JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M400604200 on May 17, 2004

J. Biol. Chem., Vol. 279, Issue 29, 30210-30218, July 16, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/29/30210    most recent
M400604200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Vougier, S.
Right arrow Articles by Ladant, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vougier, S.
Right arrow Articles by Ladant, D.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Essential Role of Methionine Residues in Calmodulin Binding to Bordetella pertussis Adenylate Cyclase, as Probed by Selective Oxidation and Repair by the Peptide Methionine Sulfoxide Reductases*

Stéphanie Vougier{ddagger}§, Jean Mary{ddagger}, Nathalie Dautin§, Joëlle Vinh||, Bertrand Friguet{ddagger}**, and Daniel Ladant¶{ddagger}{ddagger}

From the {ddagger}Laboratoire de Biologie et Biochimie Cellulaire du Vieillissement, EA 3106, IFR 117, Université Paris 7, Denis Diderot, 2 place Jussieu, Paris 75005, the Unité de Biochimie des Interactions Macromoléculaires, CNRS URA 2185, Département de Biologie Structurale et Chimie, Institut Pasteur, 28 rue du Dr. Roux, Paris 75015, and the ||Laboratoire de Neurobiologie et Diversité Cellulaire, CNRS UMR 7637, Ecole Supérieure de Physique et Chimie Industrielles, 10 rue Vauquelin, Paris 75005, France

Received for publication, January 20, 2004 , and in revised form, May 17, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Bordetella pertussis, the causative agent of whooping cough, secretes among other virulence factors an adenylate cyclase (AC) toxin that is able to enter into eukaryotic cells where it is activated upon binding to endogenous calmodulin (CaM) and synthesizes supraphysiological cAMP levels. In vivo, the AC toxin, through its specific interaction with the CD11b/CD18 integrin, primarily targets phagocytic cells such as neutrophils and macrophages. Because neutrophil priming and activation result in the production of reactive oxygen species that may cause intracellular oxidation, we have examined the biological consequences of the oxidation of CaM methionines upon its interaction with AC. We show here that the interaction of CaM with AC is dependent on the reduced state of methionines, because oxidation of all methionine residues of CaM dramatically decreases its affinity for AC. Peptide methionine sulfoxide reductases A (MsrA) and B (MsrB) were able to partially reduce the oxidized CaM, and these partially "repaired" forms could interact with AC nearly as efficiently as the native protein. We further showed that the CaM·AC complex is resistant to oxidation with tert-butylhydroperoxide, and we identified methionine residues 109, 124, and 145 as critical for binding to AC. The resistance of the AC·CaM complex to oxidation and the ability of AC to be efficiently activated by partially oxidized CaM molecules should allow the toxin to exert its cytotoxic effects on activated neutrophils and contribute to the host colonization.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Proteins are sensitive to reactive oxygen species that lead to the oxidation of certain amino acid residues. Within proteins, the sulfur-containing amino acids cysteine and methionine are the most sensitive to oxidation by reactive oxygen species. Methionine oxidation is generally associated with a loss of protein function as observed for the {alpha}1-antitrypsin inhibitor and calmodulin (CaM)1 (1, 2). CaM is a key calcium sensor protein implicated, in eukaryotic cells, in many regulatory pathways (3). This protein, 148 residues long, is composed of two globular domains linked by a central flexible {alpha}-helix. Each globular domain contains two helix-loop-helix motifs, named EF-hands, that bind calcium ions. Upon calcium binding, CaM undergoes major conformational changes that permit its specific interaction with protein targets such as protein kinases, protein phosphatases, phosphodiesterases, ions channels, among others. Due to their side-chain flexibility and hydrophobicity, methionine residues play a crucial role in the CaM/protein effector recognition, and these interactions can be sensitive to methionine oxidation.

Oxidation of the two methionine residues at positions 144 and 145 was shown to abolish calcium-binding. Oxidized CaM is unable to interact properly with and activate the plasma membrane Ca2+-ATPase (4). Similarly oxidized CaM exhibits a reduced affinity for nitric-oxide synthases (5). Biophysical studies indicated that the helical-content of CaM decreases upon methionine oxidation, an effect that is accentuated in the absence of calcium (6). Therefore, calmodulin function can be modulated by the redox status of its methionine residues.

Oxidized protein build-up is a hallmark of aging and neurodegenerative diseases. In particular, oxidized CaM has been detected in aged bovine brain (7). Accumulation of oxidized proteins is generally prevented either by their disposal by proteasome-dependent degradation (8) or by selective repair achieved by dedicated enzymatic systems. Oxidation of methionine residues leads to both S and R diastereoisomers of methionine sulfoxide (MetO) that can be reduced specifically by the peptide methionine sulfoxide reductases A and B enzymes (MsrA and MsrB), respectively (9–14). This ubiquitous enzymatic repair system is essential to maintain the pool of native proteins. It has also been proposed to play an important role as an antioxidant system by reducing exposed oxidized methionines on the protein surface. In addition to its repair and antioxidant roles, a regulatory function has been attributed to MsrA, because reduction of oxidized CaM by MsrA can partially restore its ability to activate the plasma membrane Ca2+-ATPase (15). Similarly, transient inactivation of the voltage-dependent potassium channel was shown to be modulated by methionine oxidation and reversed upon reduction by MsrA or MsrB (14, 16).

In this study, we have examined the role of methionine redox status in the interaction between mammalian CaM and the adenylate cyclase (AC) toxin from Bordetella pertussis, the causative agent of whooping cough (17, 18). AC is one of the major virulence factors of this microorganism. This toxin is able to enter into eukaryotic target cells where it is activated by the endogenous CaM (19) and synthesizes supraphysiological levels of cAMP (20). Previous studies identified the CaM-dependent catalytic domain within the 400 N-terminal residues of the toxin (21, 22). More recently, it was shown that, in vivo, the AC toxin primarily targets phagocytic cells such as neutrophils and macrophages as a result of its specific binding to the CD11b/CD18 integrin (23). Because neutrophil priming and activation result in the production of reactive oxygen species, potential intracellular oxidation might have important physiological consequences for the CaM-AC interaction. In a pioneering work, Wolff et al. (19) indeed observed that oxidation of CaM abolishes activation of B. pertussis AC.

Here, we report that in vitro oxidation of all nine methionine residues of CaM dramatically reduced its affinity for AC. After treatment with either MsrA or MsrB, oxidized CaM was partially "repaired," exhibiting a maximum of seven reduced methionines out of a total of nine. This partially repaired CaM could nevertheless interact with AC almost as efficiently as the native protein as monitored by binding and activity assays. We further identified methionine residues 109, 124, and 145 as essential for binding to AC and activation of its enzymatic activity. The ability of AC to be fully activated by partially oxidized CaM molecules, as documented here, might contribute to the pathophysiological role of this toxin by allowing it to exert its cytotoxic effects in activated neutrophils.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Plasmids—All DNA manipulations were performed according to standard protocols (24) using XLI-Blue strain (Stratagene) as recipient cells.

Plasmid pDLTCaM41 is a bacterial expression vector for human CaM. The CaM cDNA coding sequence was amplified by PCR from a human cDNA library (Stratagene) by using appropriate primers and subcloned between the NdeI and BamHI restriction sites of plasmid pTrec2 (25). In the recombinant plasmid pDLTCaM41, the CaM gene is expressed under the control of the {lambda} phage PR promoter. The plasmid also encodes the thermosensitive {lambda} repressor cI857 that strongly represses gene transcription at the {lambda} PR promoter at temperature below 32 °C. Expression of CaM is triggered by shifting cells to 37 °C or higher temperature.

Plasmid pTRAC384CS is an expression vector for AC384CS, a truncated form of AC encompassing residues 1–384 of wild-type AC followed by a cysteine and a serine residues (native AC has no cysteine residue). Such a truncated form exhibited the same catalytic activity (specific activity of about 2000 units/mg) and affinity for CaM (half-maximal activation at a CaM concentration of 0.1 nM) as the 399 residues long catalytic domain (data not shown). Plasmid pTRAC384CS was constructed as follows. First, a DNA fragment coding for the AC catalytic domain was amplified by PCR from plasmid pDIA5240 (26) by using appropriate primers and subcloned between the NdeI and EcoRI sites of plasmid pTrec2. Then, the PstI site of the recombinant plasmid obtained, pTRAC1, was eliminated by PstI digestion, T4 polymerase treatment, and re-ligation. In the resulting plasmid, pTRAC{Delta}PstI, the 1.06-kb, SacII/EcoRI fragment was replaced by the corresponding DNA fragment from plasmid pACM384 (26) to yield plasmid pTRAC384. Finally, two complementary oligonucleotides, (5'-GCTAAG-3' and 5'-AATTCTTAGCTGCA-3') were hybridized and ligated into pTRAC384 digested with PstI and EcoRI. In the resulting plasmid pTRAC384CS, a Cys, Ser, and TAA stop codon were inserted immediately downstream to the Pro384 codon of AC. Similarly, two complementary oligonucleotides, 5'-CGAAGTTCTCGCCGGATGTACTGGAAACGGTGCCCGGGAAATAAG-3' and 5'-AATTCTTATTTCCCGGGCACCGTTTCCAGTACATCCGGCGAGAACTT-3', were hybridized and ligated into pTRAC-384CS digested with BstBI and EcoRI to construct plasmid pTRAC-384GK. The latter codes for the first 384 residues of the wild-type AC followed by a glycine and a lysine residues, designated below as AC. The specific activity of this AC polypeptide at saturating CaM concentration was about 2000 units/mg, and it was half-maximally activated by a CaM concentration of about 0.1 nM (see Fig. 6).



View larger version (14K):
[in this window]
[in a new window]
  		FIG. 6.
Fluorescence spectroscopy analysis of oxidized CaM binding to AC. Fluorescence emission spectra of AC (1 µM in 20 mM Hepes-Na, pH 7.5, 1 mM CaCl2) were recorded from 310 to 410 nM using an excitation wavelength of 295 nm. Native CaM or oxidized CaM added to the sample was diluted to the indicated concentrations from stock solutions of 180 µM. All spectra were corrected from background fluorescence. Plain curve 1: 1 µM AC; plain curve 2: 1 µM AC plus 1.8 µM CaMox; plain curve 3:1 µM AC plus 7.2 µM CaMox; and dotted curve:1 µM AC plus 1.8 µM CaMN.

 
Purification of Recombinant ACs and CaM—The Escherichia coli strain BLR (Novagen) transformed with plasmid pTRAC384GK (or pTRAC384CS) was grown at 30 °C in LB medium (27) containing 100 µg/ml ampicillin to an optical density (600 nm) of 0.6–0.8. Then, expression of AC (or AC384CS) was triggered by shifting the growth temperature to 42 °C. After 3 h of growth at 42 °C, the cells were collected by centrifugation (20 min, 6,000 x g, 4 °C), resuspended in 20 mM Hepes-Na, pH 7.5, and disrupted by sonication at 4 °C. The sonicated suspension was centrifuged 15 min at 13,000 x g. The supernatant was discarded, and the pellet was resuspended in 8 M urea, 20 mM Hepes-Na, pH 7.5, and agitated 2–3 h (or overnight) at 4 °C (2 mM 2-mercaptoethanol was also added in the case of AC384CS). After 10 min of centrifugation at 13,000 x g the supernatant ("urea extract"), containing the solubilized AC, was collected. AC was purified by two sequential chromatographic steps on DEAE-Sepharose. The urea extract was first loaded onto a DEAE-Sepharose column equilibrated in 8 M urea, 20 mM Hepes-Na, pH 7.5. In these conditions, AC (or AC384CS) did not bind to the resin and was essentially recovered in the flowthrough. The latter was then diluted five times in 20 mM Hepes-Na, pH 7.5, and applied to a second DEAE-Sepharose column equilibrated in 20 mM Hepes-Na, pH 7.5. AC (or AC384CS) was now retained on the resin, and, after extensive washing with 20 mM Hepes-Na, pH 7.5, the protein could be eluted in a soluble form with 20 mM Hepes-Na, 0.1–0.2 M NaCl. The protein purity was usually higher than 95% as judged by SDS-PAGE. The protein content was determined from absorption spectra using an extinction coefficient at 278 nm of 28,000 M–1·cm–1.

CaM was purified from bacterial extracts of E. coli BLR harboring plasmid pDLTCaM41 grown in LB medium/ampicillin as described above. After sonication, the soluble extract was supplemented with ammonium sulfate to a final concentration of 50% (w/v) and agitated for 30–60 min at 4 °C. After centrifugation (15 min, 13,000 x g, 4 °C), the supernatant was adjusted to pH 4.1–4.5 by addition of glacial acetic acid and agitated for 60 min at 4 °C. After centrifugation (15 min, 13,000 x g, 4 °C), the pelleted proteins were resuspended in 20 mM Hepes-Na, pH 7.5, 0.4 M NaCl, 5 mM CaCl2 and loaded onto a Phenyl-Sepharose (Amersham Biosciences) column equilibrated in the same buffer. After extensive washing with 20 mM Hepes-Na, pH 7.5, 1 mM CaCl2, the bound proteins were eluted with 20 mM Hepes-Na, pH 7.5, 3 mM EGTA. At this stage, CaM is usually more than 95% pure as judged by SDS-PAGE. In some cases, CaM was further purified by an additional stage of DEAE-Sepharose chromatography. CaM concentration was determined by measuring absorbance at 278 nm by using a molar extinction coefficient at 278 nm of 2,800 M–1·cm–1.

Expression and Purification of MsrA and MsrB—Recombinant rat MsrA enzyme was expressed and purified from E. coli as previously described by Petropoulos et al. (28). A human cDNA sequence coding for MsrB originally named Cbs1 (11) was obtained from the clone image data base (www.ATCC.org). The msrB coding region was amplified by PCR using appropriate primers and subcloned into the pET15b plasmid between the restriction sites NdeI and BglII. The recombinant plasmid, pOTB7, was transformed into E. coli BL21({lambda}DE3) (Novagen), and transformants were cultivated in LB medium containing 100 µg/ml ampicillin at 37 °C to an optical density at 600 nm of 0.8. Then, expression of MsrB was triggered by addition of 1 mM isopropyl-1-thio-{beta}-D-galactopyranoside. After 3 h of additional growth at 37 °C, the bacteria were pelleted (5,000 x g, 10 min, 4 °C), resuspended in 20 mM Tris-HCl, pH 8, 0.7 M NaCl (buffer A) containing 20 mM imidazole, and disrupted by sonication. After centrifugation (20,000 x g, 20 min, 4 °C), the supernatant was loaded onto a His-Bind resin affinity chromatography column (Novagen) equilibrated in buffer A containing 20 mM imidazole, pH 8. After washing with buffer A containing 60 mM imidazole, pH 8, the His-tagged MsrB protein was eluted with buffer A containing 250 mM imidazole, pH 8. About 1 mg of His Tag-MsrB was usually obtained from 1 liter of culture.

In Vitro Methionine Oxidation of CaM and CaM·AC—60 µM of purified CaM in 50 mM Mes, pH 5.15, 1 mM MgCl2, 100 mM KCl, was incubated during 23 h with 50 mM hydrogen peroxide. To stop the oxidation reaction, the sample was loaded onto a PD10 chromatography column (Amersham Biosciences) to exchange the buffer against 10 mM sodium phosphate, pH 7, 100 mM NaCl. Then, 20 µM of CaM with 30 µM of AC in a buffer A containing 20 mM Hepes, pH 7.4, 100 mM NaCl, 2 mM CaCl2 was incubated with 50 mM tert-butylhydroperoxide (tBHP) for 23 h at room temperature. The integrity of the complex was verified by 7.5% native PAGE.

Purification of the CaM·AC Complex and Separation of These Two Proteins—After the oxidation reaction, the sample containing the complex was loaded onto a Superdex 75 column (Amersham Biosciences) previously equilibrated in 20 mM Tris-HCl, pH 8, 0.1 M NaCl, and 2 mM CaCl2 (buffer B) flowing at a rate of 0.2 ml·min–1. Three peaks were detected at 215 nm, and the corresponding fractions were collected. Peak 1 is composed of CaM·AC complexes, peak 2 corresponds to AC, and peak 3 is the oxidation reagent tBHP. 8 M urea was added to the CaM·AC sample to disrupt the interaction between CaM and AC. Then, CaM and AC were separated on an anionic resin of Vivapure Q chromatography column (Vivasciences) previously equilibrated in buffer B containing 8 M urea. AC was not retained and flowed through the column; CaM was eluted by addition of buffer A containing 0.5 M NaCl after washing with buffer B.

In Vitro Reduction of Methionine Sulfoxide by MsrA or/and MsrB—60 µM oxidized CaM in 10 mM sodium phosphate, pH 7, 100 mM NaCl was incubated with 1 µM MsrA and/or 4 µM MsrB (MsrA enzyme is four times more efficient catalytically than MsrB) and 15 mM DTT (used as an electron donor to reduce MsrA or MsrB) for 2 h at 37 °C. The reduction efficiency was verified by SDS-PAGE and nanoESI-Q-TOF mass spectrometry.

Electrophoretic Analyses—The oxidation state of CaM was analyzed by denaturing 15% SDS-PAGE (29). Association of CaM with AC was monitored by native gel electrophoresis carried out on 7.5% PAGE in Tris-glycine, pH 8.8. Both denaturing and native gels were stained with G-250 Coomassie Blue.

Proteins Analysis by NanoESI-Q-TOF Mass Spectrometry—The full-length proteins were desalted on a reverse phase C4 ZipTipTM (Millipore). The samples were analyzed in acetonitrile/0.2% aqueous formic acid 1:1 (v/v) by nanoelectrospray (nanoESI) mass spectrometry with a hybrid nanoESI-Q-TOF mass spectrometer (Q-TOF2, Micromass). Direct infusion of the sample was realized with metallized glass needles (medium size, Protana, MDS Sciex) through a Z-SprayTM source (Waters). Mass data were processed using the Masslynx software (Waters) to calculate the molecular mass of each protein and to deconvolute the spectra.

CaM Digestion by Trypsin, Ion Trap, and MALDI-TOF MS Analyses—Native, oxidized, or repaired CaM (1 µg) were digested with 1:40 (w/w) trypsin during 18 h at 37 °Cin0.1 M Tris-HCl, pH 8.5. The digests were desalted on a C18 ZipTipTM (Millipore).

Liquid chromatography-tandem MS (LC-MS/MS) analyses were performed with an electrospray ion-trap mass spectrometer (LCQ Advantage; ThermoElectron) coupled on-line with a Surveyor high performance liquid chromatography system (ThermoElectron). A 150- x 0.18-mm ThermoElectron HyPURITY C18 column (5-µm particle diameter; 190-Å pore size) with a mobile phase of solvent A (0.1% (v/v) formic acid in water) and solvent B (0.1% (v/v) formic acid in 80% (v/v) acetonitrile) was used with a linear gradient of 2 to 60% of mobile phase B over 60 min at a flow rate of 250 µl/min. The flow was split, and 2 µl/min was directed to the column. The electrospray needle was operated with a voltage of 2.5 kV, and the heated desolvation capillary was held at 180 °C. Nitrogen was used as the sheath gas. All scans were acquired in positive ion mode. The mass spectrometer operated in a data-dependent MS/MS mode. The top three most intense ions were selected from the full MS scan (m/z range 300–2000) for MS/MS analysis. The isolation width of the parent ions was set to 2 m/z units with 35% normalized collision energy. An m/z ratio for an ion that had been selected for fragmentation was placed in a list and dynamically excluded from further fragmentation for 1 min. Proteins were identified automatically by the computer program TurboSEQUEST (ThermoElectron).

Aliquots of 0.5 µl of each digest in aqueous acidic solution were spotted and mixed on the stainless steel target with 0.5 µl of the supernatant of a saturated solution of 2,5-dihydroxybenzoic acid in ethanol/0.1% aqueous trifluoroacetic acid 1:9 (v/v). Matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectra were obtained using a DE-STR Voyager mass spectrometer (Applied Biosystems). Analyses were realized in reflectron-positive mode with an accelerating voltage of 20 kV, a delayed extraction time of 210 ns, and an average of 200 laser pulses for each acquisition on the mass range 600–5,000 Da. MALDI-TOF spectra were processed using Data Explorer software (Applied Biosystems) and m/z-Knexus edition freeware (Proteometrics, Genomic Solutions Inc.). The calibration was first performed in external mode using a standard peptides mix.

CaM tryptic digest was also analyzed by nanoscale capillary liquid chromatography-tandem mass spectrometry (LC-MS/MS) using 2.5 µl (2 µg) of the digest diluted in 6 µl of buffer A (H2O/acetonitrile/formic acid, 96/4/0.1, v/v). A capillary LC system (Famos-Switchos-UltiMate Dionex, LC Packings, Amsterdam, Netherlands) was used online with a hybrid nanoESI-Q-TOF mass spectrometer (Q-TOF2, Waters, Micromass, Manchester, UK). Chromatographic separations were conducted on a reverse-phase capillary column (Pepmap C18, 75 µm inner diameter, 15-cm length, LC Packings) with a 200-nl/min flow. The gradient profile used consisted of a linear gradient from 100% A to 45% B (H2O/acetonitrile/formic acid, 10/90/0.085, v/v) in 60 min, followed by a linear gradient to 100% B over 15 min. Data acquisition was realized in automatic mode switching between the survey acquisition in MS mode and fragmentation acquisition in MS/MS mode on the four most intense ions detected in the former survey scan. Mass data collected during an LC-MS/MS analysis were processed and converted into a pkl file using the MasslynxTM software (Micromass) to be submitted to the search software Mascot (www.matrixscience.com/). Peptide identifications were obtained by comparison of experimental data to the NCBInr mammalian data base, allowing carbamidomethylation of cysteines and oxidation of methionines as partial modification without any restriction on taxonomy. The unidentified species that have undergone fragmentation were subsequently checked manually from the raw data.

Site-specific Biotinylation of AC384CS—Purified AC384CS was biotinylated on its unique cysteine residue (native AC has no cysteine residue) by incubating the protein in 8 M urea, 20 mM Hepes-Na, pH 7.5, with biotin-maleimide (Molecular Probes) for 2 h at room temperature. The reaction was quenched with 10 mM DTT, and the mixture was diluted five times before loading on a small CaM-Sepharose (Sigma) column. After an extensive wash in coupling buffer, the biotinylated AC384CS was eluted in 8 M urea, 20 mM Hepes-Na, pH 7.5 and kept at –20 °C. Mass-spectrometry analysis confirmed that a single biotin-maleimide molecule had been incorporated on AC384CS. The selectivity of the labeling on cysteine 385 of AC384CS was strongly supported by the observation that purified AC, lacking cysteine residues, was not biotinylated when incubated in the same conditions with biotin-maleimide.

CaM Binding Assays by Competitive Enzyme-linked Immunosorbent Assay—96-well microtiter plates were coated with 50 µlof5 µg/ml CaM in 0.05 M carbonate buffer, pH 9.8, and incubated overnight at 4 °C. All further incubations were performed at room temperature. Wells were washed three times in HBST buffer (20 mM Hepes-Na, pH 7.5, 150 mM NaCl, 1 mM CaCl2, 0.1% Tween 20 (v/v)). 300 µl of blocking buffer (5 mg/ml bovine serum albumin in HBST buffer) was added to each well, and the plates were incubated for 1–2 h. Wells were washed as before, and 40 µl of serial dilutions of native, oxidized, or repaired CaM in HBST buffer was added in each well, followed by the addition of 10 µl of 0.4 nM biotinylated AC384CS in HBST buffer. After 3 h of incubation at 30 °C, wells were washed three times with HBST buffer, and 50 µlof extravidin-alkaline phosphatase conjugate (Sigma, 1:6000 dilution in HBST buffer) was added. After 1-h incubation at 30 °C, the wells were washed as before, and 50 µl of substrate solution (1 mg/ml p-nitrophenyl phosphate in 50 mM Tris-HCl, pH 9.5, 5 mM MgCl2) was added per well. Plates were incubated at room temperature until a yellow color developed, usually between 15 min and 2 h. Absorbance readings at 405 nm were recorded using a Tecan microplate reader.

Surface Plasmon Resonance Recording—Binding experiments were performed on a BIAcore biosensor 2000 system (Biacore AB). All experiments were performed at 25 °C, at a flow rate of 5 µl/min using a running buffer containing 20 mM Hepes-Na, pH 7.5, 1 mM CaCl2 and 0.005% Tween 20 (v/v). Between binding cycles the coated surfaces were regenerated by injection of 10 µl of 20 mM NaOH, followed by injection of 10 µl of 20 mM HCl. The first flow channel of the sensor chip did not contain any immobilized ligand and served as a reference surface.

In preliminary experiments, we attempted to immobilize AC by direct coupling to a 1-ethyl-3-[3-dimethylaminopropyl]carbodiimide-activated sensorchip. However, we repeatedly observed irreversible losses of signal upon regeneration of the matrix. This prompted us to immobilize AC through a biotin tag onto a streptavidin-coated sensorchip (Sensorchip SA, BIAcore). Biotinylated AC was immobilized (up to 2500 relative units) by injection of 20 µl of 100 nM AC384CS-biotin in 20 mM Hepes-Na, pH 7.5, 1 mM CaCl2 over an SA-coupled sensorchip (BIA-core), and CaM oxiforms were used as analytes. The sensorgrams were analyzed using the BIAevaluation 3.1 software (BIAcore).

Fluorescence Spectroscopy—Fluorescence measurements were recorded on a PerkinElmer Life Sciences LS-5B luminescence spectrometer thermostated at 25 °C, using a 0.5x 1 UV-grade quartz cuvette (sample volume of 1 ml). Emission spectra ({lambda}exc = 295 nm) of AC (1 µM in 20 mM Hepes-Na, pH 7.5, 1 mM CaCl2) were recorded from 310 to 410 nM (both excitation and emission bandwidths were set at 5 nm) at 60 nm/min. Native CaM or oxidized CaM added to the sample were diluted from stock solutions (in 20 mM Hepes-Na, pH 7.5) of 180 µM. All spectra were corrected from background fluorescence.

CaM Activation of AC Enzymatic Activity—Adenylate cyclase activity was measured as previously described (30). AC was diluted to 0.1 nM in 45 µl of reaction mixture containing 50 mM Tris-HCl, pH 8, 6 mM MgCl2, 0.1 mM CaCl2, 0.1 mM [3H]cAMP (6000 cpm/assay) and various concentrations of CaM or CaM oxiforms. Reaction was initiated by addition of 5 µl of 20 mM [{alpha}-33P]ATP (1–5 x 105 cpm/assay). After 10 min of incubation at 30 °C, the reaction was stopped by addition of 100 µl of 0.5 N HCl. After neutralization with 100 µl of 0.5 N NaOH, the mixture was applied on an Al2O3 column to separate [33P]cAMP (recovered in the eluate) from [{alpha}-33P]ATP (retained on the column). One unit of AC activity corresponds to 1 µmol of cAMP formed per minute at 30 °C and pH 8. A maximal specific activity of about 2000 units/mg of AC (or AC384CS) was measured in the presence of 2 µM native CaM.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
To study the influence of calmodulin (CaM) oxidation on interaction between mammalian CaM and B. pertussis adenylate cyclase (AC), we analyzed the interaction of a truncated form of B. pertussis AC, encompassing the residues 1–384 of wild-type toxin, with various forms of CaM: native (CaMN), oxidized (CaMox) and CaMox reduced by treatment with either MsrA (CaM/MsrA), MsrB (CaM/MsrB), or both MsrA and MsrB (CaM/MsrA + MsrB).

In Vitro Oxidation of Calmodulin Methionines Drastically Reduces Its Affinity for Adenylate Cyclase—Recombinant CaM was overproduced in E. coli and purified by hydrophobic chromatography to homogeneity as shown by SDS-PAGE (Fig. 1, lane 1). MS analysis confirmed the homogeneity of the sample: the mass spectrum showed a single peak of 16,709 Da corresponding to the native form (Fig. 2A) with nine fully reduced methionine residues. When the CaM methionines were specifically oxidized in vitro in the presence of hydrogen peroxide at pH 5, the oxidized protein appeared by MS analysis (Fig. 2B), as a major peak with a molecular mass of 16,853 Da. The molecular mass increase (+144 Da) corresponds precisely to the addition of one oxygen atom to each of the nine methionine residues upon conversion of methionine to MetO. As shown in Figs. 1 and 3 (lanes 2), the oxidized CaM (CaMox) migrated slightly differently as compared with the native CaM both on native and denaturing electrophoreses, most likely as a consequence of protein conformational alterations.



View larger version (49K):
[in this window]
[in a new window]
  		FIG. 1.
SDS-PAGE analysis of different CaM oxiforms. 5 µg of CaMN (lane 1), CaMox (lane 2), CaM/MsrA (lane 3), CaM/MsrB (lane 4) and CaM/MsrA+MsrB (lane 5) were separated on a 15% (w/v) SDS-polyacrylamide gel (Laemmli (29)) containing 2 mM calcium. After migration the gel was stained with G-250 Coomassie Blue.

 



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 2.
Mass spectrometry analysis of the different CaM oxiforms. The different oxiforms of CaM: A, CaMN; B, CaMox; C, CaM/MsrA D, CaM/MsrB; and E, CaM/MsrA + MsrB were analyzed by nanoESI-Q-TOF as described under "Experimental Procedures." The peaks relative intensity (%) was recorded as a function of the ratio mass/charge (m/z). Theoretical and experimental masses of all protein species are listed.

 



View larger version (75K):
[in this window]
[in a new window]
  		FIG. 3.
Native gel analysis of complex formation between CaM and AC. The different forms of CaM were diluted in 20 mM Hepes, pH 7.5, 1 mM CaCl2 to a final concentration of 30 µM and mixed with 20 µM AC. After 30 min of incubation at room temperature the different samples were then separated on a native polyacrylamide gel (7.5%) containing 1 mM calcium. After migration, the gel was stained with G-250 Coomassie Blue. Lane 1, CaMN; lane 2, CaMox; lane 3, AC·CaMN; lane 4, AC·CaMox; lane 5, AC·CaM/MsrA; lane 6, AC·CaM/MsrB; lane 7, AC·CaM/MsrA + MsrB; and lane 8, AC.

 
Interactions of CaM and its fully oxidized form, CaMox, with AC were then characterized by different approaches. Native gel electrophoresis (Fig. 3) indicated that the recombinant CaM formed a stable complex with AC (lane 3) with an electrophoretic mobility intermediate between that of free CaM (lane 1) and that of free AC (lane 8). In contrast, no similar complex was detected when AC was mixed with the oxidized CaM (lane 4), indicating that the oxidation of methionine residues of CaM essentially abolished its ability to associate with AC in these conditions. Direct interaction between CaM and AC was also monitored by surface plasmon resonance spectroscopy (BIA-core). These experiments revealed that CaM was able to bind to AC immobilized on the surface of a BIAcore sensorchip, whereas binding of CaMox could not be detected in the same conditions (data not shown). Quantitative data about the equilibrium binding constants of the AC·CaM complex were obtained by using competitive binding assays (Fig. 4), because the very slow dissociation of CaM from the immobilized AC precluded an accurate determination of binding affinities by the BIAcore technique. Competitive binding assays (Fig. 4) indicated that the apparent binding constant of the oxidized CaM for the biotinylated AC variant (K0.5 > 10,000 nM) was more than 5000 times higher than that of the native CaM (K0.5 = 1.8 nM ± 0.2).



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 4.
Quantitative analysis of CaM oxiforms binding to AC by a competitive enzyme-linked immunosorbent assay. AC384CS-biotin (0.4 nM) was incubated for 2 h in CaM-coated microtiter plates in the presence of the indicated concentrations of CaMN (closed circles), CaMox (closed squares), CaM/MsrA (closed downward triangles), CaM/MsrB (closed upward triangles), and CaM/MsrA + MsrB (closed diamonds). After washing, the fraction (f) of AC384CS-biotin bound to CaM-coated microtiter plates was revealed by extravidin-alkaline phosphatase staining as described under "Experimental Procedures" (100% refers to the absorbance read in the absence of CaM competitors). Data points were fitted to the equation, f = 100 – [100 x C/(K0.5 + C)], where K0.5 represents the apparent dissociation constant of the equilibrium, and C is the concentration of competitors.

 
We then analyzed the capability of the oxidized CaM to activate AC enzymatic activity. As shown in Fig. 5, whereas native CaM stimulated AC at subnanomolar concentrations, micromolar concentrations of oxidized CaM were required to achieve only a partial activation of AC activity, indicating that the affinity of oxidized CaM for AC was at least 5000 times lower than that of native CaM. It is noteworthy that the concentration of CaM required to achieve half-maximal activation (EC50) of AC enzymatic activity (Fig. 5) was about 10-fold lower than the apparent equilibrium constants of the AC·CaM complex determined by the competitive binding assays (Fig. 5). This variation likely results from the different settings of these two assays: heterogenous phase binding assay with a biotinylated AC versus interaction in solution with a non-modified enzyme (see "Experimental Procedures").



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 5.
AC activation by CaM oxiforms. Adenylate cyclase activity of purified AC (final concentration of 0.02 nM) was measured in the presence of the indicated concentrations CaMN, CaMox, CaM/MsrA, CaM/MsrB, or CaM/MsrA + MsrB as described under "Experimental Procedures" and expressed as a percentage of maximal activity measured in the presence of 2 µM CaMN. Symbols: CaMN (closed circles), CaMox (closed squares), CaM/MsrA (closed downward triangles), CaM/MsrB (closed upward triangles), and CaM/MsrA + MsrB (closed diamonds).

 
From the above described experiments, we could not determine whether the AC activation, observed in the presence of micromolar concentrations of oxidized CaM species (Fig. 5), represented an intrinsic capability of the fully oxidized molecule to interact with and activate AC or whether this was due to contamination of the preparation by not fully oxidized CaM. To clarify this point, we analyzed the binding of oxidized CaM to AC by fluorescence spectroscopy (Fig. 6). AC has two tryptophan residues; one of these, Trp-242, is located in the main CaM binding site. It has been shown previously that binding of CaM (which has no Trp residues) to AC induces a blue shift in the Trp fluorescence emission spectra (22). This most likely reflects the insertion of Trp-242 within a hydrophobic pocket of CaM. As shown in Fig. 6, the fully oxidized CaM also triggered a significant change in the AC fluorescence emission spectra, although the blue shift was less pronounced than that observed upon binding of native CaM. These data clearly indicated that oxidized CaM directly interacted with AC as the minor fraction of uncompletely oxidized CaM species could not account for the changes of the fluorescence properties of AC in these conditions (i.e. 1 µM AC versus 1.8 µM oxidized CaM).

Altogether, these data indicate that the redox state of CaM methionine residues has a critical role for CaM binding to AC as the oxidation of methionine residues dramatically reduced the affinity of CaM for AC. Importantly, this implies that, under physiological concentrations of CaM, the fully oxidized CaM would be unable to activate significantly the enzymatic activity of the AC toxin.

The Peptide Methionine Sulfoxide Reductase System Restores Functional Interaction of CaMox to AC—We then studied the effect of the peptide methionine sulfoxide reductases, MsrA and MsrB, on the reduction of oxidized CaM. CaMox was incubated with either MsrA or MsrB, expressed and purified from E. coli (see "Experimental Procedures"), using DTT as an electron donor in the reduction process. Reduction of CaMox upon incubation with MsrA (CaM/MsrA) or MsrB (CaM/MsrB) could be clearly evidenced by SDS-PAGE analysis (Fig. 1): the repaired CaM migrated as several distinct bands exhibiting different electrophoretic mobilities, intermediate between that of native CaM and that of oxidized CaM. These different species/bands most likely correspond to partially and heterogeneously reduced CaM molecules, having various combinations of oxidized and reduced methionines, with different conformations and/or affinities for calcium. Partial reduction of MetO residues of CaM exposed to either MsrA or MsrB was indeed confirmed by mass spectrometry that revealed, in both cases, a heterogeneous population with a maximum of seven reduced methionines (Fig. 2, C and D). This result could be easily explained by the fact that, upon oxidation, methionine residues of CaM should be randomly converted to either the S or the R diastereoisomer of MetO, and, in each case, only the specific diastereoisomer could be reduced in the presence of only one form of the enzymes, MsrA or MsrB.

The ability of these heterogeneous populations of CaM to associate with AC was analyzed by native gel electrophoresis, which revealed that CaM/MsrA and CaM/MsrB can form a stable complex with AC (Fig. 3, lanes 5 and 6). BIAcore analysis also demonstrated that CaM/MsrA and CaM/MsrB could bind specifically to AC-biotin immobilized on sensorchip (data not shown). Competition assays (Fig. 4) indicated a slightly lower affinity of the partially reduced forms of CaM for AC-biotin, as compared with that of the native activator (K0.5 of 10 and 8 nM for CaM/MsrA and CaM/MsrB, respectively, versus a K0.5 of 1.8 nM for CaMN). AC activation assays confirmed a 4- to 5-fold decrease in the apparent affinity of CaM/MsrA and CaM/MsrB for AC, as compared with that of the native CaM. It is noteworthy that both partially reduced CaM/MsrA and CaM/MsrB species, at saturating concentration (i.e. > 100 nM), activated AC to an identical specific activity as the wild-type CaM (i.e. about 2000 µmol of cAMP·min–1·mg–1).

Altogether these results indicated that incomplete reduction of oxidized CaM by either MsrA or MsrB was sufficient to almost fully restore AC activation with only a slight decrease of the apparent affinity. Importantly, the fact that the partially reduced forms, CaM/MsrA or CaM/MsrB, exhibited affinities for AC and activation properties close to that of native CaM, although they still contained numerous oxidized methionine residues, suggest that only few of the CaM methionine residues are critical to establish functional interaction with AC.

Complete reduction of methionine residues of oxidized CaM was achieved upon incubation in the presence of both MsrA and MsrB: the resulting CaM/MsrA + MsrB migrated like the native CaM on SDS-PAGE (Fig. 1). It suggests that all nine MetO had been reduced, and indeed a single species with a molecular mass of 16,709 Da corresponding to the native CaM (Fig. 2E) was detected by mass spectrometry. AC binding to CaM/MsrA + MsrB was enhanced as compared with CaM/MsrA or CaM/MsrB, although it did not fully recover the potency of native untreated CaM (Figs. 4 and 5). This might be a consequence of the different treatments to which the polypeptides have been subjected.

The C-terminal Methionine Residues of CaM Are Essential for the Interaction with AC—We then attempted to identify the specific methionine residues of CaM that are essential for interaction with AC. CaM was completely oxidized by hydrogen peroxide and then partially reduced by treatment with MsrA alone, as described above. This heterogeneous population of CaM with mixed oxidized/reduced methionines was then chromatographed onto AC covalently bound to a Sepharose resin, to separate CaM molecules competent for AC binding from those unable to do so. Unbound CaM molecules were collected and, after washing, the bound CaM was eluted from AC-Sepharose resin in denaturing conditions (8 M urea in Hepes buffer). Both unbound and bound fractions were then digested by trypsin and analyzed by MALDI-TOF to identify, among the nine methionine residues, those that are important for the interaction with AC. As shown in Table I, the methionine residues located in the N terminus moiety of CaM (i.e. Met-36, Met-51, Met-72, Met-73, and Met-77) were found to be either reduced or oxidized, in both the unbound and bound fractions. That methionine sulfoxides were found in each N terminus peptide derived from the bound fraction, strongly suggested that the oxidation of the corresponding residues did not prevent CaM-AC binding. On the other hand, methionine residues in the C terminus part of CaM were found mostly oxidized in the unbound fraction (Table I), suggesting that the oxidized state of these methionine residues (i.e. Met-109, Met-124, Met-144, and Met-145) might have prevented the association of CaM with AC. Unfortunately, the peptides corresponding to the C-terminal part of CaM in the bound fraction were not detected by mass spectrometry.


View this table:
[in this window]
[in a new window]
  		TABLE I
Oxidation status of methionine residues in partially reduced CaM fractionated by AC-Sepharose chromatography

The heterogeneous population of oxidized CaM repaired by MsrA (CaMA) was chromatographed on an AC-Sepharose column. Both unbound CaM and bound CaM fractions were then digested by Trypsin, and the resulting peptide mixtures were analyzed by MS (MALDI-TOF).

 
To further characterize the methionine residues of CaM that participate in the interaction with AC, we then explored a complementary approach and attempted to identify the methionine residues that might be protected from oxidation when CaM is complexed to AC. CaM was mixed with an excess of AC (i.e. molar ratio of 1:1.5) and oxidized for 23 h at pH 7 with tBHP, a mild oxidizing reagent selective for solvent-exposed methionines in proteins (31). Native gel electrophoresis (not shown) showed that the AC·CaM complex was not dissociated by this treatment indicating that, as expected, the methionine residues critical for interaction had been protected from oxidation. In the same experimental conditions, when free CaM was exposed to tBHP, an almost complete oxidation of methionine residues was observed. The tBHP-oxidized AC·CaM was submitted to a gel filtration chromatography to separate the AC·CaM complex from the excess free AC (Fig. 7A). The mixture was resolved in two separated peaks: peak 1 corresponded to the oxidized AC·CaM complex, and peak 2 corresponded to the excess unbound AC. The oxidized AC·CaM complex isolated in peak 1 was found to be enzymatically active (with a specific activity higher than 50% of the native complex) indicating that, within the complex, the essential methionine residues on both proteins had been protected from oxidation.



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 7.
Characterization of the CaM·AC complex oxidized by tBHP. A, CaM (20 µM) was mixed with AC (30 µM) and incubated for 23 h with 50 mM tBHP. The mixture was then loaded on a 25-ml Superdex 75 column equilibrated in 20 mM Hepes, pH 7.5, 100 mM NaCl, and eluted at a flow rate of 0.2 ml/min. Elution was monitored by the absorbance at 215 nm (upper trace). The chromatograms shown below correspond to similar Superdex 75 chromatographies of the native AC·CaM complex, native AC, or oxidized CaM (from top to bottom, respectively), which were performed as control experiments. Native CaM exhibited the same elution profile than the oxidized CaM (not shown). Peak 1 corresponds to the AC·CaM complex, peak 2 to the excess of AC, and peak 3 to the oxidation reagent, tBHP. B, SDS-PAGE (15%) analysis of CaMN (lane 1), CaM oxidized with tBHP (lane 2), ACN (lane 3), AC oxidized with tBHP (lane 4), CaM·AC oxidized with tBHP (lane 5), peak 2 of Superdex 75 (lane 6), and peak 1 of Superdex 75 (lane 7).

 
The oxidized CaM molecules from peak 1 fraction were then separated from AC by chromatography on an anionic-exchanger matrix (see "Experimental Procedures"). ESI-Q-TOF analysis revealed that the recovered CaM contained at most 3–4 reduced Met (data not shown). This sample was further digested by trypsin, and the resulting peptides were analyzed by mass spectrometry. As shown in Table II, the MS analysis revealed that the methionine residues from the N-terminal moiety of CaM were mostly oxidized. This result corroborated our previous observations that the N terminus residues of CaM are not essential for the interaction with AC. It also showed that, within the AC·CaM complex, these residues are accessible to oxidation by tBHP. In contrast, specific methionine residues from the C-terminal part of CaM were protected from the oxidation when the protein was bound to AC. As shown in Table II, the MS analysis by MALDI-TOF or LC-MS/MS/ion trap indicated that both Met-109 and Met-124 were completely reduced, whereas Met-144 and Met-145 displayed an heterogeneous oxidation status. The peptide encompassing residues 127–148 and containing both Met-144 and Met-145 was detected as two ions having molecular masses corresponding to either two reduced Met or to a combination of one oxidized Met and one reduced Met. To determine more precisely the oxidation status of these last two Met residues, the tryptic digests of the AC·CaM complex were analyzed by LC-MS/MS (Q-TOF). Direct mass spectrometry sequencing of peptide 127–148 containing a single methionine sulfoxide revealed that Met-145 was fully reduced, whereas Met-144 was found in an oxidized state (Table II). All together, our results clearly indicated that Met-144 was largely exposed to oxidation within the AC·CaM complex, whereas Met-109, Met-124, and Met-145 were protected. This suggests that these latter CaM methionines residues are the ones that are critical for functional interaction with AC.


View this table:
[in this window]
[in a new window]
  		TABLE II
Oxidation status of CaM methionine residues in the oxidized AC·CaM complex

Oxidized CaM was isolated from the Superdex 75 peak 1 fraction (Fig. 7) on a Vivapure Q column (see "Experimental Procedures"), digested by trypsin, and analyzed by MS (MALDI-TOF) or MS/MS (Q-TOF and/or Ion Trap) to determine the redox state of methionines.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
The interaction of AC with mammalian CaM is critical for the cytotoxicity of this essential toxin from B. pertussis. The AC toxin contributes to the inactivation of the host defense mechanisms and is therefore critical for allowing the successful establishment of the bacterial infection (18, 32). In this study, we examined the functional effect of CaM oxidation on the interaction with and activation of AC. We showed that oxidation of methionine residues of CaM resulted in a drastic decrease in affinity for AC. A high affinity interaction with AC was restored upon reduction of the methionine sulfoxide residues by the peptide methionine sulfoxide reductases, MsrA and MsrB. Reduction of oxidized CaM was incomplete when carried out in the presence of only one of these two enzymes, because oxidation of methionine residues leads to a mixture of S and R diastereoisomers of methionine sulfoxide (Met-S(O) and Met-R(O)), which are reduced specifically by MsrA and MsrB, respectively. This partial reduction resulted in a heterogeneous population of CaM molecules harboring from one to seven methionine sulfoxides (see Fig. 2, C and D). This mixed population of partially reduced CaM was nevertheless able to fully activate AC with a slight decrease of in apparent affinity as compared with native CaM.

Complete reduction of all methionine sulfoxide residues of the oxidized CaM could be achieved in the presence of both MsrA and MsrB enzymes, as shown by mass spectrometry study of the repaired protein (Fig. 2E). The full reduction of oxidized CaM restored a native-like conformation, as indicated by both native and denaturing gel electrophoresis analyses (Figs. 1 and 3). Yet, the affinity for AC of this fully repaired CaM was still slightly lower than that of untreated CaM. The fact that the heterogenous population of partially oxidized CaM obtained in the presence of MsrA or MsrB could fully activate AC with a minor decrease in apparent affinity, as compared with the native activator, suggests that only a small number of the methionine residues have to be in a reduced form to allow functional interaction of CaM with AC.

We took advantage of this heterogeneous population of CaM molecules, obtained after partial reduction with MsrA, to identify the methionine residues that might be critical for interaction with AC. An AC-affinity chromatography was used to separate CaM molecules able to bind AC from those that could not. Mass spectrometry analysis revealed that, in CaM molecules unable to interact with the immobilized AC, the methionine residues from the C-terminal domain were all oxidized. This suggests that the oxidized state of these residues (i.e. Met-109, Met-124, Met-144, and Met-145) prevented CaM binding to AC. These data were confirmed by a different experimental approach in which we identified the methionine residues of CaM that became protected from oxidation when associated to the AC target. The CaM·AC complex was found to be resistant to oxidation when exposed to tert-butylhydroperoxide. Mass spectrometry analysis revealed that methionines 36, 51, 71, 72, 76, and 144 of CaM were mainly oxidized showing that (i) these Met residues were accessible to oxidation within the complex and (ii) methionine sulfoxide at these positions did not disrupt the AC-CaM interaction. In contrast, Met-109, Met-124, and Met-145 of CaM were found only in their reduced state, indicating that they were preserved from oxidation within the AC·CaM complex.

Taken together with the observation that complete oxidation of CaM methionine residues drastically reduced the affinity of this activator for AC, these data strongly suggest that methionines at positions 109, 124, and 145 are involved in the formation of a high affinity complex with AC. These results highlight the essential role of the C-terminal moiety of CaM in the binding to and activation of AC, in good agreement with those obtained by Wolff et al. (33) who showed that the C-terminal half of CaM was able to activate AC as efficiently as full-length protein with only a 10-fold decrease in affinity. More recently, Shen et al. (34) confirmed that the C-terminal domain of CaM could activate AC, but at variance with Wolff et al. (33), they found that the N-terminal moiety of CaM also exhibited a potent AC activating capability. Yet, our data indicated that, when native CaM is complexed to AC, the methionine residues from the C-terminal part are preferentially protected from oxidation as compared with those from the N-terminal half. This suggests that the C-terminal domain contributes the main interface with AC while the N terminus might be more exposed. Shen et al. (34) also reported that modification of the C-terminal CaM moiety, by engineering an intramolecular disulfide bond, affected more dramatically its interaction with AC than the corresponding modification within the N-terminal part.

The role of individual methionine in the functional interaction of CaM with various target enzymes has been examined in several recent studies. The Squier group previously showed that oxidation of CaM resulted in an inability to fully activate the plasma membrane Ca2+-ATPase, and recently, Bartlett et al. (4) used a mutational approach to demonstrate that oxidation of Met-144 is largely responsible for the decreased extent of enzyme activation. The same methionine residues at positions 144 and 145 of CaM were also shown to contribute to the interaction with both neuronal nitric-oxide synthase (NOS) and endothelial NOS-constitutive isoforms of NOS, because their site-specific oxidation resulted in changes in affinity for these enzymes (5). Kondo et al. (35) also highlighted the critical role of the hydrophobic amino acid side chain at position 144 of CaM in NOS activation. Balog et al. (36) have studied the effect of CaM oxidation on the regulation of the skeletal muscle ryanodine receptor Ca2+ release channel (RyR1) and found that complete oxidation of CaM abolished its functional interaction with RyR1. By mutating individual CaM methionine residues to glutamine (a mutation that somewhat mimics the oxidation of methionine by introducing an oxygen atom at the same position in the side chain as the sulfoxide), they demonstrated that Met-109 and Met-124 were critical for functional interaction of CaM with the ryanodine receptor.

The fact that the B. pertussis AC is efficiently activated by partially oxidized CaM might have important implications for the biological function of this essential toxin. Indeed, the AC toxin, through its specific interaction with the CD11b/CD18 integrin, primarily targets phagocytic cells like neutrophils and macrophages (23, 32). Because neutrophil priming and activation result in the production of reactive oxygen species that may cause intracellular oxidation, the ability of AC to be fully activated by partially oxidized CaM molecules should allow the toxin to exert its pathophysiological effects in activated neutrophils. Moreover, our study has revealed that the methionine residues of CaM implicated in the recognition and activation of the AC toxin are essentially the same as those that are important for the regulation of some other mammalian enzymes. The oxidation state of these critical methionine residues should be under a tight control by the MsrA/MsrB enzymes, given the central role of CaM in cell function. The B. pertussis AC might have evolved to recognize the same structural features in CaM to be able to find a functional activator in the host cells it invades.


   FOOTNOTES
 
* This work was supported in part by the Institut Pasteur and the Centre National de la Recherche Scientifique (CNRS URA 2185, Biologie Structurale et Agents Infectieux) and the French Ministère de l'Education Nationale, de la Recherche et de la Technologie (EA 3106, Biologie et Biochimie Cellulaire du Vieillissement). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

§ Supported by fellowships from the French Ministère de l'Education Nationale, de la Recherche et de la Technologie and the Agence Nationale de Recherche sur le Sida (ANRS). Back

** To whom correspondence may be addressed. Tel./Fax: 33-1-44-27-82-34; E-mail: bfriguet{at}paris7.jussieu.fr. {ddagger}{ddagger} To whom correspondence may be addressed. Tel.: 33-1-45-68-84-00; Fax: 33-1-40-61-30-43; E-mail: ladant{at}pasteur.fr.

1 The abbreviations used are: CaM, calmodulin; MetO, methionine sulfoxide; AC, adenylate cyclase; MsrA, peptide methionine sulfoxide reductase A; MsrB, peptide methionine sulfoxide reductase B; CaMN, native CaM; CaMox, oxidized CaM; CaM/MsrA, oxidized calmodulin reduced with MsrA; CaM/MsrB, oxidized calmodulin reduced with MsrB; CaM/MsrA + MsrB, oxidized calmodulin reduced with MsrA and MsrB; MS, mass spectrometry; MALDI-TOF, matrix-assisted laser desorption/ionization time-of-flight; nanoESI-Q-TOF, nanoelectrospray ionization quadrupole time-of-flight; LC-MS/MS, liquid chromatography/tandem mass spectrometry; tBHP, tert-butylhydroperoxide; Mes, 4-morpholineethanesulfonic acid; DTT, dithiothreitol; NOS, nitric-oxide synthase. Back


   ACKNOWLEDGMENTS
 
We thank A. Ullmann for critical reading of the manuscript. We are also grateful to Pr. J. Rossier (CNRS UMR 7637 Neurobiologie et Diversité Cellulaire) for his support and helpful discussion and to V. Labas, I. Haddad, and J. P. Le Caer for mass spectrometry analyses.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Levine, R. L., Berlett, B. S., Moskovitz, J., Mosoni, L., and Stadtman, E. R. (1999) Mech. Ageing Dev. 107, 323–332[CrossRef][Medline] [Order article via Infotrieve]
  2. Yin, D., Kuczera, K., and Squier, T. C. (2000) Chem. Res. Toxicol. 13, 103–110[CrossRef][Medline] [Order article via Infotrieve]
  3. Chin, D., and Means, A. R. (2000) Trends Cell Biol. 10, 322–328[CrossRef][Medline] [Order article via Infotrieve]
  4. Bartlett, R. K., Bieber Urbauer, R. J., Anbanandam, A., Smallwood, H. S., Urbauer, J. L., and Squier, T. C. (2003) Biochemistry 42, 3231–3238[CrossRef][Medline] [Order article via Infotrieve]
  5. Montgomery, H. J., Bartlett, R., Perdicakis, B., Jervis, E., Squier, T. C., and Guillemette, J. G. (2003) Biochemistry 42, 7759–7768[CrossRef][Medline] [Order article via Infotrieve]
  6. Gao, J., Yin, D. H., Yao, Y., Sun, H., Qin, Z., Schoneich, C., Williams, T. D., and Squier, T. C. (1998) Biophys. J. 74, 1115–1134[Abstract/Free Full Text]
  7. Michaelis, M. L., Bigelow, D. J., Schoneich, C., Williams, T. D., Ramonda, L., Yin, D., Huhmer, A. F., Yao, Y., Gao, J., and Squier, T. C. (1996) Life Sci. 59, 405–412[CrossRef][Medline] [Order article via Infotrieve]
  8. Ferrington, D. A., Sun, H., Murray, K. K., Costa, J., Williams, T. D., Bigelow, D. J., and Squier, T. C. (2001) J. Biol. Chem. 276, 937–943[Abstract/Free Full Text]
  9. Weissbach, H., Etienne, F., Hoshi, T., Heinemann, S. H., Lowther, W. T., Matthews, B., St John, G., Nathan, C., and Brot, N. (2002) Arch. Biochem. Biophys. 397, 172–178[CrossRef][Medline] [Order article via Infotrieve]
  10. Grimaud, R., Ezraty, B., Mitchell, J. K., Lafitte, D., Briand, C., Derrick, P. J., and Barras, F. (2001) J. Biol. Chem. 276, 48915–48920[Abstract/Free Full Text]
  11. Kuschel, L., Hansel, A., Schonherr, R., Weissbach, H., Brot, N., Hoshi, T., and Heinemann, S. H. (1999) FEBS Lett. 456, 17–21[CrossRef][Medline] [Order article via Infotrieve]
  12. Lescure, A., Gautheret, D., Carbon, P., and Krol, A. (1999) J. Biol. Chem. 274, 38147–38154[Abstract/Free Full Text]
  13. Kryukov, G. V., Kumar, R. A., Koc, A., Sun, Z., and Gladyshev, V. N. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 4245–4250[Abstract/Free Full Text]
  14. Jung, S., Hansel, A., Kasperczyk, H., Hoshi, T., and Heinemann, S. H. (2002) FEBS Lett. 527, 91–94[CrossRef][Medline] [Order article via Infotrieve]
  15. Sun, H., Gao, J., Ferrington, D. A., Biesiada, H., Williams, T. D., and Squier, T. C. (1999) Biochemistry 38, 105–112[CrossRef][Medline] [Order article via Infotrieve]
  16. Ciorba, M. A., Heinemann, S. H., Weissbach, H., Brot, N., and Hoshi, T. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 9932–9937[Abstract/Free Full Text]
  17. Mock, M., and Ullmann, A. (1993) Trends Microbiol. 1, 187–192[CrossRef][Medline] [Order article via Infotrieve]
  18. Ladant, D., and Ullmann, A. (1999) Trends Microbiol. 7, 172–176[CrossRef][Medline] [Order article via Infotrieve]
  19. Wolff, J., Cook, G. H., Goldhammer, A. R., and Berkowitz, S. A. (1980) Proc. Natl. Acad. Sci. U. S. A. 77, 3841–3844[Abstract/Free Full Text]
  20. Confer, D. L., and Eaton, J. W. (1982) Science 217, 948–950[Abstract/Free Full Text]
  21. Ladant, D., Michelson, S., Sarfati, R., Gilles, A. M., Predeleanu, R., and Barzu, O. (1989) J. Biol. Chem. 264, 4015–4020[Abstract/Free Full Text]
  22. Gilles, A. M., Munier, H., Rose, T., Glaser, P., Krin, E., Danchin, A., Pellecuer, C., and Barzu, O. (1990) Biochemistry 29, 8126–8130[CrossRef][Medline] [Order article via Infotrieve]
  23. Guermonprez, P., Khelef, N., Blouin, E., Rieu, P., Ricciardi-Castagnoli, P., Guiso, N., Ladant, D., and Leclerc, C. (2001) J. Exp. Med. 193, 1035–1044[Abstract/Free Full Text]
  24. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd Ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
  25. Ray, S., Zozulya, S., Niemi, G. A., Flaherty, K. M., Brolley, D., Dizhoor, A. M., McKay, D. B., Hurley, J., and Stryer, L. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 5705–5709[Abstract/Free Full Text]
  26. Ladant, D., Glaser, P., and Ullmann, A. (1992) J. Biol. Chem. 267, 2244–2250[Abstract/Free Full Text]
  27. Miller, J. (1992) A Short Course in Bacterial Genetics, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
  28. Petropoulos, I., Mary, J., Perichon, M., and Friguet, B. (2001) Biochem. J. 355, 819–825[CrossRef][Medline] [Order article via Infotrieve]
  29. Laemmli, U. K. (1970) Nature 227, 680–685[CrossRef][Med