A bacteriocin produced by Lactobacillus plantarum LL441 was selected from the inhibitory products of 75 mesophilic lactobacilli because of its potency and broad spectrum. It is a peptide of 3.5 kDa whose amino-terminal sequence is NH2-K-K-T-K-K-N-X-S-G-D-I-. It is bactericidal and, in some cases, bacteriolytic. The peptide, called plantaricin C, retained its activity after boiling, storage, and treatment at different pHs.

Plasmid-borne resistance to fosfomycin in bacteria is due to modification of the antibiotic molecule by a glutathione S-transferase that catalyzes the formation of a covalent bond between the sulfhydryl residue of the cysteine in glutathione and the C-1 of fosfomycin. This reaction results in opening of the epoxide ring of the antibiotic to form an inactive adduct, the structure of which was confirmed by nuclear magnetic resonance. Dialyzed extracts prepared from resistant Escherichia coli strains were unable to modify fosfomycin unless exogenous glutathione was added to the reaction mixtures. Similarly, mutants defective in glutathione biosynthesis were susceptible to fosfomycin, despite harboring a resistance plasmid. Extracts of resistant but not susceptible strains could join glutathione to 1-chloro-2,4-dinitrobenzene, confirming the nature of the enzymatic activity. Adduct formation appeared to be specific for glutathione: none of the other thiols tested (cysteine, N-acetylcysteine, and dithiothreitol) could modify fosfomycin.

The enzyme that modifies fosfomycin by formation of an adduct with glutathione was purified 12-fold with a 56% activity yield by passage through DEAE Sephacel and high-performance liquid chromatography molecular exclusion columns. Its functional form was a homodimer of two 16,000-dalton polypeptides, which possibly showed an antiparallel a tertiary structure and which lacked marked hydrophobic regions. Visualization of the reaction was achieved by precolunm derivatization of glutathione and the adduct, separation by high-performance liquid chromatography, and fluorescence detection of both compounds. Temperature and pH optima were 20 to 30°C and 8.25, respectively; Mn2+, Fe2+, and Co2+ enhanced the rate of modification; and Km values were 9.4 and 11 mM for fosfomycin and glutathione, respectively. Phosphoenolpyruvate did not interfere with fosfomycin modification. The enzyme was stable at 40C for at least 6 months but progressively lost its activity upon being heated for 60 min at temperatures over 300C.Resistance to fosfornycin, an antibiotic that inhibits the synthesis of peptidoglycan by blocking the formation of N-acetylmuramic acid (11), is exerted by three mechanisms; two of them are located on the chromosome and the third is of plasmid origin. Chromosomal mutants either lack a functional uptake system for L-a-glycerophosphate or glucose-6-phosphate (10, 25) or have a phosphoenolpyruvate:UDP-N-acetylglucosamine enolpyruvyl transferase (pyruvil transferase), which discriminates between phosphoenolpyruvic acid and fosfomycin (26). Plasmid-encoded resistance to fosfomycin (17) is exerted by a gene that is located in a transposon (8,19) and that encodes a 16-kilodalton protein that is located in the cytoplasm; it also encodes constitutive synthesis (7,27). The gene is present in all strains of both environmental and clinical origin that have been isolated in our region of Spain. The presence of this gene was determined on the basis of conjugative transmissibility of fosfomycin resistance (1, 27).The gene product catalyzes the formation of an adduct between fosfomycin and glutathione through the opening of the epoxide ring of the antibiotic and the bonding of its C-1 to the sulfydryl group of the tripeptide cysteine (2). This results in inactivation of the antibiotic. The enzyme is thus a glutathione S-transferase that resembles those found in eucaryotes because of its specificity for glutathione as a substrate and its detoxifying function (14, 16), but it seems to differ from those found in eucaryotes because eucaryotic enzymes do not support fosfomycin modification (2).Here we report the isolation and characterization of fosfomycin:glutathione S-transferase and give preliminary results on the existence of a second enzyme which is responsible for the reaction between glutathione and 1-chloro-2,4-dinitrobenzene, which was previously reported as being carried out by the fosfomycin resistance protein in extracts of fosfomycin-resistant cells (2). MATERIALS AND METHODSBacterial strains. Escherichia coli K-12 JM83 ...

Sixty out of 219 fosfomycin-resistant bacteria selected from more than 7400 urinary pathogens in an epidemiological multicentre survey performed in Italy were screened for plasmid genes fosA and fosB conferring fosfomycin resistance. Only five strains, three enterobacteria and two staphylococci, carried plasmids harbouring, respectively, fosA and fosB genes. Fosfomycin resistance in the other isolates was caused by an alteration of the chromosomally encoded GlpT transport system. One strain, Morganella morganii 279, incorporated alpha-glycerolphosphate and its mechanism of fosfomycin resistance needs to be further investigated. Our study showed that PCR amplification is the most accurate, simple and rapid method for epidemiological studies of plasmid-encoded fosfomycin resistance, and that fosfomycin resistance conferred by plasmid genes (both fosA and fosB) accounts for only a low percentage of the fosfomycin-resistant strains.

Pseudomonas syringae PB-5123, a producer of fosfomycin, is resistant to high concentrations of the antibiotic. Two possible mechanisms of resistance have been detected: (i) impermeability to exogenous fosfomycin, even in the presence of sugar phosphate uptake inducers, and (ii) antibiotic phosphorylation. The gene responsible for this last activity, fosC, encodes a ca. 19,000-Da protein and is immediately followed by a second open reading frame, which shows sequence similarities to glutathione S-transferases. FosC uses ATP as a cosubstrate in an inactivation reaction that can be reversed with alkaline phosphatase. Other nucleotide triphosphates cannot be substituted for ATP in this reaction. No relationship between fosC and the previously described genes of fosfomycin resistance was found.Fosfomycin [L-(cis)-1,2-epoxy propyl phosphonic acid] is a broad-spectrum antibiotic produced by some strains of Streptomyces (15) and by Pseudomonas syringae (25). It enters cells by active transport through the L-␣-glycerophosphate and the hexose-6-phosphate uptake systems and blocks peptidoglycan biosynthesis through inhibition of the formation of N-acetylmuramic acid (18).Chromosomally encoded fosfomycin-resistant (Fo r ) strains have an impairment in fosfomycin uptake (17) or a low-affinity pyruvyl transferase (31). Plasmid-borne Fo r in gram-negative bacteria is mediated by a 141-amino-acid polypeptide (FosA) which catalyzes the fusion of fosfomycin and glutathione (for a comprehensive review, see reference 26). Another plasmidencoded Fo r determinant (fosB) was found in Staphylococcus epidermidis (10). The proteins FosA and FosB exhibit 38.3% sequence identity (35).It has been suggested that the genes encoding antibiotic resistance isolated from nosocomial strains may have originated from antibiotic-producing organisms, because analogous enzymatic activities (2, 33) and protein sequence similarities (27) have been found in both producers and clinical isolates. On the other hand, antibiotic resistance and biosynthetic genes are usually clustered, and consequently, the molecular cloning of resistance genes may also facilitate a molecular characterization of the biosynthetic pathways. In this report we describe the characterization of a DNA fragment from P. syringae PB-5123; when this fragment is introduced into Escherichia coli it leads to Fo r . The Fo r determinant, fosC, encodes a 19-kDa polypeptide that probably inactivates fosfomycin through the introduction of a phosphate group in the molecule, with ATP as a donor. MATERIALS AND METHODSBacterial strains, plasmids, and phages. P. syringae PB-5123 was used throughout the study (25). E. coli HB101 (23) and E. coli JM83 (34) were used as recipients for recombinant plasmids. E. coli XL1-Blue (24) was transfected with M13mp18 or M13mp19 phage DNA (34) containing the DNA fragments to be sequenced. Minicell experiments were conducted with E. coli AR1062 (F Ϫ leu minA minB rpsL thi thr). E. coli ESS, a fosfomycin-hypersusceptible strain, was used as an indicator organism in ...