The emergence of antibiotic-resistant pathogenic bacteria during the last decades has become a public health concern worldwide. Aiming to explore new alternatives to treat antibiotic-resistant bacteria and given that the tellurium oxyanion tellurite is highly toxic for most microorganisms, we evaluated the ability of sub lethal tellurite concentrations to strengthen the effect of several antibiotics. Tellurite, at nM or µM concentrations, increased importantly the toxicity of defined antibacterials. This was observed with both Gram negative and Gram positive bacteria, irrespective of the antibiotic or tellurite tolerance of the particular microorganism. The tellurite-mediated antibiotic-potentiating effect occurs in laboratory and clinical, uropathogenic Escherichia coli, especially with antibiotics disturbing the cell wall (ampicillin, cefotaxime) or protein synthesis (tetracycline, chloramphenicol, gentamicin). In particular, the effect of tellurite on the activity of the clinically-relevant, third-generation cephalosporin (cefotaxime), was evaluated. Cell viability assays showed that tellurite and cefotaxime act synergistically against E. coli. In conclusion, using tellurite like an adjuvant could be of great help to cope with several multi-resistant pathogens.
Citation: Molina-Quiroz RC, Muñoz-Villagrán CM, de la Torre E, Tantaleán JC, Vásquez CC, Pérez-Donoso JM (2012) Enhancing the Antibiotic Antibacterial Effect by Sub Lethal Tellurite Concentrations: Tellurite and Cefotaxime Act Synergistically in Escherichia coli. PLoS ONE 7(4): e35452. https://doi.org/10.1371/journal.pone.0035452
Editor: Mark R. Liles, Auburn University, United States of America
Received: December 23, 2011; Accepted: March 16, 2012; Published: April 20, 2012
Copyright: © 2012 Molina-Quiroz et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work received financial support from FONDECYT (Fondo Nacional de Ciencia y Tecnología) 1090097 (CCV) and 3100049 (JMPD); IFS (International Foundation for Science, Sweden) 4733-1 (JMPD); DICYT (Dirección de Investigación Científica y Tecnológica, Universidad de Santiago de Chile) (CCV, JMPD). A doctoral fellowship and grant 24100031 from CONICYT (Comisión Nacional de Ciencia y Tecnología) to RCMQ is also acknowledged. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The constant emergence of clinically-relevant pathogens exhibiting high levels of antibiotic resistance is nowadays a worldwide health problem that poses new challenges to the scientific community. Such scenario is even more worrying given the ability of some pathogens to use antibiotics as the sole carbon source .
During the last 50 years, the pharmaceutical industry has introduced only one new antibiotic into the market. However and even if new compounds with antibiotic ability are discovered, the emergence of resistant strains is only a matter of time. This is the main reason to look for new treatments, and in this context the use of compounds strengthening the antibiotic effect is a choice that worth to be evaluated . Recently, it has been reported that some antibiotics can act in a synergistic manner when used in conjunction with genetically-modified bacteriophages or organometallic compounds, among others , , .
In 2007, Kohanski et al.  showed that a common mechanism underlying the toxicity of bactericidal antibiotics involves the generation of the highly reactive oxygen species (ROS), hydroxyl radical. Overall, this observation evidences that the general mechanism(s) underlying antibiotic-toxicity are not fully understood to date. On the other hand, in 1932 Fleming reported the antibacterial properties of tellurite (TeO32−) and penicillin  and since then TeO32− has been used routinely to isolate tellurite-resistant strains as Escherichia coli O157, Proteus spp., and other bacteria .
During the last years our group has been interested in studying the underlying molecular mechanism(s) of tellurite toxicity. It has been shown that part of it results from ROS generation , damage to metabolic enzymes , , glutathione depletion  or lipid peroxidation . In this context, based on the high toxicity exhibited by TeO32− against bacteria, its numerous cell targets  and its apparent low noxiousness to eukaryotic cells , we hypothesized that TeO32− could increase significantly the antimicrobial effect of antibiotics.
In this work we report that sub lethal tellurite concentrations increase the effect of ampicillin, tetracycline, chloramphenicol or cefotaxime against E. coli and Pseudomonas aeruginosa. A similar, but reduced effect was observed with the highly tellurite- and antibiotic-resistant Staphylococcus aureus. Especially interesting was the effect with cefotaxime, a widely-used, third-generation cephalosporin, which was found to act synergistically with tellurite against E. coli.
The ability of non-lethal tellurite concentrations to increase the antibacterial effect was assessed by determining growth inhibition zones. Antibiotics targeting different cellular processes were tested in the absence or presence of sub lethal tellurite concentrations.
Bacterial species displaying distinct susceptibility to antibiotics and TeO32− were evaluated to determine if the potentiating effect was also observed with tellurite- or antibiotic-resistant bacteria. Tellurite-mediated antibiotic-potentiating effects were observed with AMP, CHL, TET, GEN and CTX when E. coli (highly sensitive to tellurite, MIC 4 µM) was grown in tellurite-amended LB plates. Approximately a 3-fold increase in growth inhibition zones was observed with TET, CHL and CTX (Fig. 1A). In turn, when P. aeruginosa was exposed to different antibiotics in the presence of 4 µM TeO32− (MIC/80), a significant potentiating-effect was observed only with CHL and GEN (Fig. 1B). Similar results were obtained with S. aureus grown in 200 µM tellurite (MIC/20)-containing plates (Fig. 1C).
Antibiotic-mediated inhibition growth zones were determined for E. coli (A), P. aeruginosa (B) and S. aureus (C) grown in the absence (white bars) or presence of the indicated tellurite (T) concentrations as described in Methods. Values represent the average of at least 4 independent trials and significance was determined using t-test analysis (p<0.05). Significance values are (*) p<0.05, (**) p<0.01 and (***) p<0.001.
Differences in growth inhibition areas observed among these bacterial species are most probably due to their different susceptibility to tellurite and antibiotics. P. aeruginosa exhibited smaller inhibition zones than E. coli, which may reflect antibiotic resistance genes that are absent in E. coli K12-derived laboratory strains . In our hands and depending on the particular antibiotic, MIC values for these antibacterials decreased 25–75% in the presence of sublethal tellurite concentrations.
Particularly interesting was the effect in E. coli exposed to CTX, where the most significant inhibition zone increase was observed in the presence of tellurite (Fig. 1A). CTX, a third-generation cephalosporin, is routinely used to treat infections caused by Gram-negative and Gram-positive pathogens and also as prophylactic strategy . Given the effect observed in sensitivity to CTX and its clinical relevance, the tellurite-dependent potentiation on CTX effect was further explored.
The minimal concentration of tellurite displaying CTX potentiation was determined. A dose-dependent effect was observed when tellurite concentrations ranging from 1/10 up to 1/1,000 of E. coli MIC were evaluated (Fig. 2). Although the maximal effect was observed at 400 nM, half of this concentration was used since the potentiating effect was still significant and because this concentration seems not to affect eukaryotic cells , , .
Inhibition growth zones were determined as described in Methods using LB plates amended with the indicated sub lethal tellurite concentrations (nM).
The cefotaxime MIC for E. coli was diminished 4 fold (0.13 to 0.03 µg/ml) when grown in the presence of tellurite. Surprisingly, the CTX MIC for the antibiotic-resistant bacteria P. aeruginosa decreased >30 fold (300 to 9.3 µg/ml) in the presence of 4 µM tellurite. Since the CTX MIC is the same for pathogenic  and laboratory E. coli strains, these results could be important in terms of future applications of tellurite-mediated cefotaxime potentiation. In this context and aiming to assess if the tellurite-potentiating antibiotic effect was also observed with pathogenic bacteria, clinical isolates were exposed to both antibacterials. Growth inhibition zones resulting from antibiotic exposure in the presence or absence of 200 or 400 nM tellurite were determined for 20 clinical coliform isolates from patients suffering urinary infection. A dose-dependent, tellurite-potentiating effect was observed with all tested antibiotics. Interestingly, the most robust effect was again observed with CTX, which was over 2 fold than that observed with other antibiotics as STR, AMK, KAN and TOB (Table 1).
To characterize the type of antimicrobial effect after exposing bacteria simultaneously to tellurite and CTX, cell viability determinations were carried out using different antibiotic concentrations in the presence of the tellurium oxyanion (Fig. 3). Growth and cell viability were not severely affected when E. coli was exposed to 200 nM tellurite. In fact, normal growth and viability was restored after 3 h exposure (Fig. 3, squares).
Growth curves (left panels) and cell viability (right panels) were determined at the indicated time intervals for E. coli exposed to 0.065 (A, sublethal), 0.13 (B, MIC) and 0.5 µg/ml (C, lethal) CTX in the absence or presence of 200 nM tellurite. Controls contained no tellurite or cefotaxime. Data represent the mean of at least 3 independent trials. Refer to inset in panel A for symbol meaning.
While growth was not affected when cells were exposed concurrently to 0.065 µg/ml CTX (MIC 0.13 µg/ml) and tellurite (200 nM), the number of viable cells was strongly decreased as determined by CFU counting (Fig. 3A). A similar result was obtained upon exposing to the antibiotic alone, suggesting that the effect observed with both CTX+tellurite depends mainly on antibiotic-mediated damage.
When cells were grown in the presence of 0.13 µg/ml cefotaxime, growth and viability recovered only after 7 h treatment. The observed potentiating-effect at 3 or 7 h exposure cannot be explained as the sum of tellurite- and CTX-independent effects (Fig. 3B). This indicates a tellurite/cefotaxime-mediated synergistic effect in E. coli.
Finally, when the potentiating effect was assessed in cells exposed to lethal CTX concentrations (0.5 µg/ml), the synergy was represented by a difference of ∼5 Δlog10 units after 20 h and growth or cell viability was not recovered at all (Fig. 3C).
Bacterial multi-resistance to different antibiotics has become a severe problem worldwide. To face this situation, the scientific community and pharmaceutical industry have made important efforts to discover new compounds exhibiting antibacterial properties. However, in the last 40 years these efforts have resulted in the discovery of only 2 new antibiotics, the oxazolidinone linezolid and the lipopeptide daptomycin , .
The conventional treatment of bacterial infections currently lies in administering antibiotics alone or in combination , or using last-generation antibiotics as the case of the multi-resistant Enterobacteriaceae with carbapenems . In spite of this, strains resistant to these new antibacterials emerge continuously, making the situation critical.
Horizontal gene transfer is the principal mode of acquiring new information by bacteria thus allowing them to cope with new antibacterial agents. In this context, the idea of using 2 different antibiotics to treat bacterial infections seems reasonable but there is still a risk of acquiring resistance determinants. To avoid multi-resistance emergence, the use of compounds exhibiting multi-target toxicity is an interesting and novel alternative, since getting a mutation or acquiring genetic determinants against these new compounds is minimal. In this context, using molecules as tellurite to potentiate the antibacterial effect seems to be a fine approach. Although information regarding TeO32− toxicity for eukaryotic cells is scarce to date, it has been shown that 50 µM tellurite (>125-fold the maximal dose used in this work) seems not to affect the viability of eukaryotic cells . In fact, in different cell lines death occurs at ∼160–1,600 µM tellurite, as compared to the E. coli 4 µM killing-dose. Despite the important effect in survival observed in neurons  and erythrocytes exposed to 100–500 µM tellurite , no significant effects have been reported when lower concentrations were used . Indeed, a therapeutic use of tellurite as a red cell antisickling agent has been proposed . Although rats receiving 8 µM tellurite daily doses did not reveal toxic effects over a year, tellurite-treated animals showed increased mortality after 19 months , . Despite these considerations, it results obvious that the real effect of tellurite on eukaryotic cells has not been well established to date.
Considering the use of tellurite as strategy to kill bacteria without affecting eukaryotic host cells, it was determined that sub lethal TeO32−concentrations increase the susceptibility of different bacterial species to various antibiotics in either LB or Müeller-Hinton media (not shown). The fact that increased growth inhibition zones were observed with most tested antibiotics in the presence of TeO32− (Fig. 1) suggests that this condition may not be related to the antibiotic's specific target. This allows hypothesizing a common mechanism underlying the observed potentiating effect. As shown by Kohanski et al. for bactericidal antibiotics  and by our group for tellurite , these compounds promote oxidative stress which could in part explain their toxicity. Experiments to address this issue are under way in our laboratory.
Tellurite-mediated potentiation of TET, GEN and CHL is probably consequence of a combined effect upon protein synthesis (mediated by the antibiotic) and tellurite-induced protein oxidation. In this context, E. coli protein misfolding/mistranslation or oxidation has been observed upon exposure to some aminoglycosides  or tellurite , respectively.
Major changes in growth inhibition zones observed with Gram negative bacteria facing simultaneously tellurite and antibiotics are probably consequence of their high tellurite susceptibility as compared to that exhibited by Gram positive microorganisms . Differences in growth inhibition areas between E. coli and P. aeruginosa may be explained because of the high antibiotic-resistance levels exhibited by the last bacterium. In spite of this, its susceptibility to antibiotics can be increased in the presence of low tellurite concentrations (Fig. 1B). On the other hand, a less robust effect was observed when S. aureus was exposed to tellurite and antibiotics, probably because the high resistance to both toxicants exhibited by this Gram positive rod.
A synergistic effect, evidenced by a difference of >2 log units in cell viability, was observed when characterizing the magnitude and the type of tellurite-mediated CTX-potentiating effect (Figs. 3B and C). This was also the case with growth curves, where an important decrease in OD600 was observed when exposing to both antimicrobials (Fig. 3). Although viability was rather unaltered, increased turbidity was observed when E. coli was exposed to 0.13 µg/ml CTX (Fig. 3B), a result that might be explained by cell filamentation upon exposition to β-lactam agents as cefotaxime . Cell viability was recovered only 7 h after exposing to a lethal cefotaxime concentration (0.5 µg/ml) (Fig. 3C), a result that may reflect a decreased antibiotic bioavailability because of covalent linkage formation with bacterial penicillin binding proteins (PBPs), as has been described for other β-lactam antibiotics .
Since hydroxyl radical and superoxide formation occurs during E. coli exposure to tellurite or bactericidal antibiotics, respectively , , the observed tellurite/cefotaxime synergistic effect would be most probably due to an oxidative stress outbreak. This idea is reinforced even with sub lethal antibiotic concentrations, where enhanced DNA damage and mutation rate are observed . Experiments to address this issue are currently being carried out in our laboratory.
The idea of using TeO32− lies on its extremely high toxicity to bacteria as compared to other metals or non metals as chromium, lead, or manganese . In addition to establishing an oxidative stress status (9), the existence of multiple tellurite cell targets (13, 14) makes the emergence of strains resistant to the antibiotic-potentiating strategy is almost negligible.
Our findings strongly suggest that the use of tellurite (or similar antimicrobials) as an antibiotic-potentiating adjuvant is a novel and feasible strategy to face the antibacterial multi-resistance problem. It is also particularly promising given that the antibacterial-potentiating effect was observed with antibiotic-resistant clinical isolates.
Finally, unveiling the molecular mechanism of the antibiotic-potentiating effect described in this work should contribute to the development of new molecules or compounds to be applied in new therapies to treat infections caused by antibiotic-resistant bacteria.
Materials and Methods
Bacterial strains and culture conditions
E. coli BW25113, P. aeruginosa PAO1 and S. aureus were routinely grown in Luria Bertani broth at 37°C with shaking. Minimal inhibitory concentrations (MIC) were determined by serial dilutions as described . Growth inhibition zones were determined as reported previously . Briefly, cells were spread on LB plates amended with TeO32− (0.2, 4 and 200 µM for E. coli, P. aeruginosa and S. aureus, respectively). Sterile filter paper disks (6 mm) containing ampicillin (100 µg, AMP), tetracycline (30 µg, TET), cefotaxime (60 µg, CTX), chloramphenicol (25 µg, CHL) or gentamicin (10 µg, GEN) were placed on the plate centres and incubated overnight at 37°C.
Antibiotic susceptibility of clinical isolates
Twenty uropathogenic E. coli were isolated from patients displaying urinary infection and purified by streaking on MacConkey and LB agar plates. Identification was carried out by conventional microbiological procedures. Cells were grown overnight in LB media (OD600∼0.6) and 50 µl were plated on Müeller Hinton plates that contained or not 200 or 400 nM tellurite. Sensidisks containing cefotaxime (CTX, 30 µg), cefalotin (CEFL, 30 µg), ampicillin (AMP, 10 µg), neomycin (NEO, 30 µg), streptomycin (STR, 10 µg), gentamycin (GEN, 10 µg), amikacin (AMK, 30 µg), kanamycin (KAN, 30 µg) or tobramycin (TOB, 10 µg) were used in disk diffusion assays as described above. Growth inhibition zones were determined for each antibiotic in the absence or presence of 200 or 400 nM tellurite for all 20 isolates and averaged. Results were expressed as the mean of 3 independent trials.
An individual, informed, written consent was obtained from each participant allowing the samples to be used in this study. Procedures for handling clinical samples were carried out in as recommended in Biosafety Laboratory Manual (3rd Ed., WHO, 2005). Microbiological procedures as well as carrying out this study were specifically approved by the Ethics Committee of Facultad de Ciencias, Universidad San Luis Gonzaga, Ica, Perú.
E. coli grown to OD600∼0.3 was exposed to tellurite (0.05 µg/ml) alone or in combination with CTX (0.065, 0.13 or 0.5 µg/ml). At different time intervals samples were serially diluted (1∶10) in PBS buffer, pH 7.0. Ten µl of each dilution were seeded on LB plates to determine the number of colony-forming units (CFU). The CFU/ml was determined using the formula: [(#colonies)*(dilution factor)]/(volume plated in ml) as previously described .
The authors thank Dr. Roberto Vidal, Instituto de Ciencias Biomédicas, Facultad de Medicina, Universidad de Chile, for helpful comments and critical reading of the manuscript.
Conceived and designed the experiments: RCMQ CCV JMPD. Performed the experiments: RCMQ CMM EdT JCT. Analyzed the data: RCMQ JMPD JCT CCV. Contributed reagents/materials/analysis tools: JCT CCV. Wrote the paper: RCMQ JMPD CCV.
- 1. Dantas G, Sommer M, Oluwasegun R, Church GM (2008) Bacteria subsisting on antibiotics. Science 320: 100–103.
- 2. Walsh C (2003) Where will new antibiotics come from? Nat Rev Microbiol 1: 65–70.
- 3. Banin E, Lozinski A, Brady KM, Berenshtein E, Butterfield PW, et al. (2008) The potential of desferrioxamine-gallium as an anti-Pseudomonas therapeutic agent. Proc Natl Acad Sci USA 105: 16761–16766.
- 4. Hemaiswarya S, Doble M (2010) Synergistic interaction of phenylpropanoids with antibiotics against bacteria. J Med Microbiol 59: 1469–1476.
- 5. Lu TK, Collins JJ (2009) Engineered bacteriophage targeting gene networks as adjuvants for antibiotic therapy. Proc Natl Acad Sci USA 106: 4629–4634.
- 6. Kohanski M, Dwyer D, Hayete B, Lawrence C, Collins J (2007) A common mechanism of cellular death induced by bactericidal antibiotics. Cell 130: 797–810.
- 7. Fleming A (1932) On the specific antibacterial properties of penicillin and potassium tellurite. Incorporating a method of demonstrating some bacterial antagonisms. J Pathol Bacteriol 35: 831–842.
- 8. Chapman P, Siddons C, Zadik P, Jewes L (1991) An improved selective medium for the isolation of Escherichia coli O157. J Med Microbiol 35: 107–110.
- 9. Pérez J, Calderón I, Arenas F, Fuentes D, Pradenas , et al. (2007) Bacterial toxicity of potassium tellurite: unveiling an ancient enigma. PLoS ONE 2: e211.
- 10. Calderón I, Elías A, Fuentes E, Pradenas G, Castro M, et al. (2009) Tellurite-mediated disabling of [4Fe-4S] clusters of Escherichia coli dehydratases. Microbiology 155: 1840–1846.
- 11. Castro M, Molina R, Díaz WA, Pradenas GA, Vásquez CC (2009) Expression of Aeromonas caviae ST pyruvate dehydrogenase complex components mediate tellurite resistance in Escherichia coli. Biochem Biophys Res Commun 380: 148–152.
- 12. Turner RJ, Aharonowitz Y, Weiner J, Taylor DE (2001) Glutathione is a target in tellurite toxicity and is protected by tellurite resistance determinants in Escherichia coli. Can J Microbiol 47: 33–40.
- 13. Pérez J, Arenas F, Pradenas G, Sandoval J, Vásquez CC (2008) Escherichia coli YqhD exhibits aldehyde reductase activity and protects from the harmful effect of lipid peroxidation-derived aldehydes. J Biol Chem 283: 7346–7353.
- 14. Chasteen TG, Fuentes DE, Tantaleán JC, Vásquez CC (2009) Tellurite: history, oxidative stress, and molecular mechanisms of resistance. FEMS Microbiol Rev 33: 820–832.
- 15. Sandoval J, Leveque P, Gallez B, Vásquez CC, Buc-Calderon P (2010) Tellurite-induced oxidative stress leads to cell death of murine hepatocarcinoma cells. Biometals 23: 623–632.
- 16. Livermore DM, Woodford N (2006) The beta-lactamase threat in Enterobacteriaceae, Pseudomonas and Acinetobacter. Trends Microbiol 14: 413–420.
- 17. Lepercq J, Treluyer J, Auger C, Raymond J, Rey E, et al. (2009) Evaluation of cefotaxime and desacetylcefotaxime concentrations in cord blood after intrapartum prophylaxis with cefotaxime. Antimicrob Agents Chemother 53: 2342–2345.
- 18. Schroeder H (1967) Effects of selenate, selenite and tellurite on the growth and early survival of mice and rats. J Nutr 92: 334–338.
- 19. Wagner M, Toews AD, Morell P (1995) Tellurite specifically affects squalene epoxidase: investigations examining the mechanism of tellurium-induced neuropathy. J Neurochem 64: 2169–2176.
- 20. Nix D, Schentag J (1995) Role of pharmacokinetics and pharmacodynamics in the design of dosage schedules for 12-h cefotaxime alone and in combination with other antibiotics. Diagn Microbiol Infect Dis 22: 71–76.
- 21. Ford CW, Zurenko GE, Barbachyn MR (2001) The discovery of linezolid, the first oxazolidinone antibacterial agent. Curr Drug Targets Infect Disord 1: 181–199.
- 22. LaPlante K, Rybak M (2004) Daptomycin - a novel antibiotic against Gram-positive pathogens. Expert Opin Pharmacother 5: 2321–2331.
- 23. Ison CA, Hussey J, Sankar KN, Evans J, Alexander S (2011) Gonorrhoea treatment failures to cefixime and azithromycin in England, 2010. Euro Surveill 16: pii = 19833. Available: http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=19833.
- 24. Kumarasamy K, Toleman M, Walsh T, Bagaria J, Butt F, et al. (2010) Emergence of a new antibiotic resistance mechanism in India, Pakistan, and the UK: a molecular, biological, and epidemiological study. Lancet Infect Dis 10: 597–602.
- 25. Asakura T, Shibutani Y, Reilly MP, DeMeio RH (1984) Antisickling effect of tellurite: a potent membrane-acting agent in vitro. Blood 64: 305–307.
- 26. Guillamet E, Creus A, Farina M, Sabbioni E, Fortaner S, et al. (2008) DNA-damage induction by eight metal compounds in TK6 human lymphoblastoid cells: results obtained with the alkaline Comet assay. Mutat Res 654: 22–28.
- 27. Schroeder HA, Mitchener M (1972) Selenium and tellurium in mice. Effects on growth, survival, and tumors. Arch Environ Health 24: 66–71.
- 28. Schroeder HA, Mitchener M (1971) Selenium and tellurium in rats: effect on growth, survival and tumors. J Nutr 101: 1531–1540.
- 29. Kohanski M, Dwyer D, Wierzbowski J, Cottarel G, Collins J (2008) Mistranslation of membrane proteins and two-component system activation trigger antibiotic-mediated cell death. Cell 135: 679–690.
- 30. Horii T, Mase K, Suzuki Y, Kimura T, Ohta M, et al. (2002) Antibacterial activities of beta-lactamase inhibitors associated with morphological changes of cell wall in Helicobacter pylori. Helicobacter 7: 39–45.
- 31. Nicola G, Tomberg J, Pratt R, Nicholas R, Davies C (2010) Crystal structures of covalent complexes of beta-lactam antibiotics with Escherichia coli penicillin-binding protein 5: towards an understanding of antibiotic specificity. Biochemistry 49: 8094–8104.
- 32. Kohanski MA, DePristo MA, Collins JJ (2010) Sublethal antibiotic treatment leads to multidrug resistance via radical-induced mutagenesis. Mol Cell 37: 311–320.