Rowth of the mutants in the presence of ampicillin at different levels (Fig. 4A and Fig. S2). Deletion of SO0541, SO0914, ampC, SO3054, SO3474 and SOA0149 resulted in a phenotype that was comparable to that of the isogenic parental strain. In contrast, loss of blaA substantially increased sensitivity to ampicillin, with no growth at 0.125 mg/ml. The DblaA strain failed to measurably grow when penicillin or carbenicillin at 1 mg/ml was added, whereas resistance of the other mutants to these two agents remained unaltered (Table 2). Expression of blaA in trans from the multiple-copy plasmid, pHG101, conferred the DblaA strain with resistance to ampicillin exceeding that of the wild type (Fig. 4A), presumably due to overproduction of BlaA [29]. In parallel, ectopic expression of blaA increased the MIC values of the mutant to ampicillin and prevented cell lysis (Table 2) (Fig. 4A). Moreover, similar results were obtained with the susceptibility test (Fig. 4B). These data indicate that the resistance to ampicillin can mainly beattributed to BlaA and that other putative PD1-PDL1 inhibitor 1 b-lactamases are not relevant under the conditions used.BlaA is induced by ampicillin at high concentrationsGiven that BlaA is largely responsible for the resistance of S. oneidensis to ampicillin, we hypothesized that this b-lactamase may be induced substantially by the addition of ampicillin at high, but not low, levels. To test this, we employed a lacZ-reporter system 
to assess the promoter activity of the blaA gene under various conditions [30]. As shown in Fig. 5A, expression of b-galactosidase driven by the blaA promoter in cultures supplemented with 50 mg/ ml ampicillin was almost 10 times that with 2.5 mg/ml ampicillin 2 hour after inoculation (,0.1 of OD600). Transcription declined with time, coinciding with reduction of the remaining ampicillin (Fig. 3D). In contrast, expression of lacZ in the presence of 2.5 mg/ ml ampicillin was constant and only slightly higher 
than that observed in cultures free of the antibiotic. Similar results were obtained with qRT-PCR when we examined expression of the blaA gene in samples treated with 50 mg/ml ampicillin (diamonds in Fig. 5A), confirming that blaA promoter is induced by ampicillin only at high concentrations. We then measured b-lactamase activity directly using the iodometric assay [31,32]. Penicillin instead of ampicillin was chosen as the substrate for the assay because of significant spontaneous hydrolysis of ampicillin [32]. As shown in Fig. 5B, when the order (-)-Calyculin A antibiotic was added at 50 mg/ml, penicillin hydrolysis recorded by reduction of the optical density (decolorization) became evident about 1.5 hours after inoculation and was much more rapid than that with penicillin at 2.5 mg/ml. In contrast, penicillin at 2.5 mg/ml was not removed until 2.5 h after inoculation. In both cases, the DblaA strain was unable to hydrolyze the antibiotic (data not shown) further confirming the critical role of BlaA. As the number of cells used in these assays was comparable, these data suggest that the amount of BlaA determines resistance to ampicillin/penicillin. Overall, we conclude that cell lysis induced by ampicillin at lysing concentrations is due to the delayed removal of the antibiotic, which resulted from an insufficient amount of BlaA.Table 1. Susceptibility of S. oneidensis to various antibiotics.Concentration (mg/ml) of antibioticsa Resistant Ampicillin Chloramphenicol Ciprofloxacin Erythromycin Gentamycin Kanamycin Neomycin Rifam.Rowth of the mutants in the presence of ampicillin at different levels (Fig. 4A and Fig. S2). Deletion of SO0541, SO0914, ampC, SO3054, SO3474 and SOA0149 resulted in a phenotype that was comparable to that of the isogenic parental strain. In contrast, loss of blaA substantially increased sensitivity to ampicillin, with no growth at 0.125 mg/ml. The DblaA strain failed to measurably grow when penicillin or carbenicillin at 1 mg/ml was added, whereas resistance of the other mutants to these two agents remained unaltered (Table 2). Expression of blaA in trans from the multiple-copy plasmid, pHG101, conferred the DblaA strain with resistance to ampicillin exceeding that of the wild type (Fig. 4A), presumably due to overproduction of BlaA [29]. In parallel, ectopic expression of blaA increased the MIC values of the mutant to ampicillin and prevented cell lysis (Table 2) (Fig. 4A). Moreover, similar results were obtained with the susceptibility test (Fig. 4B). These data indicate that the resistance to ampicillin can mainly beattributed to BlaA and that other putative b-lactamases are not relevant under the conditions used.BlaA is induced by ampicillin at high concentrationsGiven that BlaA is largely responsible for the resistance of S. oneidensis to ampicillin, we hypothesized that this b-lactamase may be induced substantially by the addition of ampicillin at high, but not low, levels. To test this, we employed a lacZ-reporter system to assess the promoter activity of the blaA gene under various conditions [30]. As shown in Fig. 5A, expression of b-galactosidase driven by the blaA promoter in cultures supplemented with 50 mg/ ml ampicillin was almost 10 times that with 2.5 mg/ml ampicillin 2 hour after inoculation (,0.1 of OD600). Transcription declined with time, coinciding with reduction of the remaining ampicillin (Fig. 3D). In contrast, expression of lacZ in the presence of 2.5 mg/ ml ampicillin was constant and only slightly higher than that observed in cultures free of the antibiotic. Similar results were obtained with qRT-PCR when we examined expression of the blaA gene in samples treated with 50 mg/ml ampicillin (diamonds in Fig. 5A), confirming that blaA promoter is induced by ampicillin only at high concentrations. We then measured b-lactamase activity directly using the iodometric assay [31,32]. Penicillin instead of ampicillin was chosen as the substrate for the assay because of significant spontaneous hydrolysis of ampicillin [32]. As shown in Fig. 5B, when the antibiotic was added at 50 mg/ml, penicillin hydrolysis recorded by reduction of the optical density (decolorization) became evident about 1.5 hours after inoculation and was much more rapid than that with penicillin at 2.5 mg/ml. In contrast, penicillin at 2.5 mg/ml was not removed until 2.5 h after inoculation. In both cases, the DblaA strain was unable to hydrolyze the antibiotic (data not shown) further confirming the critical role of BlaA. As the number of cells used in these assays was comparable, these data suggest that the amount of BlaA determines resistance to ampicillin/penicillin. Overall, we conclude that cell lysis induced by ampicillin at lysing concentrations is due to the delayed removal of the antibiotic, which resulted from an insufficient amount of BlaA.Table 1. Susceptibility of S. oneidensis to various antibiotics.Concentration (mg/ml) of antibioticsa Resistant Ampicillin Chloramphenicol Ciprofloxacin Erythromycin Gentamycin Kanamycin Neomycin Rifam.