Technical Note pubs.acs.org/ac
Thinking Outside the “Bug”: A Unique Assay To Measure Intracellular Drug Penetration in Gram-Negative Bacteria Ying Zhou,*,†,# Camil Joubran,*,† Lakshmi Miller-Vedam,‡,◇ Vincent Isabella,‡,§ Asha Nayar,‡ Sharon Tentarelli,† and Alita Miller‡ †
Department of Chemistry, Infection Innovative Medicines Unit, AstraZeneca-US, 53 Gatehouse Drive, Waltham, Massachusetts 02451, United States ‡ Department of Biosciences, Infection Innovative Medicines Unit, AstraZeneca-US, Waltham, Massachusetts 02451, United States ABSTRACT: Significant challenges are present in antibiotic drug discovery and development. One of these is the number of efficient approaches Gram-negative bacteria have developed to avoid intracellular accumulation of drugs and other cell-toxic species. In order to better understand these processes and correlate in vitro enzyme inhibition to whole cell activity, a better assay to evaluate a key factor, intracellular accumulation of the drug, is urgently needed. Here, we describe a unique liquid chromatography (LC)-mass spectrometry (MS) approach to measure the amount of cellular uptake of antibiotics by Gram-negative bacteria. This method, which measures the change of extracellular drug concentration, was evaluated by comparing the relative uptake of linezolid by Escherichia coli wild-type versus an efflux pump deficient strain. A higher dosage of the drug showed a higher accumulation in these bacteria in a dosing range of 5−50 ng/mL. The Escherichia coli efflux pump deficient strain had a higher accumulation of the drug than the wild-type strain as predicted. The approach was further validated by determining the relative meropenem uptake by Pseudomonas aeruginosa wild-type versus a mutant strain lacking multiple porins. These studies show great promise of being applied within antibiotic drug discovery, as a universal tool to aid in the search for compounds that can easily penetrate bacterial cells.
O
compounds lacking suitable chromophores, MS can be utilized to analyze compounds and is more sensitive and specific than LC detection. A few LC-MS methods were described for measurement of nonradiolabeled compound accumulation in Pseudomonas aeruginosa,18 Mycobacterium smegmatis,19 and other bacteria.20 Wash steps, to remove compound in the residue medium and on the cell surface, were involved in these methods, which greatly increases the risk of membrane disruption and possible disturbance in drug equilibrium, thereby increasing potential inaccuracy. Laborious approaches using a mixture of 1-bromododecane and 1,6-dibromodecane21 or silicon oil22 were reported to rapidly separate the cells from medium prior to cell lysis, thereby not disturbing the internal/ external drug equilibrium. To meet the pace of an iterative chemistry program, a simple, reproducible, and higher throughput assay is urgently needed. One pioneer study should be emphasized herein. Forty years ago, Nikaido measured the extracellular compound concentration using ultraviolet−visible (UV) detection, established the kinetics of influx, and discussed effect of temperature.23 However, in the absence of currently available technologies, this method was limited to evaluation of
ver the past decades, infections due to drug-resistant Gram-negative bacteria have continuously increased.1 Although there is an urgent need to address the treatment of multidrug-resistant bacteria using novel combination approaches or new antibacterial drugs, the pharmaceutical industry has reduced the investment in antimicrobial drug discovery due to considerable scientific challenges, increased costs, and lengthy clinical trials.2,3 Major resistance mechanisms include: drug inactivation by periplasmic enzymes,4,5 (lack of) cell permeability,6 compound efflux,7,8 reduced susceptibility due to biofilm formation,9,10 and the occurrence of mutator11 and persister cells.12 One key factor linking the structure− activity relationship (SAR) between the in vitro enzyme inhibition and the whole cell activity is the intracellular accumulation of the drug. Most existing techniques to determine cellular accumulation of compounds are based on either fluorescent13−15 or radiometric16,17 detection. However, these two methods are not suitable for compounds that lack fluorescent moieties or for which radiolabeled analogs are not available, which is very common for many novel antibacterial candidates currently in preclinical discovery. For candidate drugs containing basic or acidic sites, mass spectrometry (MS) coupled with high performance liquid chromatography (HPLC) can be used to quantitate drug levels at natural isotopic abundances. For © 2015 American Chemical Society
Received: December 31, 2014 Accepted: March 10, 2015 Published: March 10, 2015 3579
DOI: 10.1021/ac504880r Anal. Chem. 2015, 87, 3579−3584
Technical Note
Analytical Chemistry compounds that contain UV chromophores. It also required higher dosing concentrations due to the inherent insensitivity of UV compared to MS. A recent study on bacterial multidrug exporters utilized LC-MS technology and measured the drug concentration decrease in the medium as a result of drug accumulation in the cells.24 This study demonstrated the need and consensus for developing a thorough approach to measure the change of drug concentration in the external medium upon cellular uptake. Here, we described such an assay utilizing wellconsidered control groups to achieve better accuracy. The method was validated with wild-type Escherichia coli and effluxpump deficient (ΔtolC) strains using linezolid and was further successfully applied to estimate the accumulation of meropenem in Pseudomonas aeruginosa PAO1 and a mutant derivative strain lacking five major porins. This simple method has great potential to be further developed as a universal approach for analysis of drug uptake for other bacterial species.
■
EXPERIMENTAL SECTION Chemicals and Materials. Linezolid and meropenem were obtained from Sigma Chemical Co. (St. Louis, MO). All solvents were HPLC grade solvents (Merck, Darmstadt, Germany). Preparation of Cells. The experimental design is shown in Figure 1. Strains used are listed in Table 1. E. coli strains and wildtype PAO1 were obtained from the AstraZeneca culture collection. The multiple porin deletion mutant of PAO1 was generated by Microbiotix (Worcester, MA). Overnight cultures of E. coli and P. aeruginosa (Table 1) were grown by shaking (225 rpm) in LB medium (10 g of Bacto-tryptone, 5 g of yeast extract, 10 g of NaCl, 15 g of agar/L) overnight at 37 °C. Cultures were diluted 1:50 in fresh LB medium and grown to mid-log phase (OD600 ∼0.4−0.6), at which time cells were collected by centrifugation (10 min at 5000g). The cell pellet was washed with an equal volume of M9 medium containing 0.2% glucose (M9G; Amresco) and centrifuged again. The resulting cell pellet was resuspended in M9 medium to bring cells to an OD600 of 4.0 for the uptake assay. The assay was carried out in 96-deep-well 2 mL plates (VWR catalog # 40002-014), in a final volume of 1 mL, composed of 0.5 mL of resuspended cells and 0.5 mL of M9 medium containing a certain amount of drug. Blanks without cells were included for each temperature condition to account for nonspecific binding of the compound to the assay plate, compound degradation, and evaporation. One set of six replicate samples was incubated at 37 °C, while another set of samples was incubated at −7 to −4 °C using an ice-salt bath. The final concentrations of the compounds were below their respective minimum inhibitory concentrations (MICs): linezolid ranged from 5 to 50 ng/mL and that of meropenem was 20 ng/mL. After incubation, cells were harvested by centrifugation at 3700g for 10 min at 0 °C. The supernatant was collected for HPLC−MS analysis. Instrumentation: HPLC−MS. Sample analysis was performed using an Agilent 6490 Triple Quadrupole MS equipped with an Agilent 1290 liquid chromatography (Agilent Technologies, Victoria, Australia). The Waters Acquity HSS T3 column (2.1 × 50 mm, 1.8 μm, Waters, Milford, MA) was coupled. A secondary LC-MS system, Waters Xevo Triple Quadruple MS coupled with Waters Acquity UPLC, was utilized, depending on the availability of the instruments. Mobile phase A was water (0.1% formic acid) and B was
Figure 1. Drug accumulation assay. Bacteria were grown in LB medium and resuspended into M9 medium for this assay. The same amount of bacteria was incubated at warm (37 °C) and cold conditions and compared to the blanks (without cells) at these temperatures. The samples were centrifuged after incubation, and the supernatants of the samples were collected for direct LC-MS analysis. Green squares represent Gram-negative bacterial cells, and the black dots represent compounds.
Table 1. Minimum Inhibitory Concentration (MIC) of the Bacterial Strains Used in this Study as Determined in M9 Medium parent strain ATCC27325 ATCC27325 PAO1 PAO1 a
relevant genotype
linezolid MIC (μg/mL)
Escherichia coli wild-type ΔtolC Pseudomonas aeruginosa wild-type Δ5 (ΔoprD ΔopdP ΔopdC ΔopdB ΔopdT)
meropenem MIC (μg/mL)
64 8
NDa ND
ND ND
0.031 2
ND: Not determined.
acetonitrile (0.1% formic acid). The gradient applied was as follows: 0% B from 0 to 0.5 min, 0−95% B from 0.5 to 2 min, and 95% B held from 2 to 2.5 min. The aqueous phase was adjusted back to 100% to re-equilibrate the system for the next run. The flow rate was set to 0.5 mL/min. Divert valve function was applied to send the solvent front to the waste (0−0.6 min). All measurements were performed at room temperature, and 3580
DOI: 10.1021/ac504880r Anal. Chem. 2015, 87, 3579−3584
Technical Note
Analytical Chemistry
such as α-hemolysin.25 Herein, E. coli ΔtolC was used in comparison to wild-type E. coli for the determination of compound uptake. The minimum inhibitory concentration (MIC) of linezolid for E. coli wild-type is >64 μg/mL in M9 medium and 8 μg/mL for the ΔtolC strain (Table 1). E. coli ΔtolC strain was predicted to have higher accumulation of the compound (linezolid). The samples were compared with the blanks (not containing cells) and incubated at the same temperature, and the differences were calculated as the amount taken up by the bacteria. The blank served as a control to normalize for nonspecific binding to the assay plate, compound degradation, and solvent evaporation. Gram-negative bacteria such as E. coli and P. aeruginosa can survive a wide spectrum of growth temperatures. A temperature lower than their minimal growth temperature (4 °C) will significantly reduce the crossmembrane traffic in both active and passive transportations. Therefore, the samples incubated at the low temperature (−7 to −4 °C) were used to normalize the nonspecific binding of drug to bacterial cell walls. Since the experiments typically yielded values in the low end of the calibration curve, special attention was placed on experimental variation. Therefore, six samples were collected for each condition; the tested drug (linezolid) was from the same pool that was diluted in advance in M9 medium, and the same volumes of cell culture and drug solution were mixed. Standard linezolid was evaluated in M9 medium and LB medium using LC-MS; LB medium was found to significantly suppress the signal (data not shown). Because the components of M9 medium were mostly salts and other inorganic compounds, they did not affect the sensitivity or quantification of linezolid during LC-MS analysis. Similar results were obtained for other compounds as well (data not shown). As a result, bacteria were transferred to M9 medium after exponential growth in LB medium. As shown in Figure 2A, the concentration of linezolid in the supernatant was determined using the developed LC-MS method. Relative standard deviations (RSDs) of the measurement within each group were less than 5%. In addition, statistical significance was shown between the E. coli ΔtolC and wild-type strains at 37 °C (p < 0.05), suggesting that linezolid uptake by E. coli ΔtolC cells was significantly greater than the wild-type cells. Calculation of the final uptake amount was done using eq 3.
the autosampler was maintained at 4 °C. Five μL per sample was injected. The optimized parameters were: gas temperature of 200 °C, gas flow of 19 L/min, and the nebulizer gas pressure of 45 psi. Capillary voltage was set at 3500 V. The following multiple reaction monitoring (MRM) transitions were monitored for the analytes of interest: linezolid (+) 338.2 → 148.2 and meropenem (+) 384.0 → 141.0. Collision energy for linezolid and meropenem was set to 18 and 12 V, respectively. Calibration curves were acquired by plotting the standards concentration against its peak area. The quantitation concentration limit was 0.2 ng/mL for linezolid and 2.5 ng/ mL for meropenem. The concentrations injected ranged from 1 to 200 ng/mL for linezolid and 2.5 to 250 ng/mL for meropenem in M9 medium. R2 for each standard curve was at least 0.99. Intraday and interday validation was performed to make sure the LC-MS methods were stable for these compounds. Calculation of Intracellular Accumulation of the Compound. The control in the cold condition (−7 to −4 °C) is a measure for nonspecific binding to the bacterial cell wall (surface). As a result, the intracellular accumulation of the compound can be expressed as Sin = Sin + surface − Ssurface = Sex ‐ change ‐ 37 − Sex ‐ change ‐ cold (1)
where S corresponds to drug amount and subscripts ex, in, and surface refer to the extracellular compartment (medium), intracellular compartment, and the cell wall attachment, respectively. In addition, the subscripts 37 and cold indicate the two temperature conditions. At either 37 °C or the cold condition, the change of the drug amount in the extracellular space (Sex‑change) can be calculated as Sex ‐ change = (C blank − Cex ) × Vex = (C blank − Cex ) × (Vtotal − Vin) (2)
Here, C and V correspond to drug concentration and volume of the compartment, respectively. Cblank and Cex refer to the concentration of the blank control without cells and the concentration measurement of the supernatant after incubation. Vtotal is the total volume of the incubation, which is 1 mL in this report. Vin is the total cell volume. For OD600 of 2, 2 × 109 bacteria are expected in 1 mL. On the basis of the single bacterial volume estimation of 1 μm3, Vin can be considered as 0.002 mL.
Sin = C blank ‐ 37 − Cex ‐ 37 − C blank ‐ cold + Cex ‐ cold
Vtotal − Vin ≈ 1 − 0.002 = 0.998mL ≈ 1mL
Sin(wild ‐ type) = 9.92 − 8.90 − 9.97 + 9.08 = 0.13ng
Combining the two equations above, the total accumulation can be calculated as
Sin(ΔtolC) = 9.92 − 8.35 − 9.97 + 9.48 = 1.08ng
As a result, the total intracellular uptake of linezolid is shown in Figure 2B. After 30 min of incubation with 10 ng/mL linezolid, about a 1 ng (3 pmol) intracellular accumulation was observed for 2 × 109 E. coli ΔtolC cells, and a 0.1 ng (0.3 pmol) uptake of linezolid was detected for 2 × 109 E. coli wild-type cells. After calculation, about 1800 and 180 molecules of linezolid were taken up by each E. coli ΔtolC and wild-type bacterial cell, respectively. Therefore, E. coli cells without a functional TolC efflux pump accumulated more linezolid than the wild-type cells, which matched the whole cell activity result (MIC). Each bacterium is estimated to contain about 20 000 ribosomes, so it could be that all these molecules are associated with their targets. However, whether or not a given compound is completely associated with the target is not the objective of the current study; it is evaluating the ability of antibiotics/drug
Sin = [(C blank ‐ 37 − Cex ‐ 37) × (Vtotal − Vin)] − [(C blank ‐ cold − Cex ‐ cold) × (Vtotal − Vin)]
■
Sin = C blank ‐ 37 − Cex ‐ 37 − C blank ‐ cold + Cex ‐ cold
(3)
RESULTS AND DISCUSSION Attributes of the Method. Unlike conventional methods that measure intracellular accumulation of drugs in bacteria, we designed a fast and convenient approach to determine the reduction of compound concentration in the external culture medium. The E. coli TolC protein acts as a channel in the transport of molecules across the outer membrane and is involved in the export of various molecules including antibiotics, bile salts, organic solvents, and even large molecules 3581
DOI: 10.1021/ac504880r Anal. Chem. 2015, 87, 3579−3584
Technical Note
Analytical Chemistry
Figure 3. (A) Accumulation time courses of 35 ng/mL linezolid in wild-type and pump knockout E. coli. Gray diamonds and the black squares represent the intracellular accumulation for wild-type and the pump knockout (ΔtolC) strains, respectively. (B) Calculated intracellular uptake for various dosing amounts of linezolid (from 5 (X-5) to 50 (X-50) ng/mL) in wild-type and pump knockout E. coli for a 30 min incubation. The results were based on the calculation from the average result of 6 samples; therefore, no error bar was shown.
Figure 2. Extracellular concentration measurement of linezolid that was incubated with E. coli wild-type (wt) and ΔtolC strains resulted in the determination of intracellular accumulation of the drug. Ten ng/ mL linezolid was dosed. (A) Extracellular concentration of linezolid was determined using LC-MS for each condition. A significant difference (p < 0.05) was shown between the wild-type (wt) and pump knockout (ΔtolC) strains at 37 °C (X-37), labeled with an asterisk (∗). B stands for the blank without cells. (B) Calculated total intracellular accumulation amount of linezolid when 10 ng/mL (X-10) drug was dosed using eq 3. The results were based on the calculation from the average result of each group in (A); therefore, no error bar was shown.
showed a linear uptake when it was dosed from 5 to 50 ng/mL (R2 of 0.992 for 5 dosing points). Interestingly, the accumulation for the wild-type strain was extremely low for 5 and 10 ng/mL dosing but much higher for the experiments with more than 15 ng/mL linezolid incubation. Considering the time course result (Figure 3A), it is highly possible that the rate of drug uptake by Gram-negative bacteria is concentration related. Another explanation could be that the efflux pump was saturated at a dose of 15 ng/mL. Importantly, positive concentration of drugs inside the cells is necessary for this assay. Finally, it is possible that, at higher drug concentrations, the bacteria require more time to trigger the self-defense system in response to the antibiotic. It has been reported that cell volume is growth condition dependent.26 However, compared to the total incubation volume (medium) of 1 mL, the error of the total cell volume estimation can be ignored. This can also be considered an advantage to measuring extracellular concentrations of compounds. An important difference between the method reported herein and conventional measurements of intracellular drug concentrations is the elimination of the wash step that is necessary to remove extracellular drug before isolating the cells to lyse. The wash step greatly increases the risk of membrane disruption and possible disturbance in drug equilibrium, thereby increasing potential inaccuracy. Prior to analysis, penetration was rapidly halted by cooling down the cell culture at 0 °C. Separate incubation at warm and cold conditions in 96-well plates could
candidates to permeate bacteria, which is an important factor in determining both spectrum and antibacterial activity. Method Validation. Reproducibility was evaluated on the basis of three independent experiments that were performed by different operators on different days. Two sets of experiments were evaluated: E. coli wild-type bacteria were incubated with 40 ng/mL linezolid and E. coli ΔtolC cells were dosed using 10 ng/mL linezolid based on predetermined MICs. The RSDs for calculated intracellular accumulation of these two experiments were both within 7%. A time course study of compound accumulation was performed with 35 ng/mL linezolid at both warm and cold conditions with E. coli wild-type and ΔtolC strains (Figure 3A). A continuous accumulation was shown in the cells up to 30 min for both the wild-type and the mutant strains; however, a slight decrease was observed at the 1 h point for the wild-type strain. This phenomenon could be a result of adaptation to drug pressure by wild-type cells. The phenomenon was reported previously for radioactive ciprofloxacin in P. aeruginosa wildtype bacteria.18 Also, for an environment of 35 ng/mL linezolid, the efflux mutant cells reached an equilibrium at about 30 min. Cell accumulation for a range of dosing amounts (from 5 to 50 ng/mL) for 30 min was also monitored in this study (Figure 3B). As the dosing amount was increased, the intracellular concentration of drug also increased. The E. coli ΔtolC strain 3582
DOI: 10.1021/ac504880r Anal. Chem. 2015, 87, 3579−3584
Technical Note
Analytical Chemistry
Δ5 strain had much less accumulation than that of its parent PAO1 at 30 and 60 min. Interestingly, meropenem took longer to get into the P. aeruginosa cells than linezolid did into E. coli bacteria. These results also suggested that this assay and/or all the cell accumulation assays need to be optimized on the basis of accumulation time (incubation time) for specific bacterial species and drugs.
be more manageable for different operators and more convenient when handling multiple samples. Utilizing a low dosage of the drug in our method ensured accurate detection of the change of the drug in medium before and after incubation with bacteria. One question that could be raised is how the cell would behave at high versus low drug concentrations. However, all current assays that were developed for cell accumulation were performed at sub-MIC concentrations to ensure the survival of the cells; the correlation between the detailed compound uptake progress and the drug concentration is not fully understood. The assay developed herein helps pave the way toward linkage of the SAR of in vitro enzyme inhibition and bacterial penetration. This approach is more amenable to higher throughput deployment based on its utilization of 96well plates for sample incubation and preparation as well as fewer steps required before LC-MS detection. Extended experiments showed that increased cell concentration could improve the signal-to-noise ratio and decrease the variability of the measurements (data not shown). Large amounts of cells and low amounts of drug were critical to the success of this approach, which has the potential of becoming a typical assay used in the drug optimization process. Biological Application of the New Method. We also applied this approach to determine meropenem accumulation in P. aeruginosa wild-type (PAO1) and porin-deficient (Δ5) strains. Limited classes of antibiotics remain useful for the treatment of P. aeruginosa infections, owing to both intrinsic and acquired resistance determinants. The highly impermeable pseudomonal outer membrane has been one of the most complex challenges to overcome in the development of new drugs against this organism. The carbapenem class of β-lactam antibiotics remains important in controlling pseudomonal infection due to their resistance to degradation by extended spectrum β-lactamases and their high affinity uptake through the pseudomonal outer membrane porin, OprD. However, resistance to carbapenems has been on the rise, in large part due to the loss of the oprD allele through mutational events. The MIC of meropenem against these strains is shown in Table 1. P. aeruginosa Δ5 strain was obtained from the parent strain PAO1 with 5 porin genes knocked out, including oprD and opdP, which are important for meropenem uptake.27,28 Figure 4 shows the different uptake trends between P. aeruginosa PAO1 and Δ5 strains. As predicted, the P. aeruginosa
■
CONCLUSIONS Our method of measuring the intracellular accumulation indirectly offers several advantages, as follows. (1) Our assay has fewer steps and is less laborious compared to currently published methods. (2) Our assay is less likely to yield erroneous data due to the reduced number of steps. (3) This method is useful even for highly potent compounds, as intracellular accumulation of highly potent compounds can be below their detection limit. However, detection of a change in low doses is usually straightforward. (For instance, meropenem against P. aeruginosa PAO1 strain has an MIC of 0.031 μg/mL. It is common to dose sub-MIC amounts of drug to the cells for methods that directly measure intracellular accumulation. As a result, the drug concentration in the cell lysate may fall under the detection limit of quantification.) (4) Due to the elimination of the wash step, our method does not disturb the drug equilibrium between the outside/inside of cell. However, our approach might be less applicable for compounds that are rapidly metabolized in the cell; this remains a problem for all current methods. Further optimization of the assay could alleviate this problem by assessing shorter incubation times. Also note that it is likely the method will need optimization for nonspecific binding for each compound as they vary greatly (data not shown). Herein, the method was tested against Gram-negative bacteria. It is believed that this method can be further developed for use evaluating compound uptake by other bacterial species. We propose that this assay can be used for understanding intracellular concentration of compounds to address questions related to drug accumulation, lack of cellular potency, and resistance mechanisms.
■
AUTHOR INFORMATION
Corresponding Authors
*Tel.: (617)-923-9975. Fax: (617)-923-9980. E-mail:
[email protected]. *Tel.: (781)-839-4605. E-mail: camil.joubran@astrazeneca. com. Present Addresses #
Y.Z.: Ambergen Inc., 313 Pleasant Street, Watertown, Massachusetts, 02472, United States. ◇ L.M.-V.: University of California San Francisco, San Francisco, California, United States. § V.I.: Synlogic, Cambridge, Massachusetts, United States. Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes
The authors declare no competing financial interest.
■
Figure 4. Accumulation of 20 ng/mL meropenem in wild-type (PAO1) and porin knockout (Δ5) P. aeruginosa after a 30 (X-30) or 60 (X-60) min incubation. The results were based on the calculation from the average result of 6 samples; therefore, no error bar was shown.
ACKNOWLEDGMENTS The research leading to these results was conducted as part of the Translocation consortium (www.translocation.eu) and has 3583
DOI: 10.1021/ac504880r Anal. Chem. 2015, 87, 3579−3584
Technical Note
Analytical Chemistry received support from the Innovative Medicines Initiative (IMI) Joint Undertaking under Grant Agreement no. 115525, resources which are composed of financial contribution from the European Union’s seventh framework programme (FP7/ 2007-2013) and EFPIA companies in kind contribution.
■
REFERENCES
(1) Fraimow, H.; Nahra, R. Crit. Care Clin. 2013, 29, 895−921. (2) Fair, R. J.; Tor, Y. Perspect. Med. Chem. 2014, 6, 25−64. (3) Tomayko, J. F.; Rex, J. H.; Tenero, D. M.; Goldberger, M.; Eisenstein, B. I. Clin. Pharmacol. Ther. 2014, 96, 166−168. (4) Bush, K. J. Infect. Chemother. 2013, 19, 549−559. (5) Rennie, R. P. Handb. Exp. Pharmacol. 2012, 45−65. (6) Page, M. G. Handb. Exp. Pharmacol. 2012, 67−86. (7) Nikaido, H.; Pages, J. M. FEMS Microbiol. Rev. 2012, 36, 340− 363. (8) Schweizer, H. P. Expert Opin. Drug Discovery 2012, 7, 633−642. (9) Alhede, M.; Bjarnsholt, T.; Givskov, M.; Alhede, M. Adv. Appl. Microbiol. 2014, 86, 1−40. (10) Lazar, V.; Chifiriuc, M. C. Roum. Arch. Microbiol. Immunol. 2010, 69, 95−107. (11) Canton, R.; Morosini, M. I. FEMS Microbiol. Rev. 2011, 35, 977−991. (12) Wood, T. K.; Knabel, S. J.; Kwan, B. W. Appl. Environ. Microbiol. 2013, 79, 7116−7121. (13) Piddock, L. J.; Jin, Y. F.; Ricci, V.; Asuquo, A. E. J. Antimicrob. Chemother. 1999, 43, 61−70. (14) Ricci, V.; Piddock, L. J. Antimicrob. Chemother. 2003, 52, 605− 609. (15) Ricci, V.; Piddock, L. J. Antimicrob. Agents Chemother. 2000, 44, 2361−2366. (16) Mortimer, P. G.; Piddock, L. J. J. Antimicrob. Chemother. 1991, 28, 639−653. (17) Williams, K. J.; Piddock, L. J. J. Antimicrob. Chemother. 1998, 42, 597−603. (18) Cai, H.; Rose, K.; Liang, L. H.; Dunham, S.; Stover, C. Anal. Biochem. 2009, 385, 321−325. (19) Bhat, J.; Narayan, A.; Venkatraman, J.; Chatterji, M. J. Microbiol. Methods 2013, 94, 152−158. (20) Davis, T. D.; Gerry, C. J.; Tan, D. S. ACS Chem. Biol. 2014, 9, 2535−2544. (21) McLaggan, D.; Naprstek, J.; Buurman, E. T.; Epstein, W. J. Biol. Chem. 1994, 269, 1911−1917. (22) Schumacher, A.; Trittler, R.; Bohnert, J. A.; Kummerer, K.; Pages, J. M.; Kern, W. V. J. Antimicrob. Chemother. 2007, 59, 1261− 1264. (23) Nikaido, H. Biochim. Biophys. Acta 1976, 433, 118−132. (24) Nakashima, R.; Sakurai, K.; Yamasaki, S.; Hayashi, K.; Nagata, C.; Hoshino, K.; Onodera, Y.; Nishino, K.; Yamaguchi, A. Nature 2013, 500, 102−106. (25) Hinchliffe, P.; Symmons, M. F.; Hughes, C.; Koronakis, V. Annu. Rev. Microbiol. 2013, 67, 221−242. (26) Volkmer, B.; Heinemann, M. PLoS One 2011, 6, No. e23126. (27) Hancock, R. E.; Brinkman, F. S. Annu. Rev. Microbiol. 2002, 56, 17−38. (28) Huang, H.; Hancock, R. E. J. Bacteriol. 1993, 175, 7793−7800.
3584
DOI: 10.1021/ac504880r Anal. Chem. 2015, 87, 3579−3584
