The influence of pH and dissolved organic carbon on the ecotoxicity of ampicillin and clarithromycin.

The impacts of water chemistry properties including pH and dissolved organic carbon (DOC) on the ecotoxicity of active pharmaceutical ingredients (APIs) are increasingly evident. These impacts are a result of alterations in API bioavailability: pH regulates the bioavailability of many ionizable APIs via chemical speciation, whereas DOC interacts with several APIs to inhibit the APIs from traversing the membrane system of organisms. In this study, we examined the influences of pH and DOC on the bioavailability of ampicillin (AMP) and clarithromycin (CLA) with the help of a bioavailability model. The effects on bioavailability were quantified by ecotoxicity observed in cyanobacteria growth inhibition tests with Microcystis aeruginosa PCC7806. The median effect concentration (96 h-EC50total) of AMP increased by 5-fold when pH raised from 7.4 to 9.0, suggesting the zwitterionic AMP+/- species being higher in bioavailability than the negatively charged AMP- species. CLA ecotoxicity showed no significant pH-dependency, suggesting CLA+ and CLA0 species to be equally bioavailable, albeit it correlated significantly with M. aeruginosa growth rate in negative controls. In addition, DOC demonstrated no significant effects on the ecotoxicity of AMP or CLA. Overall, together with earlier results on ciprofloxacin, our data show that bioavailability relations with pH and DOC are variable among different antibiotics. Factors other than chemical speciation alone could play a role in their bioavailability, such as their molecular size and polarity.

Thermally activated persulfate oxidation of ampicillin: Kinetics, transformation products and ecotoxicity.

The heat-activated persulfate system showed encouraging results for the destruction of the widely used antibiotic Ampicillin (AMP). AMP removal follows exponential decay, and the observed kinetic constant was enhanced with persulfate (PS) dosage at the range 50-500 mg L-1 and temperature (40-60 °C), while AMP thermolysis at 60 °C was almost negligible. The apparent activation energy was estimated to 124.7 kJ mol-1. Alkaline conditions, water matrix constituents like bicarbonates, humic acid, and real water matrices retarded AMP oxidation. Experiments performed with tert-butanol and methanol as scavengers demonstrated the contribution of sulfate radicals as the dominant reactive species. Seven transformation products (TPs) of AMP have been identified from AMP destruction. An EC50 value equal to 187 mg L-1 was calculated for 72 h of exposure of the microalgae Chlorella sorokiniana to AMP. According to the ecotoxicity experiments that conducted after treatment of AMP with PS for different reaction times, no important inhibition of microalgae was noticed for contact time of 72 h and 10 d. These results indicate the formation of no toxic AMP by-products for the applied experimental conditions.

Probiotics protect against gut dysbiosis associated decline in learning and memory.

Gut-Brain-Axis imbalance due to gut dysbiosis by antibiotics may lead to neurobehavioral changes. Here we determine neuroprotective effect of probiotic against gut dysbiosis associated decline in learning and memory. Oral Ampicillin was used to induce gut dysbiosis while probiotic was administered along with antibiotic as treatment in Swiss albino mice. Antibiotic decreased Lactobacillus, Bifidobacterium, Firmicutes and Clostridium level. This was followed by reduced cognition, hippocampal neuronal density, intestinal crypt depth, villus length and increased corticohippocampal acetylcholinesterase, myeloperoxidase activity and oxidative stress which were partially reversed by probiotic treatment. Probiotic protected hippocampal neurons from gut dysbiosis associated inflammatory and oxidative damage in mice.

Sonocatalytic removal of ampicillin by Zn(OH)F: Effect of operating parameters, toxicological evaluation and by-products identification.

Zinc hydroxyfluoride (Zn(OH)F) sonocatalyst was prepared by using solvothermal method and was characterized by using various techniques. The sonocatalytic degradation of ampicillin (AMP) in water by sonolysis, bare ZnO and Zn(OH)F was investigated in terms of AMP removal, mineralization, detoxification of solution, and remaining by-products at the end of process. Results revealed that the sonocatalytic performance of Zn(OH)F was significantly greater than that of bare ZnO. Under the optimum conditions, the removal percentage of AMP by Zn(OH)F was ∼97% after 90 min reaction, while 51% and 36% COD and TOC removal were achieved after 120 min reaction, respectively. The study of Zn(OH)F stability revealed that the degradation efficiency of AMP was reduced by only 5% even after being reused for four experiments. The toxicity of initial and treated solutions was assessed by using agar-well diffusion method and ToxTrak™ toxicity assay, and the results indicated a substantial reduction in the toxicity of solution after the treatment. The formation of some by-products during the sonolysis and sonocatalysis was evaluated by LC-HR-MS/MS method. LC-HR-MS/MS results showed that the concentration of most by-products, which were produced after 90 min treatment by US/Zn(OH)F process, was considerably lower than those obtained during sonolysis and US/ZnO processes.

Combined effects of binary antibiotic mixture on growth, microcystin production, and extracellular release of Microcystis aeruginosa: application of response surface methodology.

The interactive effects of binary antibiotic mixtures of spiramycin (SP) and ampicillin (AMP) on Microcystis aeruginosa (MA) in terms of growth as well as microcystin production and extracellular release were investigated through the response surface methodology (RSM). SP with higher 50 and 5% effective concentrations in MA growth was more toxic to MA than AMP. RSM model for toxic unit approach suggested that the combined toxicity of SP and AMP varied from synergism to antagonism with SP/AMP mixture ratio decreasing from reversed equitoxic ratio (5:1) to equitoxic ratio (1:5). Deviations from the prediction of concentration addition (CA) and independent action (IA) model further indicated that combined toxicity of target antibiotics mixed in equivalent ratio (1:1) varied from synergism to antagonism with increasing total dose of SP and AMP. With the increase of SP/AMP mixture ratio, combined effect of mixed antibiotics on MA growth changed from stimulation to inhibition due to the variation of the combined toxicity and the increasing proportion of higher toxic component (SP) in the mixture. The mixture of target antibiotics at their environmentally relevant concentrations with increased total dose and SP/AMP mixture ratio stimulated intracellular microcystin synthesis and facilitated MA cell lysis, thus leading to the increase of microcystin productivity and extracellular release.

Use of a standardized JaCVAM in vivo rat comet assay protocol to assess the genotoxicity of three coded test compounds; ampicillin trihydrate, 1,2-dimethylhydrazine dihydrochloride, and N-nitrosodimethylamine.

As part of the Japanese Center for the Validation of Alternative Methods (JaCVAM)-initiative international validation study of the in vivo rat alkaline comet assay (comet assay), our laboratory examined ampicillin trihydrate (AMP), 1,2-dimethylhydrazine dihydrochloride (DMH), and N-nitrosodimethylamine (NDA) using a standard comet assay validation protocol (v14.2) developed by the JaCVAM validation management team (VMT). Coded samples were received by our laboratory along with basic MSDS information. Solubility analysis and range-finding experiments of the coded test compounds were conducted for dose selection. Animal dosing schedules, the comet assay processing and analysis, and statistical analysis were conducted in accordance with the standard protocol. Based upon our blinded evaluation, AMP was not found to exhibit evidence of genotoxicity in either the rat liver or stomach. However, both NDA and DMH were observed to cause a significant increase in % tail DNA in the rat liver at all dose levels tested. While acute hepatoxicity was observed for these compounds in the high dose group, in the investigators opinion there were a sufficient number of consistently damaged/measurable cells at the medium and low dose groups to judge these compounds as genotoxic. There was no evidence of genotoxicity from either NDA or DMH in the rat stomach. In conclusion, our laboratory observed increased DNA damage from two blinded test compounds in rat liver (later identified as genotoxic carcinogens), while no evidence of genotoxicity was observed for the third blinded test compound (later identified as a non-genotoxic, non-carcinogen). This data supports the use of a standardized protocol of the in vivo comet assay as a cost-effective alternative genotoxicity assay for regulatory testing purposes.

Core-shell structure microcapsules with dual pH-responsive drug release function.

We report dual pH-responsive microcapsules manufactured by combining electrostatic droplets (ESD) and microfluidic droplets (MFD) techniques to produce monodisperse core (alginate)-shell (chitosan) structure with dual pH-responsive drug release function. The fabricated core-shell microcapsules were size controllable by tuning the synthesis parameters of the ESD and MFD systems, and were responsive in both acidic and alkaline environment, We used two model drugs (ampicillin loaded in the chitosan shell and diclofenac loaded in the alginate core) for drug delivery study. The results show that core-shell structure microcapsules have better drug release efficiency than respective core or shell particles. A biocompatibility test showed that the core-shell structure microcapsules presented positive cell viability (above 80%) when evaluated by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. The results indicate that the synthesized core-shell microcapsules were a potential candidate of dual-drug carriers.

Construction and analysis of a human hepatotoxicity database suitable for QSAR modeling using post-market safety data.

Drug-induced liver injury (DILI) is one of the most common drug-induced adverse events (AEs) leading to life-threatening conditions such as acute liver failure. It has also been recognized as the single most common cause of safety-related post-market withdrawals or warnings. Efforts to develop new predictive methods to assess the likelihood of a drug being a hepatotoxicant have been challenging due to the complexity and idiosyncrasy of clinical manifestations of DILI. The FDA adverse event reporting system (AERS) contains post-market data that depict the morbidity of AEs. Here, we developed a scalable approach to construct a hepatotoxicity database using post-market data for the purpose of quantitative structure-activity relationship (QSAR) modeling. A set of 2029 unique and modelable drug entities with 13,555 drug-AE combinations was extracted from the AERS database using 37 hepatotoxicity-related query preferred terms (PTs). In order to determine the optimal classification scheme to partition positive from negative drugs, a manually-curated DILI calibration set composed of 105 negatives and 177 positives was developed based on the published literature. The final classification scheme combines hepatotoxicity-related PT data with supporting information that optimize the predictive performance across the calibration set. Data for other toxicological endpoints related to liver injury such as liver enzyme abnormalities, cholestasis, and bile duct disorders, were also extracted and classified. Collectively, these datasets can be used to generate a battery of QSAR models that assess a drug's potential to cause DILI.

Response of bacterial isolates from Antarctic shallow sediments towards heavy metals, antibiotics and polychlorinated biphenyls.

The response of bacterial isolates from Antarctic sediments to polychlorinated biphenyls (Aroclor 1242 mixture), heavy metal salts (cadmium, copper, mercury and zinc) and antibiotics (ampicillin, chloramphenicol, kanamycin and streptomycin) was investigated. Overall, the ability to growth in the presence of Aroclor 1242 as a sole carbon source was observed for 22 isolates that mainly belonged to Psychrobacter spp. Tolerance to the heavy metals assayed in this study was in the order of Cd > Cu > Zn > Hg and appeared to be strictly related to the metal concentrations, as determined during previous chemical surveys in the same area. With regards to antibiotic assays, the response of the isolates to the tested antibiotics ranged from complete resistance to total susceptibility. In particular, resistances to ampicillin and chloramphenicol were very pronounced in the majority of isolates. Our isolates differently responded to the presence of toxic compounds primarily based on their phylogenetic affiliation and secondarily at strain level. Moreover, the high incidence of resistance either to metal or antibiotics, in addition to the capability to grow on PCBs, confirm that bacteria are able to cope and/or adapt to the occurrence pollutants even in low human-impacted environments.

Horizontal gene transfer of stress resistance genes through plasmid transport.

The horizontal gene transfer of plasmid-determined stress tolerance was achieved under lab conditions. Bacterial isolates, Enterobacter cloacae (DGE50) and Escherichia coli (DGE57) were used throughout the study. Samples were collected from contaminated marine water and soil to isolate bacterial strains having tolerance against heavy metals and antimicrobial agents. We have demonstrated plasmid transfer, from Amp(+)Cu(+)Zn(-) strain (DGE50) to Amp(-)Cu(-)Zn(+) strain (DGE57), producing Amp(+)Cu(+)Zn(+) transconjugants (DGE(TC50→57)) and Amp(+)Cu(-)Zn(+) transformants (DGE(TF50→57)). DGE57 did not carry any plasmid, therefore, it can be speculated that zinc tolerance gene in DGE57 is located on chromosome. DGE50 was found to carry three plasmids, out of which two were transferred through conjugation into DGE57, and only one was transferred through transformation. Plasmid transferred through transformation was one out of the two transferred through conjugation. Through the results of transformation it was revealed that the genes of copper and ampicillin tolerance in DGE50 were located on separate plasmids, since only ampicillin tolerance genes were transferred through transformation as a result of one plasmid transfer. By showing transfer of plasmids under lab conditions and monitoring retention of respective phenotype via conjugation and transformation, it is very well demonstrated how multiple stress tolerant strains are generated in nature.

Adsorption and breakdown of penicillin antibiotic in the presence of titanium oxide nanoparticles in water.

The fate and transport of antibiotics in natural water systems is controlled in part by interactions with nanometer (10(-9)m) metal oxide particles. Experiments were performed by mixing solutions of ampicillin (AMP), a common, penicillin-class human and veterinary antibiotic, with 25 nm-TiO(2) (anatase) nanoparticles at different pH conditions. Both sorption and degradation of AMP were observed in the AMP-nanoparticle solutions. For AMP concentrations from ~3 µM to 2.9 mM the overall AMP removal from solution can be described by linear isotherms with removal coefficients (K(r)) of 3028 (±267)L kg(-1) at pH 2, 11,533 (±823)L kg(-1) at pH 4, 12,712 (±672)L kg(-1) at pH 6, and 1941 (±342)L kg(-1) at pH 8. Mass spectral analysis of AMP solutions after removal of the solid nanoparticles yielded ions that indicate the presence of peniclloic acid, penilloic acid and related de-ammoniated by-products as possible compounds resulting from the degradation of AMP at the TiO(2) surface.

Analyses of gene expression and physiological changes in Microcystis aeruginosa reveal the phytotoxicities of three environmental pollutants.

When the concentrations of ampicillin (Amp), atrazine (Atr) and cadmium chloride (Cd) reach excessive quantities, they become toxic to aquatic organisms. Due to the acceleration of the industrialization and the intensification of human activities, the incidence and concentrations of these types of pollutants in aquatic systems are increasing. The primary purpose of this study was to evaluate the short-term effects of Amp, Atr and Cd on the physiological indices and gene expression levels in Microcystis aeruginosa. These three pollutants significantly induced antioxidant activity but continuously accelerated the cellular oxidative damage in microalgae, which suggests an imbalance between the oxidant and the antioxidant systems. Amp, Atr and Cd also decreased the transcription of psaB, psbD1 and rbcL; the lowest transcription of these genes was only 38.1, 23.7 and 7% of the control, respectively. These three pollutants affected nitrogen (N) and phosphorous (P) uptake by inhibiting the transcription of N or P absorbing and transporting related genes, and they down regulated the transcription of microcystin-related genes, which caused a decrease of microcystin levels; and the lowest level of microcystin was only 42.4% of the control. Our results suggest that these pollutants may cause pleiotropic effects on algal growth and physiological and biochemical reactions, and they may even affect secondary metabolic processes.

Effect of a high dose of three antibiotics on the reproduction of a parthenogenetic strain of Folsomia candida (Isotomidae: Collembola).

Folsomia candida Willem (Isotomidae: Collembola) is an edaphic parthenogenetic species commonly used in ecotoxicity studies. We exposed F. candida to a high dose of three antibiotics, tylosin, ampicillin, and oxytetracycline, that target different bacterial groups. Possible toxic effects were assessed through egg production, hatching, and body size. All three antibiotics caused toxic effects. Treatment with oxytetracycline proved the most toxic. This group showed the smallest body size and lowest number of eggs laid, likely the result of a combination of antibiotic toxicity and avoidance of the antibiotic spiked food. Active toxin avoidance by F. candida in toxicological assays may play a role in minimizing their exposure to toxic compounds. Despite the administration of high doses of oxytetracycline, F. candida individuals remained infected with the intracellular bacteria Wolbachia indicating that this strain is resistant to this antibiotic or that the host or its gut flora had detoxified the compound. An increase in percent egg hatch with time was seen in the ampicillin and oxytetracycline treatments, indicating a possible accommodation of the host and/or gut-flora to these antibiotics.

From damaged genome to cell surface: transcriptome changes during bacterial cell death triggered by loss of a restriction-modification gene complex.

Genetically programmed cell deaths play important roles in unicellular prokaryotes. In postsegregational killing, loss of a gene complex from a cell leads to its descendants' deaths. With type II restriction-modification gene complexes, such death is triggered by restriction endonuclease's attacks on under-methylated chromosomes. Here, we examined how the Escherichia coli transcriptome changes after loss of PaeR7I gene complex. At earlier time points, activation of SOS genes and sigma(E)-regulon was noticeable. With time, more SOS genes, stress-response genes (including sigma(S)-regulon, osmotic-, oxidative- and periplasmic-stress genes), biofilm-related genes, and many hitherto uncharacterized genes were induced, and genes for energy metabolism, motility and outer membrane biogenesis were repressed. As expected from the activation of sigma(E)-regulon, the death was accompanied by cell lysis and release of cellular proteins. Expression of several sigma(E)-regulon genes indeed led to cell lysis. We hypothesize that some signal was transduced, among multiple genes involved, from the damaged genome to the cell surface and led to its disintegration. These results are discussed in comparison with other forms of programmed deaths in bacteria and eukaryotes.

The Rhizobium etli opt operon is required for symbiosis and stress resistance.

Rhizobium etli is a Gram-negative root-colonizing soil bacterium capable of fixing nitrogen while living in symbiosis with its leguminous host Phaseolus vulgaris. A genome-wide screening for R. etli symbiotic mutants revealed a R. etli operon encoding an oligopeptide ABC-transporter (Opt), two redA homologous genes and one redB gene. Expression analysis showed this opt operon to be transcribed both under free-living and symbiotic conditions and expression levels were demonstrated to be growth-phase-dependent. Plants nodulated by R. etli opt mutants showed a reduced symbiotic nitrogen fixation activity (approximately 50% reduction). Growth experiments with opt mutants in the presence of oligopeptides as the sole nitrogen source confirmed the involvement of the opt genes in oligopeptide uptake. Further phenotypic analysis of the opt mutants revealed them to display an enhanced resistance to the oligopeptide antibiotic bacitracin, an increased susceptibility to the beta-lactam antibiotic ampicillin and a decreased osmotolerance. In conclusion, our results demonstrate that the opt operon plays a crucial role during symbiosis and stress resistance.

Antibiotic amoxicillin induces DNA lesions in mammalian cells possibly via the reactive oxygen species.

Amoxicillin is a commonly prescribed drug for anti- bacterial infection. In this study, we are interested in the effect of the drug on the cellular DNA integrity. Amoxicillin was added to the human or hamster cells in culture, and the DNA lesions induced by the drug were assessed by a comet assay with nuclear extract incubation (Wang et al., 2005 Anal Biochem 337: 70-75). Amoxicillin at 5mM rapidly induced DNA lesions in human AGS cells. The level of DNA lesions attained a maximum at about 1h, and then declined steadily and reached almost the basal level at 6h following the drug treatment. Similar induction pattern of DNA lesions was found with amoxicillin-related antibiotics such as ampicillin but not with the unrelated antibiotics such as kanamycin. For studying the repair kinetics, the cells were treated with amoxicillin for only 1h and continued culture in the absence of the drug for a certain period of time before subsequent analysis. Repair of the amoxicillin-induced DNA lesions was essentially completed within 4h. Such repair may not involve nucleotide excision repair (NER) pathway because the repair was completed with similar kinetics in both NER proficient Chinese hamster CHO-K1 cells and its isogenic NER deficient UV24 cells. Instead, the repair may involve base excision repair (BER) pathway because immunodepletion of OGG1/2, glycosylases involved in BER rendered the nuclear extract unable to excise DNA lesions induced by amoxicillin in the modified comet assay. Furthermore, amoxicillin induced intracellular reactive oxygen species (ROS) at the tempo similar to that of DNA lesions induction. Thus, we hypothesize that amoxicillin causes oxidative DNA damage in mammalian cells via ROS.

The effects of organic pesticides on inner membrane permeability in Escherichia coli ML35.

We have tested whether some pesticides might cause inner membrane leakage in ML35 Escherichia coli cells, which express beta-galactosidase (lacZ; EC 3.2.1.23) constitutively but lack the permease (lacY) required for substrate entry. The activity of beta-galactosidase (indicative of substrate leakage through the inner membrane) was increased by various concentrations of pesticides, including the organometallic fungicides maneb and mancozeb, the insecticide Thiodan, and the herbicide Ally, as well as by antibiotics such as ampicillin, gramicidin D, and the calcium ionophore A23187. The enzyme activity was increased by up to approximately 30% when the E. coli ML35 strain was exposed to various concentrations (between 50 and 250 ppm) of both fungicides. Thiodan had only a slight effect on beta-galactosidase activity (increase of 12.8%), whereas, among the antibiotics, the calcium ionophore at 20 microg/ml caused a significant increase in enzyme activity by up to 61.8%. This effect is similar to that of sodium dodecyl sulfate, used as positive control ( approximately 70% increase). Accumulation of maneb and mancozeb by bacterial cells was also studied taking advantage of their metal content and using atomic absorption spectrophotometry. In parallel with the increase in enzyme activity, both fungicides accumulated in the cells as a function of their concentration. Time course experiments (3, 6, and 9 h) of fungicide accumulation and of bacterial growth at various pesticide concentrations were also carried out. Maneb seems to inhibit the bacterial growth better than mancozeb. In addition, maneb uptake increases with time up to 9 h at all tested concentrations, whereas the accumulation of mancozeb is similar at all the exposure times tested. This indicates a different uptake and/or metabolizing strategy by E. coli cells for the two fungicides.

Quantitative and kinetic study of oxidative stress regulons using green fluorescent protein.

Potentially damaging reactive oxygen species (ROS) are involved in a number of pathways ranging from signal transduction to apoptosis. Cells have adapted this alteration in redox status into a complex regulatory mechanism. ROS are specifically able to induce the expression of a multitude of genes. We constructed and characterized "oxidative stress probes" consisting of promoter fusions of several ROS-induced genes and the green fluorescent protein (GFP) reporter gene. Specifically, the sodA, fumC, zwf, acnA, acrAB, and soxS genes from the SoxRS regulon and the katG and ahpC genes from OxyR regulon, which respond to the superoxide anion and hydrogen peroxide, were studied. Our results revealed not only different levels of background transcription, but different induction levels both in terms of timing and strength. These systematic studies were performed under a uniform parallel platform and have provided insight into the complicated gene regulation of the oxidative stress regulons.

Evaluation of antimicrobial agents for veterinary use in the ecotoxicity test using microalgae.

The influence of antimicrobial agents approved as veterinary drugs in Japan on the growth of green algae, Selenastrum capricornutum and Chlorella vulgaris, was studied in accordance with the OECD guidelines for testing chemicals. Among the agents tested, growth inhibitory activity was very varied, i.e. erythromycin showed the strongest activity (EC50, 50% effective concentration, = 0.037 mg/l), sulfa drugs had activity to some extent (EC50s of sulfamethoxazole, sulfadiazine, and sulfadimethoxine were 1.5, 2.2, and 2.3 mg/l, respectively), but ampicillin and cefazolin did not inhibit growth (EC50s>1000 mg/l). We also investigated synergistic effect of combining sulfa drugs with trimethoprim or pyrimethamine, which are commonly used as a combined drug. By adding trimethoprim, the growth inhibitory activity of sulfamethoxazole and sulfadiazine was significantly enhanced. Growth inhibition by sulfa drugs was reduced by the addition of folic acid, indicating that they inhibit folate synthesis in green algae.

In vitro and in vivo antipseudomonal activity, acute toxicity, and mode of action of a newly synthesized fluoroquinolonyl ampicillin derivative.

Compounds N-(6,7-difluoroquinolonyl)-ampicillin (AU-1) and N-(6-fluoroquinolonyl)-ampicillin (FQ-1), synthesized by coupling of the carboxyl group of 6,7-difluoroquinolone (FP-3) and 6-fluoroquinolone (FP4), respectively, with the alpha-amino-group of ampicillin side chain, exhibit antipseudomonal activity similar to and lower acute toxicity than that of norfloxacin, whereas neither ampicillin nor the fluoroquinolone moieties, compound FP-3 or FP4, alone have such activity. Also, AU-1 and FQ-1 are active against tested clinical isolates of Pseudomonas aeruginosa that are highly resistant to norfloxacin, gentamicin, or both. The therapeutic efficacies of FQ-1 and norfloxacin were assessed and compared in neutropenic mice infected with a 90% lethal dose of P aeruginosa. Mice intraperitoneally administered FQ-1 (10 mg/kg) 4, 8, 24, and 48 hours after infection had survival rates as high as 80%, comparable to those of mice treated with norfloxacin at the same dosage and dosing schedule. The study of protoplast formation revealed that FQ-1 did not inhibit cell-wall biosynthesis but did induce cell filamentation of Bacillus subtilis at a level close to its minimal inhibition concentration. Both AU-1 and FQ-1 were able to intercalate into the double-stranded DNA. However, that FQ-1 lost such activity after it was treated with penicillinase suggests that the lactam-ring structure in ampicillin moiety of FQ-1 was hydrolyzed by penicillinase and that the hydrolyzed structure of FQ-1 does not own DNA-intercalation activity.