We describe here the mechanism of inhibition of two new siderophore-13-lactam conjugates against Escherichia coil X580. One conjugate is a spermidine-based catechol siderophore-carbacephalosporin (JAM-2-263), and the other is an N5-acetyl-Ns-hydroxy-L-ornithine tripeptide hydroxamate siderophore-carbacephalosporin . In an agar diffusion test, both conjugates produced large inhibitory zones against strain X580. Resistant strains (i.e., JAMR and EKD") could be isolated after exposure of X580 to the conjugates JAM-2-263 and EKID-3-88, respectively. No cross-resistance was observed in these individual isolates. JAMR and EKDR were studied further to elucidate the mechanism of inhibition of each conjugated drug. The affinities of JAM-2-263 and EKD-3-88 for penicillin-binding proteins (PBPs) of isolated inner membranes were determined by a competition assay with mI-penicillin V. JAM-2-263 targeted primarily PBPs IA/B and 5/6, while EKD-3-88 targeted PBPs lA/B and 3. Strains X580, JAMR, and EKDR showed similar PBP afinities for the conjugates. However, marked changes were observed in the iron-regulated outer membrane proteins of resistant isolates grown on agar plates depleted of iron. EKDR lost the expression of FhuA (78 kDa) and its sensitivity to phages Ti and T5, whereas JAMR lost the expression of Cir (74 kDa) and its sensitivity to colicin Ia. These results revealed the requirement of FhuA and Cir for the inhibitory activities of EKD-3-88 and JAM-2-263, respectively. In an antibiotic diffusion assay, ferrichrome (1 FLM) strongly antagonized the activities of both conjugates against X580 and JAMR, including the residual activity of JAM-2-263 against JAMR. However, the susceptibility of strain EKDR lacking the ferrichrome receptor (FhuA-) to the two conjugates remained the same in the presence of ferrichrome. The antagonistic effect of ferrichrome on the activity of JAM-2-263 may also indicate a role for FhuA in the activity of this 13-lactam conjugate. A FhuACir-double mutant confirmed this hypothesis, since it showed a higher level of resistance to JAM-2-263. To reproduce iron-restricted in vivo growth conditions, we grew X580 and EKDR cells in diffuision chambers implanted in the peritoneal cavities of rats. Strain EKDR showed impaired growth in such a cultivation system.

Cell composition and surface properties of Escherichia coli were modified by using various growth media to investigate the role of yet uncharacterized components in ceftazidime susceptibility. An eightfold dilution of Luria broth was used as the basic growth medium and was supplemented with up to 4% phosphate, 5% glucose, or 12% L-glutamate. Decreases in cephaloridine and ceftazidime susceptibility, of two-and eightfold, respectively, were observed only in the glucose-enriched medium. The outer membrane permeability to ceftazidime and cephaloridine was evaluated by crypticity indices. Indices were unchanged under all growth conditions. Fluorometry of whole cells with 1-N-phenylnaphthylamine showed that glucose does not affect the interaction of this hydrophobic probe with the membranes but showed that elevated concentrations of phosphate or glutamate cause a marked increase in cell hydrophobicity, which, in turn, correlates with an increase in the susceptibility of E. coli to nalidixic acid. Growth in phosphate-or glutamate-enriched media caused an augmentation in major phospholipid species and may explain the increased hydrophobicity and susceptibility of E. coli to nalidixic acid. These data showed that E. coli susceptibility to ceftazidime is not influenced by cell surface hydrophobicity and suggested that the contribution of a nonspecific lipophilic diffusion route for entry of ceftazidime into cells is not likely to occur or is distinct from that of more hydrophobic molecules such as nalidixic acid. Finally, the penicillin-binding proteins of the E. coli cells were also investigated. Penicillin-binding protein 8 was only markedly labeled with 125I-penicillin V in inner membranes extracted from cells grown with glucose. Results of this study suggest that the unexpected change in penicillin-binding protein 8 observed in the presence of glucose may be responsible for the increase in MICs of cephaloridine and ceftazidime.In general, the susceptibility of gram-negative bacteria to ,B-lactam antibiotics results from the relative effect of the outer membrane permeability and the periplasmic ,-lactamase on the P-lactam periplasmic concentration. In turn, the periplasmic antibiotic concentration needed to affect the antibiotic inner membrane cell targets (the penicillin-binding proteins [PBPs]) depends on the affinities of these PBPs for the P-lactam (33).Ceftazidime is a very potent broad-spectrum cephalosporin against members of the family Enterobacteriaceae and Pseudomonas spp. (9, 25), although it has been reported to have an unusually slow diffusion rate through the outer membrane of gram-negative bacteria (34). Recently, Nikaido and Normark (21) constructed a mathematical model to predict 3-lactam MICs from the antibiotic penetration rate through the outer membrane and the P-lactamase hydrolysis rate in the periplasm. In the case of ceftazidime,'the observed MIC was significantly different from the predicted value, and it was suggested that ceftazidime penetrates the outer membrane by an additional non-porin p...

No abstract

This paper describes a two-cycle bacteria energy recovery system (BERS) to power two embedded sensors: an ultra-low portable pH sensor and a sound sensor. The designed unit can handle up to seven microbial fuel cells (MFCs) to charge a super-capacitor. This allows the BERS to provide a constant 0.14 mW without further electrical components for signal conditioning. The two cycles were driven with a 100 kΩ load and a 10 Hz frequency. The BERS is also self-powered with an integrated start-up unit to be self-activated when the MFCs charge the energy-storing unit after three days. The BERS powered pH sensor has an error below 5% at 25 ∘C and is able to work continuously while being activated for 4 h. The performances of the pH and sound sensors were determined based on a compromise between accuracy and power consumption.

No abstract