these results strongly suggest that the kinase-phosphatase switch is linked to the oligomeric state of the enzyme. Moreover, BIACORE technology may be used to follow the oligomeric states of the enzyme under different physiological conditions.

The Enzyme Activities for the Two Substrates, HPr and Crh, Are Differentially Affected by Temperature

The temperature effect on enzyme activity was carried out over a temperature range between 4 and 70 °C using 50 mM MOPS (pH 8.0), since this buffer is more stable with temperature (pK_a/°C = 0.006) (33). Using an incubation time of 10 min, the optimum temperature for the kinase activity was recorded to be between 37 and 45 °C for HPr and 45 °C for Crh (Fig. 4). Furthermore, the enzyme activity with HPr and Crh, respectively, differs with a higher degree of phosphorylation for HPr at temperatures below 45 °C than for Crh. The reverse is observed for temperatures above 45 °C (i.e. a higher degree of phosphorylation for Crh than for HPr). In the phosphatase assay, the optimal temperature appears to be 50 °C. No major differences in the behavior of the activity between the two different substrates were observed in this case.

Species of Bacillus are the only Gram-positive bacteria with low GC content, which in addition to HPr, is provided with a second protein substrate, Crh (15). The kinase activity was more stimulated at temperatures below 45 °C using HPr as a substrate, whereas temperatures above 45 °C favored phosphorylation of Crh. The proposed background for B. subtilis when provided with two different protein substrates for HPrK/P may deal with CCR/CCA in different ecological environments. In this context, an observation worth noticing is that the gene encoding Crh has been localized in a genome area including operons dealing with catabolic degradation of complex substrates found in roots (7).

The Metabolic State of the Cell, through Different Glycolytic Intermediates, Modulates the Kinase/Phosphatase Balance

Several glycolytic intermediates were tested for potential allosteric effects of HPrK/P (Fig. 5). The concentrations were chosen according to representative values recorded for B. subtilis as well as other Gram-positive bacteria. Thus, four intermediates were evaluated at concentrations found in vivo in Streptococcus mutans, another Gram-positive bacterium with low GC content: 0.5 mM D-glucose 6-phosphate, 1 mM NAD⁺, 0.1 mM NADH, and 1 mM pyruvate (57). FBP was initially tested at 2 mM, a concentration found in B. subtilis when grown on D-glucose (58), and F2,6BP was also initially tested at a similar concentration, 1 mM. DL-Glyceraldehyde 3-phosphate was included at a concentration of 1 mM, since intracellular concentrations of this intermediate were found to be as high as 0.6 mM in Streptoccus bovis (59). Acetyl coenzyme A and acetyl phosphate were tested at concentrations of 0.5 and 1 mM, respectively. Intracellular concentrations of these two metabolites have been reported to be ~0.2 mM in the Gram-positive bacteria Corynebacterium glutamicum (60).

View larger version (12K): [in this window] [in a new window]   Fig. 5.   The effect of different glycolytic intermediates on kinase (Graph A) and phosphatase activity (Graphs B and C) for HPr (black bars) and Crh (white bars). The intermediates were tested at the following concentrations: 1 mM acetyl phosphate, 2 mM FBP, 20 and 40 mM FBP in a phosphatase assay, 1 mM F2,6BP, 20 and 40 mM F2,6BP in a phosphatase assay, 0.5 mM D-glucose 6-phosphate, 1 mM DL-glyceraldehyde 3-phosphate, 1 mM NAD+, 0.1 mM NADH, 0.5 mM acetyl coenzyme A, and 1 mM pyruvate. The reference samples in Graph A refer to a kinase assay without any glycolytic intermediate for the two different protein substrates. The reference samples in Graph B refer to the kinase assay (in gray) before proceeding to the phosphatase assay, and the phosphatase assay without any glycolytic intermediate (hatched). The reference sample in Graph C refers to the degree of phosphorylation (in gray) before proceeding to the phosphatase assay. When evaluating the effect of Mg2+, no inorganic phosphate was included. In the cases when testing acetyl phosphate and glyceraldehyde 3-phosphate for triggering the phosphatase activity, inorganic phosphate was also omitted. The values in Graphs B and C were normalized to the assay without any intermediate but with the addition of inorganic phosphate using the program GraphPad Prism. The data represent the mean ± S.E. (n = 3).

Regarding the kinase activity, FBP and F2,6BP were found to stimulate the phosphorylation reaction (Fig. 5A). FBP stimulated HPr phosphorylation more than F2,6BP, whereas in the case of Crh both intermediates caused the same degree of phosphorylation. On the contrary, inorganic phosphate, acetyl phosphate, and DL-glyceraldehyde 3-phosphate inhibited the kinase activity of HPrK/P. The inhibitory effect of inorganic phosphate on kinase activity has previously been reported for HPr kinases (46, 47, 61-63). To further evaluate the effect of inorganic phosphate, 1 mM inorganic phosphate was included in a kinase assay with increasing concentrations of FBP/F2,6BP. Whereas 10 mM FBP partly restored the kinase activity for both proteins, HPr and Crh, 10 mM F2,6BP had only a small effect regarding the phosphorylation of Crh, and no effect of F2,6BP was observed for the phosphorylation of HPr (data not shown). Similar results with FBP and the capacity of restoring the kinase activity in the presence of inorganic phosphate were reported for HPrK/P from Lactobacillus casei (63). A similar experiment was performed to evaluate the possibility to restore the kinase activity in an assay using ATP with 0.1 mM inorganic phosphate in the absence of effector molecules. Increasing concentrations of ATP up to 3 mM only partly restored the kinase activity (data not shown).

For the phosphatase activity, F2,6BP was found to inhibit the dephosphorylation of both HPr(Ser(P)) and Crh(Ser(P)) as well as high concentrations of FBP in the case of Crh(Ser(P)) (Fig. 5B). FBP alone only inhibited dephosphorylation of Crh(Ser(P)), and 2 mM ATP alone did not affect the dephosphorylation activity of either of the proteins, whereas 20 mM FBP with the addition of 2 mM ATP inhibited dephosphorylation of both HPr(Ser(P)) and Crh(Ser(P)) (data not shown). Similar results have been reported previously for HPrK/P from B. subtilis (9). Since acetyl phosphate and glyceraldehyde 3-phosphate inhibited the kinase activity, their role in the phosphatase assay were investigated (Fig. 5C). Both substances seem to have similar potential as a source of inorganic phosphate, thus shifting the enzyme activity to a phosphatase. As in the case with inorganic phosphate, Mg²⁺ is required for activity. However, phosphatase activity was also observed in the absence of inorganic phosphate, acetyl phosphate, or glyceraldehyde 3-phosphate but in the presence of Mg²⁺, although not to the same extent. A similar observation, that the enzyme did not require inorganic phosphate to exhibit phosphatase activity, has also been reported for HPrK/P from S. xylosus (64).

Of the evaluated glycolytic intermediates, FBP as well as inorganic phosphate are known as an allosteric activator and inhibitor, respectively, not only for the kinase activity of HPrK/P from B. subtilis (11) but also for the enzymes from, for example, L. casei (63), Lactobacillus brevis (62), S. mutans (61), and S. pyogenes (47). Furthermore, F2,6BP was also found to be an allosteric activator for the kinase activity. The opposing phosphatase activity was found to be regulated by inorganic phosphate, acetyl phosphate, or glyceraldehyde 3-phosphate. Inorganic phosphate is known as an indicator of the metabolic status of the cell (16). In addition to the role of acetyl phosphate in acetate metabolism and in the overflow mechanism of excess carbohydrates (29), acetyl phosphate has been identified as a regulator of two-component signal transduction systems such as ComA in B. subtilis (65). Acetyl phosphate has been postulated as a good sensor reflecting the intracellular metabolic state of the cell as an indicator of glucose availability, and the synthesis of acetyl phosphate was found to be necessary for glucose-starved cells of E. coli to survive glucose starvation (66). Thus, reflecting the intracellular metabolic status of the cell in response to changing concentrations of PTS sugars, acetyl phosphate could shift the kinase activity of HPrK/P to a phosphatase. Regulatory properties were also found for glyceraldehyde 3-phosphate, which induced the phosphatase activity of the enzyme. Recently presented data suggest multiple levels of control of the gapA operon, including the gene encoding glyceraldehyde 3-phosphate dehydrogenase, gapA, involved in the conversion of glyceraldehyde 3-phosphate to 1,3-bisphosphoglycerate (67). This enzyme catalyzes an irreversible reaction, and enzymes catalyzing irreversible reactions in the glycolysis were found to be inducible by glucose. Furthermore, the gapA operon was found to be regulated by a repressor, CggR, and CcpA (67). In view of the complex regulation of the gapA operon to modulate the needs of glycolysis and the tricarboxylic acid cycle under different conditions, the conversion of glyceraldehyde 3-phosphate to 1,3-bisphosphoglycerate is strongly regulated. However, the exact role of glyceraldehyde 3-phosphate in the CCR/CCA mechanism requires further evaluation.

Taking into account the proposed relationship between the oligomeric state of the enzyme and its activity, FBP and F2,6BP are suggested to promote the dissociation of the hexamer into dimer and monomer. The presence of inorganic phosphate, acetyl phosphate, and/or glyceraldehyde 3-phosphate, on the other hand, are suggested to induce the formation of the hexamer.

The Different Substrate and Effector Sites are Strongly Coupled

The kinetic parameters for the three substrates, HPr/Crh and ATP, were determined by varying one substrate of interest and keeping the second substrate at a saturating level. When estimating the kinetic parameters for HPr and Crh, the concentration of HPr/Crh was varied between 0.1 and 200 µM, whereas the concentration of ATP, in general, was kept at 2 mM. Two graphs (Fig. 6) illustrate some of the results, and all of the kinetic parameters are listed in Table I.

View larger version (11K): [in this window] [in a new window]   Fig. 6.   Velocity curves including HPr in the kinase assay at different concentrations of HPr (Graph A) and ATP (Graph B), without effector molecule (), with the addition of 2 mM FBP (), and with 0.5 µM Pi (). For ATP without an effector molecule, the Hill coefficient was held constant due to the low affinity of the phosphate donor to the enzyme and thus made the determination of kinetic parameters problematic. The data represent the mean ± S.E. (n = 3). For all of the kinetic parameters determined in this study, refer to Table I.

Results presented in Fig. 6 (Graph B) are nearly identical to results presented in Fig. 5 of Jault et al. (17). This comparison validates the radioactive assay used in this work, whereas in most of the previous work, the enzymatic activity was measured by a gel assay (17).

The enzyme exhibited a positive homotropic response with respect to substrate binding. For the two substrates, HPr and Crh, the affinity was about 5 times higher for HPr, indicating the preference of HPr as a substrate. The addition of the effector molecule, FBP, resulted in increased values for V_max, whereas no major effect on K_0.5 was observed. F2,6BP, on the other hand, increased both V_max and K_0.5 values. Both activators increased the turnover numbers as well as k_cat/K_0.5. The degree of cooperativity decreased somewhat or was not affected in the presence of the effectors. The inclusion of the negative effector, inorganic phosphate, lowered all of the kinetic parameters except the Hill coefficient for Crh.

Furthermore, the possibility of an interaction between the HPr/Crh and ATP sites was also investigated. The concentration of ATP was decreased for the examination of whether lowered concentrations of the phosphate donor would alter the observed kinetic behavior of HPr/Crh. When the concentration of ATP was decreased 4-fold, the Hill coefficient decreased in the case with HPr and increased in the case with Crh. Thus, the binding of ATP to the enzyme seems to induce and/or influence the positive cooperative binding of HPr/Crh to HPrK/P.

To estimate the kinetic parameters for ATP, the concentration of ATP was varied between 1 µM and 1 mM, and the concentration of HPr/Crh was kept at 200 and 169 µM, respectively. In the absence of effector molecules, the enzyme exhibited low affinity for ATP. The addition of FBP/F2,6BP lowered both the K_0.5 and V_max when using HPr as the second substrate. However, the opposite effect, with increasing values for V_max, was observed when including Crh in the assay. Whereas the k_cat values decreased in the case with HPr, an increase was observed for Crh. In both cases, however, the quotient of k_cat/K_0.5 increased. For the Hill coefficient, no obvious explanation for either the increase or decrease in cooperativity after the addition of FBP/F2,6BP can be offered. Experiments to determine the kinetic parameters at pH 7.4 for the different substrates were also performed without effector molecules (data not shown). The results indicate that the positive cooperativity was more pronounced at pH 8.0 than at pH 7.4. In addition, the values for the turnover number and k_cat/K_0.5 were markedly lowered, except for the turnover number for HPr, which remained the same.

Attempts to determine kinetic parameters in a phosphatase assay failed. Under the different concentrations of phosphorylated substrates that we were able to use due to our incapacity to purify high concentrations of phosphorylated substrate, we were not able to obtain a simple and complete saturation curve. Therefore, we were not able to obtain kinetic parameters using the Hill equation (see "Experimental Procedures").

	CONCLUSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION CONCLUSION REFERENCES

A model for the proposed regulation of HPrK/P with HPr as substrate is illustrated in Fig. 7. The switch between kinase and phosphatase activity is probably correlated to the oligomeric state of the enzyme, where the hexamer favors the phosphatase activity and the trimeric form of the enzyme favors the kinase activity. Fieulaine et al. (43) proposed a model where the shift from kinase to phosphatase activity may be due to a direct competition between ATP and inorganic phosphate. An observation supporting an additional mechanism is that phosphatase activity was also observed even in the absence of inorganic phosphate but in the presence of Mg²⁺. The same phenomenon, no requirement for inorganic phosphate to exhibit phosphatase activity, was reported for HPrK/P from S. xylosus (64). However, a close relationship seems to exist between the kinase and phosphatase activity supported by point mutations of Asp-176 and Asp-177 replaced by Ala residues in HPrK/P from B. subtilis affecting both the kinase and phosphatase activity (68). In addition, several mutants of HPrK/P from B. subtilis and L. casei affect phosphatase activity, but with almost normal kinase activity, and no mutant specifically affecting kinase activity has been found (8, 69). Some of the mutations, related with reduced phosphatase activity, are located on the surface in a direct contact with another subunit, suggesting that the mutation may influence the oligomeric state of the enzyme (8, 69).

View larger version (40K): [in this window] [in a new window]   Fig. 7.   A proposed model for the regulation of HPrK/P illustrated for the substrate HPr. The switch between kinase and phosphatase activity is probably correlated to the oligomeric state of the enzyme regulated by different glycolytic intermediates (e.g. FBP, acetyl phosphate and inorganic phosphate, as well as the pH reflecting the metabolic state of the cell). F2,6B was furthermore found to inhibit the dephosphorylation of HPr(Ser(P)) as well as Crh(Ser(P)), but regarding FBP, an effect was only observed in an assay with Crh(Ser(P)). The glycolytic intermediates, acetyl phosphate and glyceraldehyde 3-phosphate, exhibit similar properties as inorganic phosphate (i.e. stimulating the phosphatase activity). Whether the two intermediates bind to the same site as inorganic phosphate requires further investigation. Both opposing activities of the enzyme required divalent cations not only to neutralize the phosphate group of ATP or HPr(Ser(P))/Crh(Ser(P)) but possibly also through a specific site by modulating the conformational state of the enzyme. The enzyme, which shows a positive cooperativity for the binding of the two substrates, HPr and ATP, also proved to have a kinetic coupling between the two different binding sites.

The internal pH of vegetative cells, ~8, will dislocate the balance to the kinase activity, whereas starvation conditions, when the internal pH is lowered, will favor the phosphatase activity. FBP and F2,6BP, possibly interacting with the same site due to their structural similarities, are probably regulating the kinase activity through allosteric modification of the interaction between the monomeric subunits of the enzyme. The kinetic parameters indicate a stronger regulation by FBP than F2,6BP. Other metabolites that have been reported to stimulate the kinase activity are 6-phosphogluconate and glycerate-2-phosphate, although to a lesser extent than FBP (11).

In addition to inorganic phosphate, the glycolytic intermediates, acetyl phosphate and glyceraldehyde 3-phosphate, were found to induce the phosphatase activity of the enzyme. Whether the latter two regulators bind to the same site as inorganic phosphate is not clear, but for both, divalent cations were necessary for enzyme activity, as in the case with inorganic phosphate. The dephosphorylation reaction does not seem to be a simple reversal of the phosphorylation reaction, since the addition of ADP did not stimulate dephosphorylation of HPr(Ser(P)) in an assay with HPrK/P from L. casei (43). Inhibition of the phosphatase activity was observed in the presence of F2,6BP for HPr(Ser(P)) and Crh(Ser(P)), and also of FBP but only when using Crh(Ser(P)) as a substrate.

The binding of the two substrates, HPr and ATP, yield sigmoidal velocity with positive cooperativity, thus increasing the affinities of the vacant binding sites. Furthermore, the results indicate a kinetic coupling between the ATP and HPr binding sites, since the Hill coefficient was changed when the concentration of ATP decreased. For both the opposing activities of the enzyme, divalent cations were necessary for the reactions not only to neutralize the phosphate group of ATP or HPr(Ser(P))/Crh(Ser(P)) but perhaps also through a specific site modulating the conformational state of the enzyme.

In this study, proteins were expressed in E. coli and fused with different tags (His₆ and Trx-His₆-S) in order to facilitate their purification and to obtain a high yield of a homogenous population of proteins. The different tags were not removed, assuming that they were not going to interfere with the behavior of the enzyme. Comparison of some specific experiments from our laboratory and other groups using different constructions support this assumption (data not shown) (39). However, it cannot be ruled out that the wild-type enzyme may behave in a somewhat different manner. Nonetheless, the proposed model is assumed to be also applicable for the wild-type enzyme.

In summary, the bifunctional HPrK/P from B. subtilis involved in the main mechanism of CCR/CCA is a strongly regulated allosteric enzyme. The differences in response obtained regarding phosphorylation and dephosphorylation of the proteins under different in vitro conditions (e.g. regulation by effector molecules and in cooperativity) could be a means by which bacteria "fine tune" the regulation of metabolic pathways. Thus, the enzyme HPrK/P is involved in a sophisticated regulation system, sensing environmental changes in the availability of carbon sources to adapt the uptake and use of nutrients in a hierarchical manner in at least two different ecological niches.

	ACKNOWLEDGEMENTS

We thank Louis Cuccia for valuable linguistic advice and Mireille Gaire for helpful scientific advice. We thank Virginie Lafont and Danièle Altschuh for technical support with the BIACORE system as well as Philippe Dumas and Bernard Lorber for support with the dynamic light-scattering apparatus.

	FOOTNOTES

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

^§ Recipient of a grant from the Swedish Academy of Pharmaceutical Sciences.

Supported by CNRS and Lilly.

^** To whom correspondence should be addressed. Tel.: 33-3-90-24-42-70; Fax: 33-3-90-24-43-12; E-mail: haiech@pharma.u-strasbg.fr.

Published, JBC Papers in Press, October 30, 2002, DOI 10.1074/jbc.M209052200

	ABBREVIATIONS

The abbreviations used are: HPrK/P, HPr kinase/phosphatase; CCR/CCA, carbon catabolite repression/activation; FBP, D-fructose 1,6-diphosphate; PTS, phosphotransferase system; ESI-MS, electrospray ionization mass spectrometry; F2, 6BP, D-fructose 2,6-diphosphate; MES, 4-morpholineethanesulfonic acid; MOPS, 4-morpholinepropanesulfonic acid; NTA, nitrilotriacetic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION CONCLUSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				K. D. Singh, M. H. Schmalisch, J. Stulke, and B. Gorke Carbon Catabolite Repression in Bacillus subtilis: Quantitative Analysis of Repression Exerted by Different Carbon Sources J. Bacteriol., November 1, 2008; 190(21): 7275 - 7284. [Abstract] [Full Text] [PDF]

				N. P. Tran, J. Gury, V. Dartois, T. K. C. Nguyen, H. Seraut, L. Barthelmebs, P. Gervais, and J.-F. Cavin Phenolic Acid-Mediated Regulation of the padC Gene, Encoding the Phenolic Acid Decarboxylase of Bacillus subtilis J. Bacteriol., May 1, 2008; 190(9): 3213 - 3224. [Abstract] [Full Text] [PDF]

				A. H. Klein, A. Shulla, S. A. Reimann, D. H. Keating, and A. J. Wolfe The Intracellular Concentration of Acetyl Phosphate in Escherichia coli Is Sufficient for Direct Phosphorylation of Two-Component Response Regulators J. Bacteriol., August 1, 2007; 189(15): 5574 - 5581. [Abstract] [Full Text] [PDF]

				J. Deutscher, C. Francke, and P. W. Postma How Phosphotransferase System-Related Protein Phosphorylation Regulates Carbohydrate Metabolism in Bacteria Microbiol. Mol. Biol. Rev., December 1, 2006; 70(4): 939 - 1031. [Abstract] [Full Text] [PDF]

				Y. Liang, F. Du, S. Sanglier, B.-R. Zhou, Y. Xia, A. Van Dorsselaer, C. Maechling, M.-C. Kilhoffer, and J. Haiech Unfolding of Rabbit Muscle Creatine Kinase Induced by Acid: A STUDY USING ELECTROSPRAY IONIZATION MASS SPECTROMETRY, ISOTHERMAL TITRATION CALORIMETRY, AND FLUORESCENCE SPECTROSCOPY J. Biol. Chem., August 8, 2003; 278(32): 30098 - 30105. [Abstract] [Full Text] [PDF]

				S. Nessler, S. Fieulaine, S. Poncet, A. Galinier, J. Deutscher, and J. Janin HPr Kinase/Phosphorylase, the Sensor Enzyme of Catabolite Repression in Gram-Positive Bacteria: Structural Aspects of the Enzyme and the Complex with Its Protein Substrate J. Bacteriol., July 15, 2003; 185(14): 4003 - 4010. [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 278/2/1174    most recent M209052200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ramström, H. Articles by Haiech, J. Search for Related Content PubMed PubMed Citation Articles by Ramström, H. Articles by Haiech, J. Social Bookmarking                      What's this?