D-cycloserine or similar physiochemical compounds may be uniquely suited for use in Bacillus anthracis spore decontamination strategies

Authors

  • T.O. Omotade,

    1. Bacteriology Division, United States Army Medical Research Institute of Infectious Diseases (USAMRIID), Fort Detrick, Frederick, MD, USA
    Search for more papers by this author
    • These authors contributed equally.

  • J.D. Heffron,

    1. Bacteriology Division, United States Army Medical Research Institute of Infectious Diseases (USAMRIID), Fort Detrick, Frederick, MD, USA
    Search for more papers by this author
    • These authors contributed equally.

  • C.P. Klimko,

    1. Bacteriology Division, United States Army Medical Research Institute of Infectious Diseases (USAMRIID), Fort Detrick, Frederick, MD, USA
    Search for more papers by this author
  • C.L. Marchand,

    1. Bacteriology Division, United States Army Medical Research Institute of Infectious Diseases (USAMRIID), Fort Detrick, Frederick, MD, USA
    Search for more papers by this author
  • L.L. Miller,

    1. Bacteriology Division, United States Army Medical Research Institute of Infectious Diseases (USAMRIID), Fort Detrick, Frederick, MD, USA
    Search for more papers by this author
  • S.A. Halasahoris,

    1. Bacteriology Division, United States Army Medical Research Institute of Infectious Diseases (USAMRIID), Fort Detrick, Frederick, MD, USA
    Search for more papers by this author
  • J.A. Bozue,

    1. Bacteriology Division, United States Army Medical Research Institute of Infectious Diseases (USAMRIID), Fort Detrick, Frederick, MD, USA
    Search for more papers by this author
  • S.L. Welkos,

    1. Bacteriology Division, United States Army Medical Research Institute of Infectious Diseases (USAMRIID), Fort Detrick, Frederick, MD, USA
    Search for more papers by this author
  • C.K. Cote

    Corresponding author
    1. Bacteriology Division, United States Army Medical Research Institute of Infectious Diseases (USAMRIID), Fort Detrick, Frederick, MD, USA
    • Correspondence

      Christopher K. Cote, Bacteriology Division, United States Army Medical Research Institute of Infectious Diseases (USAMRIID), 1425 Porter Street, Fort Detrick, Frederick, MD 21702-5011, USA. E-mail: christopher.cote2@us.army.mil

    Search for more papers by this author

Abstract

Aims

As observed in the aftermath of the anthrax attacks of 2001, decontamination and remediation of a site contaminated by the accidental or intentional release of Bacillus anthracis spores is difficult, costly and potentially damaging to the environment. The identification of novel strategies that neutralize the threat of spores while minimizing environmental damage remains a high priority. We investigated the efficacy of d-cycloserine (DCS), an antibiotic and inhibitor of the spore-associated enzyme (alanine racemase) responsible for converting l-alanine to d-alanine, as a spore germination enhancer and antimicrobial agent.

Methods and Results

We characterized the impact of DCS exposure on both germinating spores and vegetative cells of fully virulent B. anthracis by evaluating spore germination kinetics, determining the minimum inhibitory concentrations (MICs) required to affect growth of the bacteria and performing macrophage viability assays. DCS enhanced germination induced by l-alanine and also efficiently killed the newly germinated spores. Furthermore, DCS proved nontoxic to macrophages at concentrations that provided protection from the killing effects of spores. Similar tests were conducted with Bacillus thuringiensis (subspecies kurstaki and Al Hakam) to determine its potential as a possible surrogate for B. anthracis field trials. Bacillus thuringiensis spores responded in a similar manner to B. anthracis spores when exposed to DCS.

Conclusions

These results further support that DCS augments the germination response of spores in the presence of l-alanine but also reveal that DCS is bactericidal towards germinating spores.

Significance and Impact of the Study

DCS (or similar compounds) may be uniquely suited for use as part of decontamination strategies by augmenting the induction of spore germination and then rendering the germinated spores nonviable.

Introduction

Bacillus anthracis is a Gram-positive, spore-forming bacterium that is the aetiological agent of anthrax (Friedlander 2000; Mock and Fouet 2001; Cote et al. 2006a). While anthrax can occur as subcutaneous or gastrointestinal forms, inhalational anthrax is the most serious form and if untreated is nearly uniformly lethal (Friedlander 2000; Mock and Fouet 2001; Cote et al. 2006a; Turnbull 2008). One key aspect of anthrax pathogenesis is the fact that the spore is the infectious particle (reviewed in Cote et al. 2011). Bacillus anthracis spores are extremely resistant to environmental insults, including temperature variations and desiccation (Setlow et al. 2001; Driks 2002, 2003) making environmental decontamination after an accidental or intentional release challenging.

The few documented examples of wide-area contamination by B. anthracis spores demonstrate the complexities involved when trying to mitigate public health concerns and minimize environmental damage. A notable example involves British military forces purposely contaminating areas of Gruinard Island off the coast of Scotland, UK, during the 1940s. Offensive weapons testing resulted in sections of the island becoming heavily contaminated with virulent B. anthracis spores (Manchee et al. 1981, 1983). Over 40 years later, it was reported that efficient decontamination was achieved by the application of 5% formaldehyde diluted in sea water (Manchee et al. 1994). This operation required miles of tubing to distribute c. 25–50 l of formaldehyde solution per square metre (Manchee et al. 1994). Prior to the administration of the decontaminant, both physical and chemical (herbicide) means were used to remove the vegetation before the soil was saturated with the formaldehyde solution. The method used to make Gruinard Island habitable again is not practical in an emergency situation, poorly reproducible in an urban setting and fails to take environmental and public health concerns into consideration.

The most well-known example of an accidental release of B. anthracis spores occurred at a Soviet laboratory in Sverdlovsk (now Yekaterinburg, Russia) in 1979 (Meselson et al. 1994). This was determined to be a substantial release that resulted in many human mortalities (Abramova et al. 1993; Grinberg et al. 2001). As this was an unintentional release during the Cold War between the U.S. and U.S.S.R., many of the details were not obtained until many years after the incident. Unfortunately, little information has been revealed about how this spore release was remediated, and the details of any decontamination are sparse at best.

Most recently, the U.S. anthrax letter attacks in 2001 demonstrated the devastating impact that the intentional release of B. anthracis spores can have on public health, infrastructure and society as a whole. Twenty-two cases of anthrax were confirmed, and five people lost their lives as a result of these attacks (Jernigan et al. 2001, 2002). By some estimates, the combined cost of medical response and environmental/building remediation approached one billion U.S. dollars (Campbell et al. 2012; Schmitt and Zacchia 2012), and the overall impact on the U.S. has been immeasurable. These brief descriptions of the accidental or intentional release of B. anthracis spores serve as reminders that the concept of wide-area decontamination after such an incident is neither trivial nor improbable.

The concept of inducing germination to render B. anthracis spores susceptible to decontamination strategies is not novel (Gould et al. 1968; Giebel et al. 2009; Indest et al. 2009; Ghosh and Setlow 2010; Nerandzic and Donskey 2010). However, attention has been newly placed on this strategy to devise decontamination methods that are both effective and less detrimental to the environment. By inducing spores to germinate, secondary decontamination would be considerably easier than it would be when dealing with ungerminated spores; in fact, we hypothesize that the majority of the germinated spores would also be susceptible to normal environmental conditions (i.e. sunlight, desiccation). We have previously shown that germinated spores are significantly less infectious than ungerminated spores in a mouse intranasal model of inhalational anthrax (Cote et al. 2009). Collectively, these facts suggest that the potential hazard to first responders at a spore-release site could be significantly reduced by germinating the spores as soon as possible. With this strategy in mind, we have re-evaluated the utility of the antibiotic d-cycloserine (DCS).

DCS (commercially available as seromycin) is an antibiotic compound derived from Streptomyces and is a vegetative cell wall synthesis inhibitor (Noda et al. 2004a,b). The antibiotic inhibits alanine racemase, which is an enzyme responsible for the conversion of l-alanine to d-alanine (a major bacterial cell wall component) (Noda et al. 2004a,b). DCS has been used for treating tuberculosis and other bacterial infections starting in the mid-1950s (Epstein et al. 1955, 1956). It is currently a second line of choice for tuberculosis treatment and would be prescribed in combination with other antibiotics, such as rifampin or isoniazid (American Thoracic Society, Centers for Disease Control and Prevention, and Infectious Disease Society of America 2003; Di Perri and Bonora 2004; Gordin and Masur 2012). DCS is not a preferred treatment for bacterial infections due in part of the discovery of much more effective antibiotics but also because of the central nervous system-associated side effects associated with DCS (i.e. headache, depression, irritability, etc.). These side effects have been shown to be, in part, caused by the binding of DCS to N-methyl-d-aspartate receptors that are responsible for synaptic plasticity (Rouaud and Billard 2003; Billard and Rouaud 2007). Interestingly, these unintended effects have resulted in DCS being used for the treatment of personality/anxiety disorders as well as in investigational studies on its effects on neurodegenerative diseases (Tsai et al. 1999; Bailey et al. 2007; Goff et al. 2008; Ho et al. 2011).

Alanine racemase activity has been associated with germination inhibition in spores of Bacillus species. The alanine racemase enzyme has been located in both the exosporium and coat layers of B. anthracis spores (Boydston et al. 2006; Steichen et al. 2007). l-alanine is a significant inducer of spore germination, whereas d-alanine is a potent (yet transient) inhibitor of spore germination (Ireland and Hanna 2002; Moir 2003; Akoachere et al. 2007; McKevitt et al. 2007). It is currently thought that by converting l-alanine to d-alanine, spores delay germination until the environment is best suited for the subsequent outgrowth of vegetative bacilli. It has been demonstrated that ungerminated Bacillus spores are capable of producing amounts of d-alanine that are sufficient to autoinhibit their germination (Anmuth et al. 1956; Fey et al. 1964; Yasuda and Tochikubo 1985; Titball and Manchee 1987; McKevitt et al. 2007). Accordingly, DCS has potential dual functions pertaining to B. anthracis decontamination. In theory, it should have antibiotic activities against this organism, but this should be preceded by enhancing spore germination in the presence of the germinant l-alanine (Gould 1966; McKevitt et al. 2007).

As field trials of novel decontamination strategies will be necessary, we have examined the effects of DCS on spores derived from Bacillus thuringiensis subsp. kurstaki and subsp. Al Hakam. Bacillus thuringiensis is a Gram–positive, spore-forming bacterium (member of the Bacillus cereus group of bacteria) characterized by the ability to produce entomicidal crystal (CRY) proteins during sporulation (Schnepf et al. 1998). This unique attribute has made it a fixture as the most prominent agricultural biopesticide worldwide (Crickmore 2006).

Members of the B. cereus group are thought to be derived from one species due to the striking chromosomal homology that exists between them (Daffonchio et al. 2000; Helgason et al. 2000; Read et al. 2003). As the only nonmammalian pathogen within the group, B. thuringiensis has received more attention as a suitable candidate for surrogacy in B. anthracis studies/field trials (Greenberg et al. 2010; Buckley et al. 2012; Buhr et al. 2012; Emanuel et al. 2012; Van Cuyk et al. 2012). In addition, the widespread commercial exploitation of B. thuringiensis CRY proteins was preceded and subsequently supported by extensive tests conducted to ensure its safety, and it has been found to pose little to no harm towards mammalian/vertebrate health (Green et al. 1990; Bernstein et al. 1999). In this report, we examine the impact of DCS on germinating spores and bacilli of strains of B. anthracis and B. thuringiensis.

Materials and methods

Bacteria and media

The fully virulent Ames strain of B. anthracis was used in most experiments (Little and Knudson 1986). Additionally, a geographically diverse strain panel of B. anthracis (USAMRIID Culture Collection) was also used to perform the MIC determination. Bacillus anthracis spores were produced in Leighton and Doi medium and purified with Hypaque/Omnipaque (GE Healthcare, Silver Spring, MD, USA) as previously described (Leighton and Doi 1971; Cote et al. 2006b). Bacillus thuringiensis subsp. kurstaki strain T1B1 was provided by Dr. Henry Gibbons, Biosciences Division, U.S. Army Edgewood Chemical Biological Center (ECBC) Aberdeen Proving Ground, MD, USA (Buckley et al. 2012). Bacillus thuringiensis subsp. Al Hakam was obtained from the Unified Culture Collection (UCC) at USAMRIID. Bacillus thuringiensis spores were made by following the same protocol used for B. anthracis, but cultures were allowed to sporulate in liquid culture for c. 72–96 h at a temperature of 30°C (Yan et al. 2007; Abdoarrahem et al. 2009) as opposed to c. 48 h at 37°C for B. anthracis. Unless otherwise noted, media used in assays described below include Luria-Bertani (LB) (Becton Dickenson and Co., Franklin Lakes, NJ, USA), Dulbecco's minimal essential medium (DMEM) (Thermo Scientific, Lenexa, KS, USA), minimal alanine and inosine germination medium (AI) containing 0·25 mmol l−1 l-alanine (Acros Organics; Thermo Fisher Scientific, Waltham, MA, USA) and 1 mmol l−1 inosine (Sigma-Aldrich, St. Louis, MO, USA), and cation-adjusted Mueller-Hinton broth (CA-MHB) (Becton Dickenson and Co.). In some instances, heat-inactivated foetal bovine serum (FBS) was added to media at a final concentration of 10%. Sheep blood agar (SBA) (Remel-Thermo Scientific, Lenexa, KS, USA) or LB plates were used for determining colony-forming units (CFU) in samples.

Germination assays

We employed two assays to monitor germination. The semi-automated fluorescence assay was described in detail previously (Welkos et al. 2004) and used to evaluate highly purified B. anthracis spore preparations. Bacillus anthracis spores were germinated in germinant solutions containing the fluorescent nucleic acid stain Syto-9 (Molecular Probes; Life technologies, Carlsbad, CA, USA); and germination was detected by fluorescent dye uptake by the spores. Germination was allowed to proceed for 1 h at ambient temperature, while fluorescence was detected by a Spectramax M2E (Molecular Devices, Sunnyvale, CA, USA) plate reader programmed with softmax pro 5.2 LS software to record fluorescence (485/530, Excitation/Emission) readings in relative fluorescence units (RFUs) every minute for 60 min. The germination kinetics of spores treated were analysed by using four-parameter logistic regression (graphpad prism v5.00; GraphPad Software, Inc., La Jolla, CA, USA) (Welkos et al. 2004). Multiple regression curves were compared for significance by performing an anova followed by post-anova multiple pairwise comparisons using the Tukey t-test method, as detailed in Fig. 3. The fluorescence dye assay was not used to monitor germination of B. thuringiensis spores because the crystals altered the sensitivity of the assay.

Germination was also measured in B. anthracis and B. thuringiensis by exposing germinating spores at various time points to elevated temperatures (65°C for 30 min) suitable to kill germinated spores, and samples (both pre- and postheat exposure) were enumerated on SBA or LB plates. Plate counts of the samples indicated the quantity of spores that remained heat resistant and thus ungerminated. Additionally, total (unheated) plate counts were used to evaluate the antibiotic effect of DCS (Sigma-Aldrich) on newly germinated spores when compared with samples not containing DCS.

Minimum inhibitory concentration (MIC) assays

MICs for DCS were determined by the broth microdilution method in CA-MHB, according to the methodology of the Clinical and Laboratory Standards Institute (CLSI) using USAMRIID standard operating procedures, as described previously (Heine et al. 2010). Briefly, several colonies from fresh SBA plate cultures of 29 geographically diverse strains of B. anthracis were suspended in CA-MHB at a concentration of c. 106 CFU ml−1. Escherichia coli ATCC25922, Pseudomonas aeruginosa ATCC27853 and Staphylococcus aureus ATCC29213 were used as quality control strains. As an additional control, a MIC assay was performed using ciprofloxacin. The plate was then incubated at 37°C for c. 24 h, and the MICs were then determined visually. Modified MIC assays were also performed with c. 1 × 107 to 1 × 108 CFU ml−1 of vegetative cells or heat-activated purified spores of B. anthracis Ames or B. thuringiensis (subsp. kurstaki or subsp. Al Hakam) that were aliquoted to 96-well round-bottom plates containing LB. Depending upon the experiment, DCS was added in concentrations ranging from 0 to 10 210 μg ml−1. Plates were covered and incubated for 24 h at 37°C. Immediately after inoculating the microtiter plates, the bacterial samples were serially diluted and plated to calculate the number of CFU ml−1 per inoculum. After 24 h incubation, the MICs were determined visually when possible (this was not the case for B. thuringiensis subsp. kurstaki), and in some cases, well contents were resuspended with a pipette and the OD600 nm per well was measured using a plate reader (Molecular Devices). Control wells receiving sterile medium were used as blanks. Data for each test sample were normalized to the OD600 nm of cultures grown without DCS, averaged and subjected to statistical analysis. Additionally, to determine the viability of the cultures after 24 h, representative aliquots from the wells of the 96-well plate were serially diluted and enumerated on LB or SBA plates. In the case of B. thuringiensis (kurstaki), the CRY formation confounded visual changes in optical density; thus, CFU ml−1 counts were essential in verifying MIC values.

Macrophage protection assay

Murine RAW 264.7 cell line macrophages were infected with heat-activated, ungerminated spores of B. anthracis at a multiplicity of infection (MOI) of either 5 or 42 spores per macrophage in wells containing cover slips, the samples were then washed and processed as described previously (Welkos et al. 2002; Cote et al. 2004, 2008). The medium in samples to be incubated further was replaced with DMEM containing 10% FBS, 10% FBS and 5 μg ml−1 gentamycin, or 10% FBS and 1·5 mg ml−1 DCS. These were incubated at 37°C in 5% CO2 for 4 h. Subsequently, the plates were washed with PBS, and some of the sample replicates were immediately processed for staining (t4). A final set of replicate samples was again resuspended in one of the additive mixtures described above and incubated for an additional 20 h before a final PBS wash and staining (t24).

Two replicates per treatment group at each time point were stained with propidium iodide (Molecular Probes-Life Technologies, Grand Island, NY, USA) nucleic acid stain to detect living vs dead macrophages. Wells were washed with PBS and then incubated in PBS containing 7·5 μg ml−1 propidium iodide for 5 min. After washing with PBS, the wells were fixed in 2% EM-grade formaldehyde (Tousimis, Rockville, MD, USA) in PBS. After a final PBS wash, the cover slips were removed and mounted onto a slide with Vectashield (Vector Labs, Burlingame, CA, USA) and visualized with fluorescence microscopy (535 nm excitation).

Two replicates per treatment group at the t0 time point were subjected to a fluorescent immunostain to determine the number of spores phagocytosed per macrophage as previously described (Welkos et al. 2002; Cote et al. 2004, 2008). Trypan blue (Sigma Aldrich) staining was performed to assay for toxicity of gentamycin or DCS for uninfected macrophages by performing the same procedure as detailed above except that replicate wells contained no cover slips and all wells were stained as follows at t0, t4, or t24 h. After washing wells with PBS, wells were incubated in 0·08% trypan blue in PBS for 5 min and then washed again. Unstained (live) cells and stained (dead) cells were counted with an inverted light microscope.

Statistics

The germination kinetics of spores treated were analysed using a four-parameter logistic regression model (graphpad prism v5.00) and as described previously (Welkos et al. 2004). Standard methods were used to determine statistical significance and to analyse the data and included the mean, standard error of the mean (SEM) and unpaired Student's t-tests. In comparing groups, a P-value of <0·05 was considered to indicate a statistically significant difference.

Results

DCS augments germination of Bacillus anthracis spores

We expanded on the work previously published by Gould et al. and McKevitt et al. and examined the ability of DCS to augment L-alanine-induced germination (Gould 1966; McKevitt et al. 2007). When used as a sole germinant, a greater concentration of l-alanine is needed to induce germination of B. anthracis spores than what is required by other Bacillus species (e.g. B. cereus) (Barlass et al. 2002; Yi and Setlow 2010). Bacillus anthracis requires a cogerminant in addition to l-alanine for optimal germination (Ireland and Hanna 2002). As demonstrated in Fig. 1, the addition of inosine as a cogerminant significantly lowered the amount of l-alanine required to trigger spore germination. Spores incubated in the presence of 0·25 mmol l−1 l-alanine with 0·5 mmol l−1 and higher levels of inosine germinated to a greater extent than did those incubated with 100 mmol l−1 l-alanine alone (Fig. 1). Also, DCS significantly augmented the initiation of germination by spores incubated with l-alanine in either the absence or the presence of a cogerminant (Figs 2 and 3, respectively). As shown in Fig. 2, the impact of DCS on germination is lessened as the amount l-alanine is increased. When added to AI germinant, 1021 μg ml−1 DCS appeared to optimally enhance the germination of Ames strain spores (Fig. 3). Although, higher amounts of DCS (i.e. 5105 μg ml−1) seemed to inhibit germination as determined by the fluorescence-based kinetic assay. We hypothesized that greater concentrations of DCS were resulting in the killing of newly germinated spores, thus affecting the interactions of the spores with the fluorescent dye and giving the impression of decreased germination. Also, it is important to note that defined minimal germination-induction media, such as AI, are sufficient to support spore germination, but are not nutritionally complex enough to support the transition into replicating vegetative cells, suggesting that DCS was acting directly on germinating or germinated spores (Welkos et al. 2004).

Figure 1.

Fluorescent assay of Bacillus anthracis spore germination in the presence of l-alanine with different concentrations of inosine (Ino). Spores were germinated in phosphate-buffered saline, 100 mmol l−1 Ala, 0·25 mmol l−1 Ala + 0·5 mmol l−1 Ino, 0·25 mmol l−1 Ala + 1 mmol l−1 Ino, 0·25 mmol l−1 Ala + 5 mmol l−1 Ino, or 0·25 mmol l−1 Ala + 10 mmol l−1 Ino. Media components are indicated by each line. All pairwise comparisons differed significantly using an anova with Tukey multicomparison post tests (< 0·001) except the comparison of the Ala + Ino 5·0 mmol l−1 and Ala + Ino 10·0 mmol l−1 curves, which were not significantly different. Equivalent results were obtained in at least two individual experiments.

Figure 2.

The effects of d-cycloserine (DCS) on Ala-induced germination of B. anthracis spores as assessed in the fluorescent germination assay. Germination medium was supplemented with (dashed line) or without (solid lines) 1021 μg ml−1 DCS. Spores were germinated in 25 mmol l−1 Ala (a), 100 mmol l−1 Ala (b), or 200 mmol l−1 Ala (c). The data shown in each panel were significantly different with < 0·0001, < 0·0001 and = 0·0155, respectively, for the graphs in panels a, b and c. Media components are indicated by each line. Equivalent results were obtained in at least two individual experiments. Data were analysed by nonlinear regression analysis which was confirmed by unpaired t-tests using graphpad prism5 software.

Figure 3.

Bacillus anthracis spore germination augmentation by d-cycloserine (DCS). Spores were germinated in phosphate-buffered saline or in the standard concentration (AI) of 0·25 mmol l−1 Ala and 1 mmol l−1 Ino as germinants. The AI germinants were supplemented with DCS as follows: 1021, 2553 or 5105 μg ml−1. Germination in AI alone was significantly greater than that in AI + 5105 μg ml−1 DCS (< 0·001). Germination in AI + 1021 μg ml−1 DCS was significantly greater than that in AI alone or AI + 5105 μg ml−1 DCS (< 0·001); but germination in AI + 1021 μg ml−1 DCS was not significantly different from that in AI + 2553 μg ml−1 DCS. Germination in AI + 2553 μg ml−1 DCS was significantly greater than that in AI + 5105 μg ml−1 DCS (< 0·001). Media components are indicated by respective each line. Data were analysed by regression analysis and anova with Tukey pairwise comparison post tests. Equivalent results were obtained in at least two individual experiments.

We utilized other assays designed to evaluate spore germination because germination is a multistep series of events leading to vegetative outgrowth of bacterial cells (Moir 2006). In addition to the semi-automated fluorescent spore germination assay (which evaluates the kinetics of Stage I germination initiation), we examined the loss of heat resistance, a hallmark characteristic of ungerminated resistant spores. This method also demonstrated significant augmentation of germination by the presence of DCS (in a dose-dependent manner) in the germination medium (Fig. 4). We also confirmed that treatment of spores with DCS alone does not induce spore germination (data not shown).

Figure 4.

The impact of d-cycloserine (DCS) on Bacillus anthracis spore germination induced by AI as determined by the enumeration of heat-resistant colonies. l-alanine and inosine (10 mmol l−1 each) were added to spores (c. 2 × 108 spores/ml final concentration) and the effects on germination measured after 1·5 or 8-h incubation. The addition of AI + 1021 DCS (1021 μg ml−1 DCS) resulted in significant augmentation of spore germination as determined by heat sensitivity after 1·5 h germination (*= 0·006). After 8 h germination, significant augmentation of germination was observed with the addition of AI + 408 DCS (408 μg ml−1 DCS) (**= 0·003) or AI + 1021 DCS (***= 0·00001). These data represent at least three experiments and standard error of the mean is depicted. Statistical comparisons were performed with an unpaired Student's t-test.

Cycloserine kills germinating spores of Bacillus anthracis

To confirm our hypothesis regarding the antibiotic potential of DCS on newly germinated spores, we performed modified MIC determinations using spores incubated in LB medium. Nutritionally rich media, such as LB, are sufficient to support both spore germination and the transition to replicating vegetative cells. The MIC of DCS for B. anthracis Ames strain using LB medium was determined to be approximately 153 μg ml−1 DCS (Table 1).

Table 1. MIC values in LB for germinating spores or bacilli
Species and strainMediumMIC of DCS (μg ml−1)
  1. DCS, d-cycloserine; LB, Luria-Bertani; MIC, minimum inhibitory concentration.

  2. a

    Indicates the inhibitory concentration. The minimal bactericidal concentration (MBC) was 510 μg ml−1.

  3. b

    Indicates the MBC (102 μg ml−1).

Bacillus anthracis (Ames)
SporesLB153
BacilliLB102a
Bacillus thuringiensis (kurstaki)
SporesLB102
BacilliLB102b
B. thuringiensis (Al Hakam)
SporesLB102
BacilliLB102b

We further analysed these samples by plating the bacterial culture after 24 h and observed statistically significant killing of the newly germinating bacteria in a dose-dependent manner (Fig. 5a). We also performed similar plating experiments on samples that were washed before plating on agar plates. This was to ensure that the apparent loss in viability was not due to the inhibition of growth on agar by residual DCS carried over from the liquid cultures. Plating experiments using washed bacteria revealed similar levels of bactericidal activity as those observed in unwashed samples (data not shown). We consistently observed a 3- to 4-log reduction in viability when spores were germinated in the presence of an inhibitory concentration of DCS. Importantly, we observed bactericidal activity on newly germinated spores in minimal media (i.e. AI) as well as rich medium (i.e. LB) (Table 1 and Fig. 5a). These results indicate that B. anthracis spore germination is sufficient to render spores susceptible to DCS, and vegetative outgrowth is not required for DCS to be bactericidal.

Figure 5.

Measurement of Bacillus anthracis viability in samples from minimum inhibitory concentration (MIC) assays (Luria-Bertani medium). Spores (a) and vegetative bacilli (b) of B. anthracis from the MIC assays (after 24 h) were serially diluted and plated on Sheep blood agar. The viability of each sample after exposure to the indicated concentrations of d-cycloserine (DCS) was determined by counting the colony-forming units (CFU). None of the samples exhibited a decrease in viability compared to the initial inoculum when DCS was absent. When spores were used (a), the viability decreased significantly (*= 0·0007 compared to 0 μg ml−1, **= 0·0005 compared to 0 μg ml−1 and = 0·004 compared to 102 μg ml−1) in a fashion that corresponded with the inhibitory concentration of DCS as determined by OD600 nm (Table 1). Vegetative B. anthracis bacilli exhibited little to no decrease in viability after DCS treatment with similar concentrations of DCS that were significantly bactericidal against germinating spores, however, decreased viability became more apparent at higher concentrations of DCS (b). At concentrations >255 μg ml−1, there was a significant decrease in recovered CFU ml−1 (+ 0·035 compared to 0  μg ml−1). Data representative from at least two independent experiments. Error bars represent one standard deviation of the mean. Statistical comparisons were performed with an unpaired Student's t-test.

The effects of DCS on vegetative cells

We next evaluated the impact of DCS on vegetative cells of B. anthracis Ames by performing the LB-based MIC assay with viable bacilli as described above for spores. Growth of the cells, determined by increase in optical density, was inhibited by DCS at an approximate concentration of 102 μg ml−1 (Table 1). However, at this concentration, the viability of vegetative cells never declined as dramatically as was observed with germinating spores. After 24 h of DCS exposure in LB, we observed a reduction in viability of <1-log in CFU ml−1 (Fig. 5b). However, higher concentrations (c. 510 μg ml−1 or greater) exhibited a significantly greater bactericidal effect (Fig. 5b). These results support the notion that DCS is generally weakly bactericidal for vegetative B. anthracis bacilli at concentrations that are significantly bactericidal towards germinated spores (Fig. 5a,b).

To further characterize the inhibitory effects of DCS on B. anthracis, vegetative cells of geographically diverse strains were used in a CA-MHB-based MIC assay (Table 2) (Heine et al. 2010). The determined MIC values were fairly consistent across the strain collection and similar to the inhibitory concentration required for killing when the modified LB-based MIC assay was performed. These concentrations could have potential benefit when used in wide-area decontamination remediation strategies.

Table 2. MIC values of various strains of Bacillus anthracis in cation-adjusted Mueller-Hinton broth
B. anthracis strainMIC of ciprofloxacin (μg ml−1)MIC of DCS (μg ml−1)
  1. DCS, d-cycloserine; MIC, minimum inhibitory concentration.

Sterne0·125128
Ames0·03125128
K19380·125128
K59260·0625128
K70380·0625128
SK570·03125128
K79780·0625128
Africa330·03125128
K80910·03125128
K45390·0625128
K97240·01563128
SP500-530·03125128
BA10860·0312564
TX20·0625128
Colorado0·03125128
V7700·125128
NH0·0625128
Texas0·0625128
17T50·0312564
Ohio0·062564
BA00180·125256
BA10240·0625128
BA10070·03125128
K51350·0625128
K18870·03125128
K68350·12564
K27620·03125128
K76650·03125128
K42410·125128
Control strain
Escherichia coli 0·003932
Pseudomonas aeruginosa 0·0625256
Staphylococcus aureus 0·2564

DCS can protect macrophages from infection with Bacillus anthracis

RAW264.7 macrophage-derived cells were protected from an infection with B. anthracis spores when treated with DCS (used at a concentration of c. 1·5 mg ml−1). The concentration of DCS was chosen to be approximately ten times the MIC (Table 1) to account for uncertainties associated with tissue culture conditions and subsequent manipulations. When the infected cell culture was treated with gentamycin (5 μg ml−1), which served as a positive control, the cells were similarly protected (Table 3 and Fig. 6). These results demonstrate that DCS is capable of inhibiting B. anthracis infection in cell culture. Furthermore, these concentrations of DCS did not result in any appreciable toxicity towards the macrophage monolayers, as determined by staining of uninfected cells (data not shown).

Figure 6.

Fluorescent micrographs identifying dead or dying macrophages (which stain red) after infection with Bacillus anthracis spores. Spores alone (a), gentamycin treated – 5 μg ml−1 (b), d-cycloserine treated – 1.5 mg ml−1 (c). These representative micrographs, from three replicates each, were taken from t4 h samples using an approximate multiplicity of infection of 42 spores/macrophage. The presence of macrophage populations of similar cell densities in the different samples was verified by Diff-Quik staining of replicate cover slips.

Table 3. Macrophage protection assaya
TimebMOIcTreatmentdMacrophage viability (%)
  1. DCS, d-cycloserine; MOI, multiplicity of infection.

  2. a

    As measured by PI stain Propidium iodide stain.

  3. b

    Hours of incubation.

  4. c

    Multiplicity of infection (spores/macrophage).

  5. d

    Spores received gentamycin (5 μg ml−1) or DCS (1·5 mg ml−1) as indicated.

  6. e

    Rare viable macrophages and cell debris replaced the cell monolayer; numerous bacteria were also present.

05Spores only>99
Spores + gentamycin>99
Spores + DCS>99
42Spores only>99
Spores + gentamycin>99
Spores + DCS>99
45Spores only16
Spores + gentamycin>99
Spores + DCS>99
42Spores only1·5
Spores + gentamycin>99
Spores + DCS>99
245Spores only e
Spores + gentamycin>99
Spores + DCS>99
42Spores only e
Spores + gentamycin>99
Spores + DCS>99

Experiments performed to evaluate Bacillus thuringiensis as a Bacillus anthracis surrogate

Bacillus thuringiensis continues to garner support as a suitable nonpathogenic surrogate for B. anthracis field tests (Greenberg et al. 2010; Buhr et al. 2012). Taking this into consideration, we evaluated the impact of DCS on germinating spores derived from B. thuringiensis, specifically strains kurstaki and Al Hakam. Spore germination was determined through a loss of heat resistance and an accompanying decrease in CFU ml−1.

After allowing B. thuringiensis spores to germinate in the presence of AI germinant with DCS (1021 μg ml−1), an appreciable increase in germination was observed compared to spores in the presence of AI alone (Fig. 7a,b). The addition of DCS significantly augments the germination of spores in a comparable manner to that observed with B. anthracis.

Figure 7.

The impact of d-cycloserine (DCS) on Bacillus thuringiensis spore germination induced by AI as determined by the enumeration of heat-resistant colonies. l-alanine and inosine (10 mmol l−1 each) were added to the spores and the effects on germination measured after 1·5 or 8 h incubation. The addition of 1021 μg ml−1 DCS resulted in significant augmentation of spore germination as determined by heat sensitivity after 1·5 h germination for both B. thuringiensis isolates, kurstaki (*< 0·005) (a) and Al Hakam (+< 0·00001) (b). After 8 h germination, significant augmentation of germination was also observed with the addition DCS (**< 0·00005 and ++< 0·00001). These data represent at least three experiments and standard error of the mean is depicted. Statistical comparisons were performed with an unpaired Student's t-test.

We further analysed the impact of DCS on both vegetative cells and germinating spores by performing an LB-based MIC. Bacillus thuringiensis appeared to be slightly more sensitive to the effects of DCS than B. anthracis. The experimental MIC for germinating B. thuringiensis spores was observed to be c. 102 μg ml−1 of DCS as compared to c. 153 μg ml−1 for B. anthracis spores (Table 1). While this increased sensitivity was slight in regard to germinating spores (Fig. 8a,c), vegetative cells were more affected by DCS resulting in a greater bactericidal effect on B. thuringiensis cells compared to that observed for B. anthracis vegetative cells (Fig. 8b,d). Although both B. anthracis and B. thuringiensis were inhibited at relatively similar concentrations of DCS, B. thuringiensis vegetative cells experienced a greater decrease in viability. A two to three log decrease, as determined by CFU ml−1, was routinely observed with B. thuringiensis, and the inhibitory concentration was c. 102 μg ml−1 of DCS (Table 1). Notably, both B. thuringiensis spores and vegetative cells experienced the same degree of sensitivity (102 μg ml−1) to DCS and a similar decrease in viability (Fig. 5c,d). Comparable results were observed between both B. thuringiensis isolates while evaluating the effects of DCS.

Figure 8.

Measurement of Bacillus thuringiensis viability in samples from minimum inhibitory concentration (MIC) assays (Luria-Bertani medium). Spores (a and c) and vegetative bacilli (b and d) of B. thuringiensis from the MIC assays (after 24 h) were serially diluted and plated on Sheep blood agar. Bacillus thuringiensis kurstaki (a and b) and B. thuringiensis Al Hakam (c and d) are depicted. The viability of each sample after exposure to the indicated concentrations of d-cycloserine (DCS) was determined by counting the colony-forming units. None of the samples exhibited a decrease in viability compared to the initial inoculum when DCS was absent. When spores were used (a and c), the viability decreased significantly in the presence of DCS (* 0·00005). Vegetative B. thuringiensis bacilli also exhibited a significant decrease in viability after DCS treatment with similar concentrations of DCS (b and d) (**< 0·00005). Data representative from at least three independent experiments. Error bars represent one standard deviation of the mean. Statistical comparisons were performed with an unpaired Student's t-test.

Discussion

We have demonstrated that the addition of DCS to media significantly enhances the rate and extent of germination of both B. anthracis and B. thuringiensis spores in the presence of limited defined germinants. We also demonstrated that DCS exhibits significant inhibitory activity against both bacilli and germinating spores, albeit at relatively high concentrations. DCS was more bactericidal towards germinating B. anthracis spores as compared to the more bacteriostatic effect on vegetative bacilli. DCS has been documented as being either bacteriostatic or bactericidal depending upon the concentration and the susceptibility of a particular bacterium (Curtiss et al. 1965; Clark and Young 1977). Additionally, Clark and Young demonstrated that early logarithmic stage cells of Bacillus subtilis were highly sensitive to DCS, whereas during late stage growth, the authors were more likely to observe resistance to DCS. It was also shown that transport of DCS decreased during these later stages of growth (Clark and Young 1977). Thus, there is precedent for DCS exhibiting differing effects on different stages of bacterial growth.

As a nonpathogenic surrogate bacterium will be required for future field trials, it is critical to document whether proposed strategies are similarly effective on both fully virulent B. anthracis and a chosen surrogate bacterium. In our studies, B. thuringiensis was slightly more susceptible to DCS than B. anthracis (Table 1). This increased sensitivity, while appreciable, was not considered to be significant enough to preclude using B. thuringiensis as a reasonable surrogate for B. anthracis. However, further evaluation is warranted. Bacillus thuringiensis subsp. kurstaki produces copious amounts of entomicidal protein (i.e. CRY) during the sporulation process. This particular characteristic dictated the techniques that could be used to examine germination. To address this, we also worked with B. thuringiensis subsp. Al Hakam, an acrystalliferous strain (CRY-deficient isolate) (Challacombe et al. 2007; Buhr et al. 2012). In both cases, we were able to analyse these spores, and the data illustrated similar effects of DCS on B. thuringiensis spores as compared to B. anthracis spores. To limit the introduction of complicating variables, these studies were performed on B. thuringiensis spores prepared in a very similar manner to the way we normally prepare B. anthracis spores. However, future efforts will focus on spores produced under conditions optimized for B. thuringiensis sporulation.

Contamination resulting from an accidental or intentional release of B. anthracis is both difficult and costly to remediate. Additionally, natural outbreaks can also prove to be difficult to resolve (Turnbull 2008). Ungerminated spores are naturally resistant to most environmental insults and all antibiotics. However, once spores germinate, the resulting cells are once again susceptible. The concept of inducing spore germination to facilitate decontamination has been previously proposed (Gould et al. 1968; Giebel et al. 2009; Indest et al. 2009; Ghosh and Setlow 2010; Nerandzic and Donskey 2010). By inducing germination in spores, the resulting germinated cells would be significantly easier to decontaminate and also significantly less infectious. It has also been demonstrated that germinated spores are susceptible to the antimicrobial activities of macrophages (Guidi-Rontani et al. 2001; Kang et al. 2005). Germinating spores rapidly lose their resistance properties, and subsequent decontamination strategies could be less severe than what would be employed if trying to decontaminate completely ungerminated spores. Additionally, we previously demonstrated that germinated spores are considerably less virulent than ungerminated spores in a murine model of inhalational anthrax (Cote et al. 2009). Thus, if spores were induced to germinate after a contamination event, the potential health hazard to first responders and remediation specialists could possibly be lessened. This would also decrease the potential hazard associated with possible reaerosolization of infectious spores during subsequent decontamination strategies (Layshock et al. 2012). Further work is required to fully elucidate the potential of including germination induction as part of a decontamination strategy; but we believe that this strategy offers great promise to mitigate the repercussions associated with an accidental or intentional release of B. anthracis spores.

Antibiotic resistance, particularly multidrug resistance, is on the forefront of all aspects of microbiology. When weighing the benefits of antibiotic use in the environment, an appropriate risk–benefit analysis should be performed. The introduction of high concentrations of any compound into the environment may negatively impact the flora or fauna (including the human population), and these potential negative impacts must be addressed. In our opinion, the potential benefit of adding DCS may outweigh the risks for several reasons. This strategy would be employed only in rare instances during an emergency clean up scenario, thus, extended periods of exposure should not be necessary. DCS is not a first-choice antibiotic for any infection (it is low on the list of second-line therapeutic options for tuberculosis) (Di Perri and Bonora 2004) and is also a relatively labile compound in certain environmental conditions (Rao et al. 1968). Therefore, any potential adverse effects associated with DCS exposure are predicted to be minimal. Lastly, DCS resistance does not appear to be associated with significant cross-resistance to other antibiotics (Stapley 1958; Howe et al. 1964; Peteroy et al. 2000), and sensitivity to DCS has been restored by growing resistant B. subtilis in the absence of DCS (Clark and Young 1977). For these reasons, we hypothesize that the addition of DCS to potential germination-induction solutions used to facilitate aspects of wide-area decontamination may be beneficial. Our future work includes optimizing a germination induction regimen that would be paired with an effective but environmentally responsible decontamination method.

Acknowledgements

The research described herein was sponsored by the Defense Threat Reduction Agency JSTO-CBD plan CBT-PHM-12-CB2-CB3831 to SLW and CKC. Opinions, interpretations, conclusions and recommendations are those of the authors and are not necessarily endorsed by the U.S. Army.

Conflict of interest

No conflict of interest declared.

Ancillary