Portal de Periódicos da CAPES

Siderophore Peptide, a New Type of Post-translationally Modified Antibacterial Peptide with Potent Activity

2004; Elsevier BV; Volume: 279; Issue: 27 Linguagem: Inglês

10.1074/jbc.m400228200

1083-351X

Xavier Thomas, Delphine Destoumieux‐Garzón, Jean Péduzzi, Carlos Afonso, Alain Blond, Nicolas Birlirakis, Christophe Goulard, Lionel Dubost, Robert Thaï, Jean‐Claude Tabet, Sylvie Rebuffat,

Microbial Natural Products and Biosynthesis

Microcin E492 (MccE492, 7886 Da), the 84-amino acid antimicrobial peptide from Klebsiella pneumoniae, was purified in a post-translationally modified form, MccE492m (8717 Da), from culture supernatants of either the recombinant Escherichia coli VCS257 strain harboring the pJAM229 plasmid or the K. pneumoniae RYC492 strain. Chymotrypsin digestion of MccE492m led to the MccE492m-(74–84) C-terminal fragment that carries the modification and that was analyzed by mass spectrometry and nuclear magnetic resonance at natural abundance. The 831-Da post-translational modification consists of a trimer of N-(2,3-dihydroxybenzoyl)-l-serine linked via a C-glycosidic linkage to a β-d-glucose moiety, itself linked to the MccE492m Ser-84-carboxyl through an O-glycosidic bond. This modification, which mimics a catechol-type siderophore, was shown to bind ferric ions by analysis of the collision-induced dissociation pattern obtained for MccE492m-(74–84) by electrospray ion trap mass spectrometry experiments in the presence of FeCl3. By using a series of wild-type and mutant isogenic strains, the three catechol-type siderophore receptors Fiu, Cir, and FepA were shown to be responsible for the recognition of MccE492m at the outer membrane of sensitive bacteria. Because MccE492m shows a broader spectrum of antibacterial activity and is more potent than MccE492, we propose that by increasing the microcin/receptor affinity, the modification leads to a better recognition and subsequently to a higher antimicrobial activity of the microcin. Therefore, MccE492m is the first member of a new class of antimicrobial peptides carrying a siderophore-like post-translational modification and showing potent activity, which we term siderophore-peptides. Microcin E492 (MccE492, 7886 Da), the 84-amino acid antimicrobial peptide from Klebsiella pneumoniae, was purified in a post-translationally modified form, MccE492m (8717 Da), from culture supernatants of either the recombinant Escherichia coli VCS257 strain harboring the pJAM229 plasmid or the K. pneumoniae RYC492 strain. Chymotrypsin digestion of MccE492m led to the MccE492m-(74–84) C-terminal fragment that carries the modification and that was analyzed by mass spectrometry and nuclear magnetic resonance at natural abundance. The 831-Da post-translational modification consists of a trimer of N-(2,3-dihydroxybenzoyl)-l-serine linked via a C-glycosidic linkage to a β-d-glucose moiety, itself linked to the MccE492m Ser-84-carboxyl through an O-glycosidic bond. This modification, which mimics a catechol-type siderophore, was shown to bind ferric ions by analysis of the collision-induced dissociation pattern obtained for MccE492m-(74–84) by electrospray ion trap mass spectrometry experiments in the presence of FeCl3. By using a series of wild-type and mutant isogenic strains, the three catechol-type siderophore receptors Fiu, Cir, and FepA were shown to be responsible for the recognition of MccE492m at the outer membrane of sensitive bacteria. Because MccE492m shows a broader spectrum of antibacterial activity and is more potent than MccE492, we propose that by increasing the microcin/receptor affinity, the modification leads to a better recognition and subsequently to a higher antimicrobial activity of the microcin. Therefore, MccE492m is the first member of a new class of antimicrobial peptides carrying a siderophore-like post-translational modification and showing potent activity, which we term siderophore-peptides. Microcins are low molecular weight antibacterial peptides secreted by enterobacteria, mostly Escherichia coli. As they are active against phylogenetically related Gram-negative bacteria, it has been suggested that they balance the intestinal microbial flora (1Baquero F. Moreno F. FEMS Microbiol. Lett. 1984; 23: 117-124Crossref Scopus (160) Google Scholar). Although many antimicrobial peptides of microbial origin are produced by large multidomain enzyme complexes, the peptide synthetases, microcins are synthesized through the ribosomal pathway. The microcin genetic systems include a series of genes encoding the microcin precursor, modification enzymes, secretion factors, and immunity proteins. These gene clusters give rise to a broad diversity of microcin structures and mechanisms of action (2Pons A.M. Lanneluc I. Cottenceau G. Sable S. Biochimie (Paris). 2002; 84: 531-537Crossref PubMed Scopus (53) Google Scholar, 3Destoumieux-Garzón D. Peduzzi J. Rebuffat S. Biochimie (Paris). 2002; 84: 511-519Crossref PubMed Scopus (46) Google Scholar). Microcin E492 (MccE492) was found for the first time in the culture medium of the Klebsiella pneumoniae RYC492 fecal strain (4de Lorenzo V. Arch. Microbiol. 1984; 139: 72-75Crossref PubMed Scopus (76) Google Scholar). It was shown to be chromosomally encoded, and the genetic determinants necessary for its production, secretion, and immunity were cloned into the pJAM229 plasmid (5Wilkens M. Villanueva J.E. Cofre J. Chnaiderman J. Lagos R. J. Bacteriol. 1997; 179: 4789-4794Crossref PubMed Google Scholar, 6Lagos R. Villanueva J.E. Monasterio O. J. Bacteriol. 1999; 181: 212-217Crossref PubMed Google Scholar). MccE492 was expressed in the recombinant E. coli VCS257 strain harboring the pJAM229 plasmid (5Wilkens M. Villanueva J.E. Cofre J. Chnaiderman J. Lagos R. J. Bacteriol. 1997; 179: 4789-4794Crossref PubMed Google Scholar) and characterized as an 84-residue peptide (GETDPNTQLLNDLGNNMAWGAALGAPGGLGSAALGAAGGALQTVGQGLIDHGPVNVPIPVLIGPSWNGSSSGYNSATSSSGSGS) of 7886 Da (7Pons A.M. Zorn N. Vignon D. Delalande F. Van Dorsselaer A. Cottenceau G. Antimicrob. Agents Chemother. 2002; 46: 229-230Crossref PubMed Scopus (29) Google Scholar). MccE492 is a membrane-active antibacterial peptide (8Lagos R. Wilkens M. Vergara C. Cecchi X. Monasterio O. FEBS Lett. 1993; 321: 145-148Crossref PubMed Scopus (61) Google Scholar, 9Destoumieux-Garzón D. Thomas X. Santamaria M. Goulard C. Barthelemy M. Boscher B. Bessin Y. Molle G. Pons A.M. Letellier L. Peduzzi J. Rebuffat S. Mol. Microbiol. 2003; 49: 1031-1041Crossref PubMed Scopus (66) Google Scholar), which depolarizes the E. coli cytoplasmic membrane (9Destoumieux-Garzón D. Thomas X. Santamaria M. Goulard C. Barthelemy M. Boscher B. Bessin Y. Molle G. Pons A.M. Letellier L. Peduzzi J. Rebuffat S. Mol. Microbiol. 2003; 49: 1031-1041Crossref PubMed Scopus (66) Google Scholar, 10de Lorenzo V. Pugsley A.P. Antimicrob. Agents Chemother. 1985; 27: 666-669Crossref PubMed Scopus (66) Google Scholar). However, this membrane activity alone is not believed to account for the peptide antibacterial properties, which are thought instead to rely on a receptor-mediated mechanism (9Destoumieux-Garzón D. Thomas X. Santamaria M. Goulard C. Barthelemy M. Boscher B. Bessin Y. Molle G. Pons A.M. Letellier L. Peduzzi J. Rebuffat S. Mol. Microbiol. 2003; 49: 1031-1041Crossref PubMed Scopus (66) Google Scholar, 11Corsini G. Baeza M. Monasterio O. Lagos R. Biochimie (Paris). 2002; 84: 539-544Crossref PubMed Scopus (24) Google Scholar). In agreement with this hypothesis, enterobactin (12Pollack J.R. Neilands J.B. Biochem. Biophys. Res. Commun. 1970; 38: 989-992Crossref PubMed Scopus (302) Google Scholar), also named enterochelin (13O'Brien I.G. Gibson F. Biochim. Biophys. Acta. 1970; 215: 393-402Crossref PubMed Scopus (279) Google Scholar), was shown to be an MccE492 antagonist (14Orellana C. Lagos R. FEMS Microbiol. Lett. 1996; 136: 297-303Crossref PubMed Google Scholar). Enterobactin is an iron-chelating compound (siderophore) produced by enterobacteria (15Payne S.M. Crit. Rev. Microbiol. 1988; 16: 81-111Crossref PubMed Scopus (148) Google Scholar) and also by some Streptomyces species (16Fiedler H.P. Krastel P. Muller J. Gebhardt K. Zeeck A. FEMS Microbiol. Lett. 2001; 196: 147-151Crossref PubMed Google Scholar). This tricatechol derivative of a cyclic triserine lactone (cyclic trimer of N-(2,3-dihydroxybenzoyl)-l-serine, DHBS)1 as well as its breakdown products, the linear DHBS trimer, dimer, and monomer, are able to transport ferric ions into enterobacteria (12Pollack J.R. Neilands J.B. Biochem. Biophys. Res. Commun. 1970; 38: 989-992Crossref PubMed Scopus (302) Google Scholar, 17O'Brien I.G. Cox G.B. Gibson F. Biochim. Biophys. Acta. 1971; 237: 537-549Crossref PubMed Scopus (101) Google Scholar, 18Scarrow R. Ecker D.J. Ng C. Liu S. Raymond K.N. Inorg. Chem. 1991; 30: 900-906Crossref Scopus (60) Google Scholar, 19Hantke K. FEMS Microbiol. Lett. 1990; 55: 5-8PubMed Google Scholar). The catecholate iron-siderophore receptors involved, Fiu, Cir, and FepA, have been suggested to recognize MccE492 (9Destoumieux-Garzón D. Thomas X. Santamaria M. Goulard C. Barthelemy M. Boscher B. Bessin Y. Molle G. Pons A.M. Letellier L. Peduzzi J. Rebuffat S. Mol. Microbiol. 2003; 49: 1031-1041Crossref PubMed Scopus (66) Google Scholar, 11Corsini G. Baeza M. Monasterio O. Lagos R. Biochimie (Paris). 2002; 84: 539-544Crossref PubMed Scopus (24) Google Scholar, 20Braun V. Patzer S.I. Hantke K. Biochimie (Paris). 2002; 84: 365-380Crossref PubMed Scopus (144) Google Scholar). In addition, this recognition was found to be TonB-dependent (9Destoumieux-Garzón D. Thomas X. Santamaria M. Goulard C. Barthelemy M. Boscher B. Bessin Y. Molle G. Pons A.M. Letellier L. Peduzzi J. Rebuffat S. Mol. Microbiol. 2003; 49: 1031-1041Crossref PubMed Scopus (66) Google Scholar). From the first description of the MccE492 structure (7Pons A.M. Zorn N. Vignon D. Delalande F. Van Dorsselaer A. Cottenceau G. Antimicrob. Agents Chemother. 2002; 46: 229-230Crossref PubMed Scopus (29) Google Scholar), it was thought that together with ColV, MccH47, and MccL, MccE492 belonged to a class of unmodified microcins (2Pons A.M. Lanneluc I. Cottenceau G. Sable S. Biochimie (Paris). 2002; 84: 531-537Crossref PubMed Scopus (53) Google Scholar). However, the lack of post-translational modification in the MccE492 structure (7Pons A.M. Zorn N. Vignon D. Delalande F. Van Dorsselaer A. Cottenceau G. Antimicrob. Agents Chemother. 2002; 46: 229-230Crossref PubMed Scopus (29) Google Scholar) was questionable because of the presence within the large genetic system of MccE492 of several genes, three of which displayed homologies with genes encoding glycosyltransferases (mceC), acyltranferases (mceI), and enterobactin esterases (mceD), whereas no homology was found for mceJ and mceE (21Lagos R. Baeza M. Corsini G. Hetz C. Strahsburger E. Castillo J.A. Vergara C. Monasterio O. Mol. Microbiol. 2001; 42: 229-243Crossref PubMed Scopus (66) Google Scholar). In recent work, inactivation of the mceC, mceI, or mceJ genes was shown to result in loss of MccE492 antibacterial activity. Because such inactivation did not result in any modification of the primary structure, the genes concerned were proposed to encode molecular chaperones involved in the acquisition of the MccE492 active conformation rather than modification enzymes (11Corsini G. Baeza M. Monasterio O. Lagos R. Biochimie (Paris). 2002; 84: 539-544Crossref PubMed Scopus (24) Google Scholar). In this study, we report for the first time the isolation and characterization of a modified form of MccE492. The structure of this post-translational modification was determined by mass spectrometry and nuclear magnetic resonance. It consists of a sugar moiety carrying a linear trimer of DHBS that is shown to bind a ferric ion. The antibacterial activity of MccE492m was examined against wild-type and mutant strains, and the membrane receptors involved in the microcin recognition at the outer membrane of the sensitive strains were determined. We show here the advantage in terms of activity of the modification carried by MccE492, which thus appears as the first member of a new class of post-translationally modified antibacterial peptides that we name siderophore-peptides. Bacterial Strains and Plasmids—Microcin production was performed in E. coli VCS257 (Stratagene), E. coli C600, E. coli C600 aroB– harboring the pJAM229 plasmid (5Wilkens M. Villanueva J.E. Cofre J. Chnaiderman J. Lagos R. J. Bacteriol. 1997; 179: 4789-4794Crossref PubMed Google Scholar), and K. pneumoniae RYC492 (4de Lorenzo V. Arch. Microbiol. 1984; 139: 72-75Crossref PubMed Scopus (76) Google Scholar). The bacteria used to establish the spectrum of antibacterial activity were those described previously (9Destoumieux-Garzón D. Thomas X. Santamaria M. Goulard C. Barthelemy M. Boscher B. Bessin Y. Molle G. Pons A.M. Letellier L. Peduzzi J. Rebuffat S. Mol. Microbiol. 2003; 49: 1031-1041Crossref PubMed Scopus (66) Google Scholar). Micrococcus luteus CIP5345 was from the Pasteur Institute collection (Paris, France). Studies of the MccE492m mechanism of action included a series of wild-type and mutant E. coli strains listed in Table I. Complementation of E. coli W3110-KP1344 (tonB–) was performed with the pMS7 plasmid (KanR, encoding TonB) (9Destoumieux-Garzón D. Thomas X. Santamaria M. Goulard C. Barthelemy M. Boscher B. Bessin Y. Molle G. Pons A.M. Letellier L. Peduzzi J. Rebuffat S. Mol. Microbiol. 2003; 49: 1031-1041Crossref PubMed Scopus (66) Google Scholar). Similarly, complementation of E. coli C600 was performed with the pHX405 plasmid (AmpR, encoding FhuA) (22Moeck G.S. Letellier L. J. Bacteriol. 2001; 183: 2755-2764Crossref PubMed Scopus (56) Google Scholar).Table IWild type and mutant E. coli strains used in the study of the MccE492m mechanism of actionStainsGenotypeSource or Ref.W3110F- IN(rrnD-rrnE)142Hill C.W. Harnish B.W. Proc. Natl. Acad. Sci. U. S. A. 1981; 78: 7069-7072Crossref PubMed Scopus (175) Google ScholarKP1344W3110 (tonB::blaM)43Larsen R.A. Thomas M.G. Postle K. Mol. Microbiol. 1999; 31: 1809-1824Crossref PubMed Scopus (155) Google ScholarW3110-6W3110 Δ(exbB-exbD)aS. P. Howard, unpublished data.C600F-thr leu fhuA lacY thi supELaboratory collectionMC4100AraD Δ (argF-lac) U169 rpsL relA flbB deoCLaboratory collectionH1443MC4100 aroB19Hantke K. FEMS Microbiol. Lett. 1990; 55: 5-8PubMed Google ScholarH873H1443 fepA::Tn10bK. Hantke, unpublished data.H1594H1443 fiu::Mud XbK. Hantke, unpublished data.H2222H1443 cir::Mud XbK. Hantke, unpublished data.H1728H1443 cir fiu::Mud X19Hantke K. FEMS Microbiol. Lett. 1990; 55: 5-8PubMed Google ScholarH1875H1443 cir::Mud X fepA::Tn1019Hantke K. FEMS Microbiol. Lett. 1990; 55: 5-8PubMed Google ScholarH1877H1443 fepA::Tn10 fiu::Mud X19Hantke K. FEMS Microbiol. Lett. 1990; 55: 5-8PubMed Google ScholarH1876H1728 fepA::Tn1019Hantke K. FEMS Microbiol. Lett. 1990; 55: 5-8PubMed Google Scholara S. P. Howard, unpublished data.b K. Hantke, unpublished data. Open table in a new tab Large Scale Purification of MccE492 and MccE492m—The microcin-producing strains E. coli VCS257 and E. coli C600 harboring the pJAM229 plasmid as well as K. pneumoniae RYC492 were grown for 16 h at 37 °C with vigorous shaking in 2 liters of M63 minimum medium containing 0.25 g/liter MgSO4 and 1 mg/liter thiamine. Alternatively, glucose or trisodium citrate (0.25%) was used as a source of carbon, and casamino acids (Difco) or bactotryptone (BD Biosciences) was used at 1 g/liter as a source of amino acids. After cell removal by centrifugation (5000 × g, 15 min, 4 °C), the supernatant was loaded onto a Sep-Pak C8 cartridge (Waters) previously equilibrated with 0.1% aqueous trifluoroacetic acid (pH 2). The cartridge was washed once with 0.1% aqueous trifluoroacetic acid, and three stepwise elutions were performed with 30, 35, and 40% ACN in 0.1% aqueous trifluoroacetic acid. The absorbance was monitored at 226 nm, and the fractionation was performed at a flow rate of 10 ml/min. The 40% ACN Sep-Pak fraction was concentrated under vacuum (Rotavapor, Büchi Labortechnik) before being lyophilized. The lyophilized fraction was resuspended in 40% ACN in 0.1% aqueous trifluoroacetic acid and subjected to RP-HPLC on an analytical Inertsil ODS2 column (5 μm, 250 × 4.6 mm, Interchim, France). Isocratic elution with 39% ACN in 0.1% aqueous trifluoroacetic acid at a flow rate of 1 ml/min allowed successive elution of MccE492 and MccE492m. Absorbance was monitored at 226 nm. Chymotrypsin Digestion and Isolation of the MccE492m-(74–84) Fragment—MccE492 or MccE492m (150 nmol) dissolved in 25 mm NH4HCO3 (pH 7.8) was digested with chymotrypsin (Sigma) at an enzyme to substrate ratio of 1:50 (w/w) for 90 min at 25 °C. The reaction was stopped by acidification with glacial acetic acid, and samples were dried under vacuum (SpeedVac, Savant). The proteolytic fragments were separated on an Inertsil ODS2 column (5 μm, 250 × 4.6 mm), with a linear gradient of 0–60% ACN in 0.1% aqueous trifluoroacetic acid over 90 min at a flow rate of 1 ml/min. Absorbance was monitored at 226 nm. SDS-PAGE—Purified MccE492 and MccE492m were loaded onto a 16.5% SDS-Tricine polyacrylamide gel (23Schägger H. von Jagow G. Anal. Biochem. 1987; 166: 368-379Crossref PubMed Scopus (10505) Google Scholar) without boiling and detected by silver staining. The molecular weight marker was the ultra-low range color marker (Sigma). Amino Acid Composition and Microsequence Analysis—Peptides (0.5 nmol) were hydrolyzed under vacuum for 24 h at 110 °Cin6 m constant boiling HCl (Sigma). Separation by ion-exchange HPLC of the resulting amino acids and post-column derivatization with ninhydrin were performed as described previously (9Destoumieux-Garzón D. Thomas X. Santamaria M. Goulard C. Barthelemy M. Boscher B. Bessin Y. Molle G. Pons A.M. Letellier L. Peduzzi J. Rebuffat S. Mol. Microbiol. 2003; 49: 1031-1041Crossref PubMed Scopus (66) Google Scholar). Lyophilized purified samples (10 nmol) of MccE492 and MccE492m and of selected chymotrypsin digest fragments of MccE492 and MccE492m were resuspended in MilliQ™ water (Millipore) or methanol. Each sample was loaded and argon-dried on a Biobrene-coated filter and then subjected to N-terminal sequencing according to Edman degradation on a Procise 492 automatic protein sequencer (Applied Biosystems, PerkinElmer Life Sciences). Liquid Growth Inhibition Assay—Antibacterial assays were performed as described previously (9Destoumieux-Garzón D. Thomas X. Santamaria M. Goulard C. Barthelemy M. Boscher B. Bessin Y. Molle G. Pons A.M. Letellier L. Peduzzi J. Rebuffat S. Mol. Microbiol. 2003; 49: 1031-1041Crossref PubMed Scopus (66) Google Scholar). Briefly, serial dilutions of MccE492 or MccE492m (10 μl) were incubated in a 96-well microtiter plate with a 90-μl suspension of bacteria in midlogarithmic growth phase diluted in poor-broth nutrient medium (1% bactotryptone (BD Biosciences), 0.5% NaCl) to a starting absorbance of 0.001 at 620 nm. Inhibition of growth was determined by measuring the absorbance at 620 nm with a Ceres 900 microplate recorder (Bioteck Instruments) after a 16-h incubation at 30 °C. Minimal inhibitory concentrations (MICs) were defined as the lowest concentrations that cause 100% growth inhibition and were determined in triplicate. Mass Spectrometry—The purified MccE492, MccE492m, and the chymotryptic peptides were analyzed for molecular mass determination by MALDI-TOF MS with a Voyager-de-Pro spectrometer (Applied Biosystems) operating in the linear or reflectron mode with positive ion detection. Samples were loaded on matrix crystals obtained by rapid evaporation of a saturated solution of α-cyano-4-hydroxycinnamic acid in 50% ACN. The dry droplet method was chosen for target preparation. Bovine insulin (MH+ at m/z 5734.59), thioredoxin (MH+ at m/z 11674.48), and apomyoglobin (MH+ at m/z 16952.56) were used for calibration. Sequencing of the purified MccE492-(74–84) and MccE492m-(74–84) and analysis of the structure of the modification were performed by using hybrid ESI-Qq-TOF (Q-Star, Applied Biosystems) and nano-ESI-IT (ESQUIRE 3000, Bruker Daltonics) instruments operating in the positive and negative ion detection modes. Peptide samples (10 pmol/μl) were prepared in 50% ACN. Formic acid (0.5%) was added for positive ion mode analysis. In Fe3+/peptide binding experiments, a solution of FeCl3 was added to obtain metal to peptide ratios varying from 1:1 to 8:1. CID experiments (MS2 and MS3) were performed from selected ions submitted to resonant excitation amplitude from 0.5 to 1.5 VP-P. 2 μl of sample were loaded into Proxeon (Odense, Denmark) nano-electrospray tips. Accurate mass measurement was performed on the ESI-Qq-TOF instrument. Internal calibration was applied from the theoretical m/z ratio of identified b ions. The Roepstorff nomenclature (24Roepstorff P. Fohlman J. Biomed. Mass Spectrom. 1984; 11: 601Crossref PubMed Scopus (2389) Google Scholar) was used to describe peptide fragmentations. NMR Spectroscopy—MccE492m-(74–84) was dissolved at a final 2.5 mm concentration either in CD3OH/H2O (50:50, v/v) or CD3OD/D2O (50:50, v/v). One- and two-dimensional NMR spectra were recorded on Bruker Avance 400 and DRX 800 spectrometers (Wissembourg, France) using a BBI probe and a TXI cryoprobe, respectively, both equipped with shielded gradients z. 1H and 13C chemical shifts were measured using the methanol methyl resonance as internal reference taken at 3.313 and 49.3 ppm, respectively. Conventional two-dimensional experiments 1H-1H double quantum filtered correlation spectroscopy, TOCSY (an 80-ms z-filtered DIPSI-2 sequence was used for Hartmann-Hahn mixing), incredible natural abundance double quantum transfer, NOESY (200 ms mixing time), as well as natural abundance 1H-13C HSQC (using adiabatic decoupling during acquisition) and HMBC (containing a low-pass filter) were performed for each sample. In both spectrometers, solvent suppression was obtained either by presaturation or by means of pulsed field gradients used in a water suppression by gradient tailored excitation scheme (25Piotto M. Saudek V. Sklenar V. J. Biomol. NMR. 1992; 2: 661-665Crossref PubMed Scopus (3539) Google Scholar) or for coherence selection. Data were processed with the XWINNMR 3.1 software. Typically, matrixes were zero filled and forward linear prediction at the indirect dimension was applied prior to square sine bell apodization (shifted by π/3) and Fourier transformation. E. coli VCS257 harboring pJAM229 was grown in M63 minimal medium containing either glucose or citrate as a source of carbon and casamino acids or bactotryptone as a source of amino acids. No major difference in terms of antibacterial activity was observed upon carbon source replacement (data not shown), glucose thus being maintained for subsequent cultures. Conversely, significantly higher activities were detected in the supernatants of cultures using bactotryptone instead of the conventionally used casamino acids-supplemented medium (data not shown). This correlated with a much faster growth of the MccE492-producing strain in the bactotryptone-containing medium, as compared with the casamino acids strain (Fig. 1A). Therefore, when the cultures were stopped after a 16-h incubation at 37 °C, the bacteria were at the end and beginning of the exponential phase of growth, respectively. MccE492 was purified by RP-HPLC from both supernatants as described under “Experimental Procedures.” The purified fractions were both active against E. coli ML35p. Surprisingly, although the microcin purified from the casamino acid-containing medium showed the expected electrophoretic behavior (9Destoumieux-Garzón D. Thomas X. Santamaria M. Goulard C. Barthelemy M. Boscher B. Bessin Y. Molle G. Pons A.M. Letellier L. Peduzzi J. Rebuffat S. Mol. Microbiol. 2003; 49: 1031-1041Crossref PubMed Scopus (66) Google Scholar) on a silver-stained SDS-polyacrylamide gel (Fig. 1B), the peptide issued from the bactotryptone-containing medium exhibited a higher apparent molecular mass (Fig. 1B). MALDI-TOF MS measurements showed that this peptide exhibited a molecular mass 831 Da higher (MH+ at m/z 8718) than the expected value (MH+ at m/z 7887) (Fig. 1C). Both peptides displayed an identical 37-residue N-terminal amino acid sequence by automatic Edman sequencing and a similar amino acid composition, with the exception of three extra serine residues in the higher molecular weight peptide (Table II). These results indicated the unknown peptide was a modified MccE492 bearing an 831-Da post-translational modification. It was consequently named MccE492m. Both the unmodified and modified peptides were obtained in high yield, using their optimized culture medium (2.1 mg/liter MccE492 and 9.5 mg/liter MccE492m for M63-casamino acids and M63-bactotryptone media, respectively). MccE492m was also purified from the culture supernatant of the wild-type K. pneumoniae RYC492 and was characterized by MALDI-TOF MS.Table IIAmino acid composition of MccE492, MccE492m, and their chymotryptic fragments MccE492-(74–84) and MccE492m-(74–84)Amino acidMccE492MccE492-(74-84)MccE492mMccE492m-(74-84)Asx10.01.110.11.1ThraValues corrected by extrapolation to zero time hydrolysis.3.91.03.91.0SeraValues corrected by extrapolation to zero time hydrolysis.9.86.012.68.9Glx4.24.2Pro6.06.0Gly19.02.119.12.1Ala9.81.09.91.1Val4.14.0Met0.90.9Ile3.03.0Leu9.19.0Tyr1.01.0His1.11.1TrpNDbND, not determined.NDa Values corrected by extrapolation to zero time hydrolysis.b ND, not determined. Open table in a new tab The antibacterial activity of MccE492m was compared with that of the unmodified microcin against Gram-negative and Gram-positive bacteria. Like MccE492, MccE492m was highly active against all the E. coli and the Salmonella enteritidis strains tested, with MIC values in the 40–80 nm range and at 150 nm, respectively (Table III). Moreover, MccE492m was 4–8 times more active than MccE492 against these strains. In addition, MccE492m was active against Enterobacter cloacae and K. pneumoniae, whereas no activity could be detected for the unmodified microcin in the range of 0.02–10 μm (Table III). As observed previously for MccE492 (9Destoumieux-Garzón D. Thomas X. Santamaria M. Goulard C. Barthelemy M. Boscher B. Bessin Y. Molle G. Pons A.M. Letellier L. Peduzzi J. Rebuffat S. Mol. Microbiol. 2003; 49: 1031-1041Crossref PubMed Scopus (66) Google Scholar), MccE492m was inactive in the same concentration range against Gram-positive bacteria and Gram-negative bacteria not belonging to enterobacteria, such as Vibrio species. Therefore, MccE492m is a more potent antibacterial peptide than MccE492, characterized by a broader spectrum of activity against enterobacteria and by lower MICs.Table IIIComparative antibacterial activity of MccE492 and MccE492mMicroorganismsMICaMeasured in triplicate.MccE492MccE492mμmμmGram-negative bacteriaE. coli B0.300.04E. coli F0.300.08E. coli ML35p0.300.08S. enteritidis1.250.15E. cloacaeNAbNA, not active in the range of 0.02-10 μm.0.60K. pneumoniaeNA2.50Vibrio alginolyticusNANAVibrio harveyiNANAVibrio penaeicidaNANAGram-positive bacteriaM. luteus CIP5345NANAStaphylococcus aureusNANAStaphylococcus haemolyticusNANABacillus megateriumNANAa Measured in triplicate.b NA, not active in the range of 0.02-10 μm. Open table in a new tab In order to localize the modification in the MccE492 sequence, both MccE492 and MccE492m samples were subjected to chymotrypsin digestion. The RP-HPLCs of the resulting digests were identical (data not shown), with the exception of (i) one fast eluting peak present in the MccE492 digest exclusively, and (ii) a group of three slower eluting peaks observed in the MccE492m digest, which was absent from the MccE492 chromatogram. The MALDI-TOF and ESI-Qq-TOF mass spectra in positive mode of the corresponding fractions showed similar [M + H]+ ions at m/z 1772.6 and [M + 2H]2+ ions at m/z 886.8 for the three MccE492m-specific peaks, whereas the MccE492-specific peak led to an [M + H]+ ion at m/z 941.4, which appeared shifted to lower m/z ratios consistent with a mass difference of 831.2 Da (data not shown). An identical amino acid sequence (NSATSSSGSGS), corresponding to the 74–84 C-terminal residues of MccE492, was determined by Edman degradation for all the MccE492- and MccE492m-specific peptides. Because the three peaks in the MccE492m digest showed identical MS and Edman sequencing results, they were assumed to be three isomers or conformers. Therefore, further analyses were performed with the most abundant peptide form. The calculated mass for the 11-residue peptide sequence perfectly matched the 940.4-Da molecular mass determined for the MccE492-specific peptide, which was termed MccE492-(74–84). Because the molecular mass of the peptide isolated from the MccE492m digest (1771.6 Da) was 831.2 Da higher than that of MccE492-(74–84), it was deduced that the whole post-translational modification was carried by the 74–84 C-terminal peptide, which was named MccE492m-(74–84). CID of both MccE492-(74–84) and MccE492m-(74–84) was performed by ESI-MS in the positive ion mode (Fig. 2), generating two independent series of acylium (b-type) and ammonium (y-type) ions. The b ion series of both peptides, as well as the y ion series of MccE492-(74–84), unambiguously described the NSATSS-SGSGS sequence. By contrast, the MccE492m-(74–84) y ion series was shifted to upper m/z ratios consistent with a mass increase of 831.2 Da as compared with the MccE492-(74–84) y ion series (Fig. 2A). The ion at m/z 937.2, which is consistent with a protonated serine with an additional m/z ratio of 831.2, was attributed to the MccE492m-(74–84) y1 ion. The y2 to y8 ion series showed systematically this same shift in m/z ratio. Altogether, these results unambiguously indicated that the entire modification of MccE492m was carried by the microcin Ser-84 C-terminal residue. Characterization of a DHBS Trimer—In addition to the b- and y-type ion series observed as positive product ions, the CID spectrum of MccE492m-(74–84) obtained with the Qq-TOF instrument showed an intense ion at m/z 224, indicative of DHBS (data not shown) (26Berner I. Greiner M. Metzger J.