In vitro evaluation of the effect of linezolid and ...

J Antimicrob Chemother 2017; 72: 417– 420 doi:10.1093/jac/dkw427 Advance Access publication 17 October 2016

In vitro evaluation of the effect of linezolid and levofloxacin on Bacillus anthracis toxin production, spore formation and cell growth Breanne M. Head1*, Michelle Alfa1,2, Daniel S. Sitar3, Ethan Rubinstein1† and Adrienne F. A. Meyers1,4,5 1

Department of Medical Microbiology and Infectious Diseases, University of Manitoba, Winnipeg, Canada; 2St Boniface Research Centre, Winnipeg, Canada; 3Department of Pharmacology and Therapeutics, University of Manitoba, Winnipeg, Canada; 4National Laboratory for HIV Immunology, JC Wilt Infectious Disease Research Centre, Public Health Agency of Canada, Winnipeg, Canada; 5Department of Medical Microbiology, University of Nairobi, Nairobi, Kenya *Corresponding author. Department of Medical Microbiology and Infectious Diseases, University of Manitoba, Winnipeg, MB, R3E 0J9, Canada. Tel: +1-204-226-1077; E-mail: [email protected] †This author is deceased.

Received 3 May 2016; returned 11 July 2016; revised 30 August 2016; accepted 12 September 2016 Background: Owing to its ability to form spores and toxins, Bacillus anthracis is considered a bioterror agent. Although current therapeutic strategies can be effective, treatment does not prevent sporulation and toxin production. Objectives: To quantify the combined effect of a protein synthesis inhibitor and a bactericidal agent on B. anthracis toxin production, sporulation and cell growth. Methods: Susceptibility and synergy titrations were conducted on B. anthracis Sterne and 03-0191 strains using linezolid and levofloxacin. The effect of antibiotic exposure on cell viability was evaluated using a continuous medium replacement model. In vitro static models were used to study the effect of linezolid and levofloxacin on sporulation and toxin production. Spores were quantified using the heat shock method. Toxin was quantified via commercial ELISA. Results: Synergy titrations indicated that the combination was synergistic or indifferent; however, in all models antagonism was observed. In the spore model, linezolid resulted in the lowest sporulation rates, while combination therapy resulted in the highest. In the toxin model, linezolid prevented toxin production altogether. Conclusions: This study advances our understanding of the effects of combination therapy on B. anthracis infection. Used alone, linezolid therapy abolishes toxin production and reduces sporulation. These results suggest that studies using a step-wise approach using linezolid initially to stop sporulation and toxin production followed by levofloxacin to rapidly kill vegetative B. anthracis can be recommended.

Introduction Bacillus anthracis (anthrax) is a sporulating bacterium that can be transmitted via contact, ingestion and aerosolization.1 Pathogenesis is attributed to two exotoxins and a capsule, encoded on the pXO1and pXO2 plasmids, respectively, which are essential for full virulence.2 A 60 day course of antibiotics is recommended when anthrax exposure is suspected. 1 Delays in treatment allow toxin accumulation in the body and the rate of survival decreases dramatically. Moreover, antibiotic-resistant strains can be bioengineered with ease.3 Therefore, better therapeutics are required. Recently, levofloxacin, a bactericidal agent that targets DNA replication, was approved as an alternative for treating anthrax.4,5

Additionally, linezolid, a protein synthesis inhibitor, has been suggested as a possible alternative for treating anthrax.6,7 To date, the combined effect of levofloxacin with linezolid on B. anthracis growth, sporulation and toxin production has not been examined. To test the hypothesis that combination treatment (a bactericidal agent + protein synthesis inhibitor) would be the most effective therapy for killing B. anthracis and preventing toxin and spore production, static and continuous medium replacement models were used.

Materials and methods Microorganisms Two B. anthracis isolates, a fully virulent clinical strain (03-0191) and the pXO2-deficient Sterne strain, were acquired from the National

# The Author 2016. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please e-mail: [email protected]

417 Downloaded from https://academic.oup.com/jac/article-abstract/72/2/417/2290947 by guest on 26 April 2018

Head et al.

Microbiology Laboratory (NML). All studies were performed in a Biosafety Level 3 laboratory at the NML. Samples from 2808C glycerol stocks (109 spores/mL) were resurrected by streaking on an LB-Miller plate and incubated overnight at 378C. For bacterial suspension, colonies were suspended in magnesium and calcium CAMHB.8 Suspensions were adjusted to contain 105 cfu/mL using a 0.5 McFarland standard (Thermo Scientific, Ontario). All studies were performed in triplicate.

Antibiotics Levofloxacin (Hospira Incorporated, GA, USA) and linezolid (Pfizer Canada, Quebec) were purchased from St Boniface Hospital (Winnipeg, Canada). Levofloxacin was supplied in a 5% dextrose intravenous solution (5 mg/ mL). Linezolid was supplied in a dextrose intravenous formulation (2 mg/mL).

Susceptibility and synergy testing Susceptibility studies were conducted on both B. anthracis strains with S. aureus ATCC 25923 for quality control. MIC and MBC were determined by the microdilution broth method.8,9 Uninoculated CAMHB and inoculated antibiotic-free CAMHB served as controls. MBC was determined via plating 1 mL from each non-turbid well onto LB-Miller plates and incubating overnight at 378C. MBC was defined as the concentration where there was a 3 log10 cfu/mL reduction versus the original inoculum. Linezolid and levofloxacin synergy was determined using a chequerboard assay (dilutions ranging from 0.03 to 32 mg/L).8,9 Bacterial suspensions were prepared in CAMHB and inoculated into each well. Uninoculated CAMHB and inoculated antibiotic-free CAMHB served as controls. Plates were incubated overnight at 378C. The fractional inhibitory concentration index (FICI) was calculated using FICA +FICB¼FICI (FICA ¼MIC of linezolid in the combination 4 MIC of linezolid; FICB ¼ MIC of levofloxacin in the combination4MIC of levofloxacin). FICI values ,0.5 indicated synergism, FICI values from 0.5 to 4 indicated indifference and FICI values .4 indicated antagonism.10 MICs and FICI determined from susceptibility and synergy testing were the concentrations used for the remainder of the study.

no antibiotic, linezolid, levofloxacin or the combination) were placed in a 378C water bath. Samples (1 mL each) were collected at 0, 2, 4, 6, 8, 24 and 48 h. Counts were quantified via the plate count method.12

Spore studies To promote sporulation, bacterial suspensions were prepared in 14 mL of minimal medium [10% Columbia broth plus 0.1 mM MnSO4 (CB-M)] with no antibiotic, linezolid, levofloxacin or the combination.13 Samples (1 mL each) were taken at 0, 24 and 48 h. Spore counts were determined by the heat shock method.14,15

Toxin (PA83) quantification Bacterial suspensions were prepared in 100 mL of CAMHB and initial inoculum and toxin were determined. Samples (1 and 4 mL for cell and toxin quantification, respectively) were taken at 24 and 48 h. Cell counts were quantified as described above. Toxin samples were filtered through a 0.2 mm cellulose acetate filter (VWR International, Ontario) to remove all bacteria and spores. Toxin was concentrated (by a factor of 4) using Amicon Ultra-0.5 mL Centrifugal Filters (EMD Millipore, MA, USA). Filtrate was collected and PA83 quantified using an ELISA (Alpha Diagnostic, TX, USA). Samples were run in parallel with the positive recombinant controls according to the manufacturer’s instructions and read at 450 nm. PA83 concentration (ng/mL) was determined using the best fit line from the standard curve (Figure S1, available as Supplementary data at JAC Online).

Statistical analysis Data are expressed as the mean+SEM. The relationships of time, treatment and strain with cell counts, sporulation or toxin formation were assessed using a three-way ANOVA. Tukey’s post hoc analyses were conducted on significant results. Two-way ANOVA was applied to nonsignificant results with post hoc tests for significant results.

Results

Continuous medium replacement model (CMRM)

Susceptibility and synergy studies

A CMRM was used to study the response of B. anthracis to antibiotic exposure (Figure 1).11 The pump (Masterflex, Cole-Parmer, IL, USA) was set to a flow rate of 1 mL/min and calibrated according to the manufacturer’s instructions. Reaction flasks (containing 250 mL of bacterial culture with

Both B. anthracis strains had linezolid MICs of 2 mg/L with MBCs .32 mg/L. The MIC and MBC of levofloxacin were 0.3 mg/L for Sterne and 0.5 mg/L for 03-0191. For Sterne, the combination (0.5 and 0.125 mg/L linezolid and levofloxacin, respectively) exhibited a borderline synergistic effect with an FICI of 0.5. For 03-0191, the combination (0.125 and 0.125 mg/L linezolid and levofloxacin, respectively) exhibited indifference with an FICI of 0.56.

Growth in the CMRM CAMHB + a) No antibiotic b) Linezolid c) Levofloxacin d) Combination

Pump

Reaction flask

Waste

Figure 1. Schematic of the CMRM. CAMHB containing either no antibiotic, linezolid, levofloxacin or the combination is pumped from the medium reservoir into the reaction flask (which contains 250 mL of active bacterial culture). As 1 mL of fresh medium is pumped into the reaction flask, used medium (plus bacterial culture) is pumped out into the waste at a similar rate.


Sterne and 03-0191 controls started with 4.1+0.1 and 4.6+ 0.1 log10 cfu/mL, respectively, and reached 5.5 log10 cfu/mL by 48 h (Figure 2a and b). No difference was noted between antibiotic therapies for bacterial killing. All treatments were equally effective (P, 0.001) and decreased counts to ,1 log10 cfu/mL by 48 h.

Effect of antibiotics on sporulation All antibiotic treatments effectively reduced sporulation (P,0.001, Figure 2c and d). Linezolid resulted in the greatest reduction in

JAC

Effects of combination therapy on B. anthracis

Sterne CMRM

03-0191 CMRM

(b)

9

9

7

7

Log10 cfu/mL

Log10 cfu/mL

(a)

5 3

5 3 1

1

0 4 8 12 16 20 24 28 32 36 40 44 48 Time (h)

0 4 8 12 16 20 24 28 32 36 40 44 48 Time (h)

Log10 cfu/mL

9

(d)

7 5 3 1

9

(f)

3

1000

1000

PA83 (ng/mL)

10000

10

Control Linezolid Levofloxacin Combination

5

10000

100

03-0191 SPORE

7

1

0 4 8 12 16 20 24 28 32 36 40 44 48 Time (h) Sterne TOXIN

(e)

PA83 (ng/mL)

Sterne SPORE

Log10 cfu/mL

(c)

0 4 8 12 16 20 24 28 32 36 40 44 48 Time (h) 03-0191 TOXIN

100 10 1

1 0 4 8 12 16 20 24 28 32 36 40 44 48 Time (h)

0 4 8 12 16 20 24 28 32 36 40 44 48 Time (h)

Figure 2. Effects of linezolid and levofloxacin on B. anthracis. (a) Sterne growth in a CMRM. (b) 03-0191 growth in a CMRM. (c) Sterne spore production. (d) 03-0191 spore production. (e) Sterne toxin production. (f) 03-0191 toxin production. The y-axis represents viable counts (log10 cfu/mL) (a, b, c and d) or PA83 (ng/mL) (e and f), while the x-axis represents time. Results are from three separate trials and are reported as the mean+SEM (error bars).

sporulation, while the combination resulted in the lowest; however, this difference was not statistically significant.

Effect of antibiotics on PA83 Both strains started with 3 ng/mL PA83 (limit of detection). Within 48 h, the Sterne and 03-0191 controls produced 1516+797 and 1716+466 ng/mL PA83 (Figure 2e and f). All antibiotic treatments significantly reduced 03-0191 toxin production (P, 0.001 versus control). For Sterne, no toxin was detected for linezolid and minimal toxin was detected for the combination (P¼ 0.001 versus control), although there was still 3.4 log10 cfu/mL after 48 h (Figure S2). Levofloxacin did not have an effect on toxin production since at 24 h PA83 levels were similar to the control. Likewise, at 48 h levofloxacin toxin levels (133+38 ng/mL) were 10-fold greater than those for linezolid or the combination (P,0.030).

Discussion Currently, in the event of an anthrax outbreak, none of the antibiotics targets the bacterium’s ability to form spores or toxins. This paper describes the effect of linezolid and levofloxacin on B. anthracis Sterne and 03-0191 toxin production, sporulation and cell viability. In accordance with initial susceptibility testing, linezolid was bacteriostatic, whereas levofloxacin was bactericidal. 4,6 Synergy titrations indicated that the combination was synergistic or indifferent; however, in all models antagonism was observed. Linezolid resulted in the lowest increase in sporulation, while the combination resulted in the highest. Linezolid also prevented toxin production altogether, while the combination only reduced it. Although studies have looked at the efficacy of single antimicrobials on B. anthracis sporulation in vitro, no studies have


Head et al.

assessed their effect in vivo. Several studies using Clostridium difficile have found that protein synthesis inhibitors are more effective than other antimicrobials in vivo and that in vitro efficacy has translated well into in vivo efficacy.16,17 It was hypothesized that the combination would be the most effective treatment since linezolid would inhibit spore and toxin production, while levofloxacin would rapidly kill the bacteria.6 However, the results herein indicate otherwise. This discrepancy might be due to the mechanisms of action of these two drugs.4,6 According to previous studies, bactericidal drugs are most potent against actively growing cells. 17 Therefore, when levofloxacin and linezolid were combined the inhibition of protein synthesis by linezolid could have resulted in reduction of levofloxacin efficacy, explaining why no synergy was observed. Despite this being the first study looking at the combined effect of linezolid and levofloxacin on B. anthracis killing, previous studies looking at other antibiotic combinations on other species have reached similar conclusions.18,19 It is reasonable to assume that fluoroquinolones in general when combined with protein synthesis inhibitors will not be truly synergistic. Although the combination of linezolid and levofloxacin was antagonistic, linezolid on its own was able to kill B. anthracis, while abolishing toxin production and reducing sporulation. Since our experimental AUCs of linezolid are concordant with in vivo pharmacokinetics in humans, it is likely that linezolid may be a promising alternative for the treatment of anthrax.20 Considering that linezolid kills slowly, in vivo studies looking at a step-wise approach using linezolid initially to stop spore and toxin production followed by levofloxacin to rapidly kill vegetative B. anthracis are recommended.

Acknowledgements We would like to thank the Department of Medical Microbiology at the University of Manitoba for funding these studies. In addition, we thank the Applied Biosafety Research Program at the NML for the use of their facilities. We would also like to acknowledge Dr Daryl Hoban and Ravi Vashisht from the Health Sciences Center Clinical Microbiology Laboratory for their guidance in the antibiotic susceptibility studies.

Funding This study was supported by internal funding.

2 Dai Z, Sirard JC, Mock M et al. The atxA gene product activates transcription of the anthrax toxin genes and is essential for virulence. Mol Microbiol 1995; 16: 1171– 81. 3 Price LB, Vogler A, Pearson T et al. In vitro selection and characterization of Bacillus anthracis mutants with high-level resistance to ciprofloxacin. Antimicrob Agents Chemother 2003; 47: 2362– 5. 4 Ortho-McNeil-Janssen Pharmaceuticals. Levaquin Full Prescribing Information 2008. http://www.accessdata.fda.gov/drugsatfda_docs/ label/2008/021721s020_020635s57_020634s52_lbl.pdf. 5 Li F, Nandy P, Chien S et al. Pharmacometrics-based dose selection of levofloxacin as a treatment for post-exposure inhalational anthrax in children. Antimicrob Agents Chemother 2010; 54: 375– 9. 6 Swaney SM, Aoki H, Ganoza MC et al. The oxazolidinone linezolid inhibits initiation of protein synthesis in bacteria. Antimicrob Agents Chemother 1998; 42: 3251– 5. 7 Louie A, VanScoy BD, Heine HS III et al. Differential effects of linezolid and ciprofloxacin on toxin production by Bacillus anthracis in an in vitro pharmacodynamic system. Antimicrob Agents Chemother 2011; 56: 13– 7. 8 Clinical and Laboratory Standards Institute. Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically— Ninth Edition: Approved Standard M07-A9. CLSI, Wayne, PA, USA, 2012. 9 Sopirala MM, Mangino JE, Gebreyes WA et al. Synergy testing by Etest, microdilution checkerboard, and time– kill methods for pan-drug-resistant Acinetobacter baumannii. Antimicrob Agents Chemother 2010; 54: 4678– 83. 10 Hall MJ, Middleton RF, Westmacott D. The fractional inhibitory concentration (FIC) index as a measure of synergy. J Antimicrob Chemother 1983; 11: 427–33. 11 Zelenitsky SA, Iacovides H, Ariano RE et al. Antibiotic combinations significantly more active than monotherapy in an in vitro infection model of Stenotrophomonas maltophilia. Diagnostic Microbiol Infect Dis 2005; 51: 39– 43. 12 Biesta-Peters EG, Reij MW, Joosten H et al. Comparison of two optical-density-based methods and a plate count method for estimation of growth parameters of Bacillus cereus. Appl Environ Microbiol 2010; 76: 1399– 405. 13 Joshi LT, Phillips DS, Williams CF et al. Contribution of spores to the ability of Clostridium difficile to adhere to surfaces. Appl Environ Microbiol 2012; 78: 7671– 9. 14 Louie A, VanScoy BD, Brown DL et al. Impact of spores on the comparative efficacies of five antibiotics for treatment of Bacillus anthracis in an in vitro hollow fiber pharmacodynamic model. Antimicrob Agents Chemother 2012; 56: 1229– 39. 15 Carlson PE, Walk ST, Bourgis AETet al. The relationship between phenotype, ribotype, and clinical disease in human Clostridium difficile isolates. Anaerobe 2013; 24: 109–16.

None to declare.

16 Mathur T, Kumar M, Barman TK et al. Activity of RBx 11760, a novel biaryl oxazolidinone, against Clostridium difficile. J Antimicrob Chemother 2011; 66: 1087– 95.

Supplementary data

17 Locher HH, Seiler P, Chen X et al. In vitro and in vivo antibacterial evaluation of cadazolid, a new antibiotic for treatment of Clostridium difficile infections. Antimicrob Agents Chemother 2014; 58: 892– 900.

Figures S1 and S2 are available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/).

18 Jawetz E, Gunnison J. Antibiotic synergism and antagonism: an assessment of the problem. Pharmacol Rev 1953; 5: 175– 92.

References

19 Frean J, Klugman KP, Arntzen L et al. Susceptibility of Bacillus anthracis to eleven antimicrobial agents including novel fluoroquinolones and a ketolide. J Antimicrob Chemother 2003; 52: 297– 9.

1 Inglesby T, O’Toole T, Henderson DA et al. Anthrax as a biological weapon, 2002: updated recommendations for management. JAMA 2002; 287: 2236– 52.

20 Pharmacia and Upjohn. Zyvox. Pfizer Inc., New York, NY, USA, 2005. http:// www.fda.gov/ohrms/dockets/ac/06/briefing/2006-4254b_02_05_KP% 20LinezolidFDAlabel52005.pdf.

Transparency declarations
