Effects of Endogenous d-Alanine Synthesis and Autoinhibition of Bacillus anthracis Germination on In Vitro and In Vivo Infections

Abstract

Bacillus anthracis transitions from a dormant spore to a vegetative bacillus through a series of structural and biochemical changes collectively referred to as germination. The timing of germination is important during early steps in infection and may determine if B. anthracis survives or succumbs to responsive macrophages. In the current study experiments determined the contribution of endogenous d-alanine production to the efficiency and timing of B. anthracis spore germination under in vitro and in vivo conditions. Racemase-mediated production of endogenous d-alanine by B. anthracis altered the kinetics for initiation of germination over a range of spore densities and exhibited a threshold effect wherein small changes in spore number resulted in major changes in germination efficiency. This threshold effect correlated with d-alanine production, was prevented by an alanine racemase inhibitor, and required l-alanine. Interestingly, endogenous production of inhibitory levels of d-alanine was detected under experimental conditions that did not support germination and in a germination-deficient mutant of B. anthracis. Racemase-dependent production of d-alanine enhanced survival of B. anthracis during interaction with murine macrophages, suggesting a role for inhibition of germination during interaction with these cells. Finally, in vivo experiments revealed an approximately twofold decrease in the 50% lethal dose of B. anthracis spores administered in the presence of d-alanine, indicating that rates of germination may be directly influenced by the levels of this amino acid during early stages of disease.

Bacillus anthracis spores transit from the lungs to the bloodstream during early stages of inhalational anthrax (7), and current dogma suggests that spores are engaged by resident macrophages during this step in infection (16). However, unlike vegetative B. anthracis, spores are resistant to killing by the macrophage (10, 14). Moreover, Hu et al. demonstrated that preventing spore germination using an antigerminant protects B. anthracis from macrophage-mediated destruction (10). These findings indicate that delayed germination could work to the advantage of B. anthracis by preventing macrophage-specific killing of the organism during the transition from localized to systemic infection. Hence, factors that influence the efficiency of germination may determine if B. anthracis is destroyed early in infection or survives to cause systemic disease.

The influence of extrinsic and intrinsic factors on B. anthracis spore germination has been the focus of several studies (3, 5, 6, 11, 12, 26, 27). Yet the mechanisms by which these factors impact the timing and the location of B. anthracis germination during infection are poorly understood. More importantly, the factors that could delay germination of B. anthracis during early stages of inhalational anthrax have not been defined. At the most fundamental level, the efficiency of B. anthracis germination is influenced by the available concentrations of germinants and antigerminants. B. anthracis germinants include specific amino acids, sugars, and nucleosides (11, 27). The prominent antigerminant is d-alanine, which is known to repress B. anthracis germination (27). d-Alanine is rarely detected in animals but is found abundantly in the peptidoglycan of eubacteria. d-Alanine is also found in the supernatants of Bacillus spores, and if the level of this amino acid becomes significantly high, germination of the spores is repressed (4). Hence, under appropriate conditions Bacillus species naturally produce d-alanine that can act as an antigerminant, essentially producing its own antagonist of germination. This phenomenon is known as autoinhibition of germination and was first described more than 50 years ago by Anmuth and colleagues (2).

The mechanism of autoinhibition of germination has been studied in the environmental Bacillus species Bacillus subtilis (30), but little is known about this process in the pathogenic Bacillus. The extent of our understanding of autoinhibition of germination in B. anthracis comes from a study by Titball and Manchee, which found that at high density B. anthracis spores germinated less efficiently than when spores were maintained at a much lower density, and this effect was reversed by an alanine racemase inhibitor (23). However, the enzymatic characteristics and the contribution of autoinhibition of germination to pathogenesis have not been investigated. Indeed, this naturally occurring system, which endogenously reduces the efficiency of B. anthracis germination, could contribute to the timing of germination during early stages of inhalational anthrax.

To gain insight into these issues, as described herein, experiments examined the contribution of endogenous d-alanine production to the efficiency of germination and the establishment of B. anthracis infection. The findings from this study indicate that autoinhibition of germination correlates with the density of B. anthracis spore cultures but occurs in a nonlinear fashion, does not require germination, and is due to the production of d-alanine. In addition, experiments show that alanine-racemase activity can alter the outcome of B. anthracis survival in a spore-macrophage assay. Finally, in vivo data are presented supporting the idea that levels of d-alanine can alter the course of infection. Overall, the current findings provide new insight into a mechanism by which the timing of B. anthracis spore germination could be modulated during early stages of inhalational anthrax and alter the outcome of disease.

MATERIALS AND METHODS

Bacterial strains, cell lines, and reagents.

Unless otherwise noted, standard reagents were purchased from Sigma (St. Louis, MO). B. anthracis Sterne strain 7702 (obtained from Theresa Koehler) was used as the reference strain for germination studies and as the background strain for constructing mutants. Abelson murine leukemia virus-transformed murine macrophages derived from ascites of BALB/c mice (termed RAW 264.7) were obtained from the American Type Culture Collection (Manassas, VA). RAW 264.7 cells were maintained in tissue culture grade T-75 flasks grown in the presence of RPMI 1640 medium (Gibco) supplemented with 10% fetal bovine serum (ATCC) and antibiotics (penicillin, streptomycin, and gentamicin). Cells were maintained in a humidified incubator in the presence of 6.0% CO₂ at 37°C and passaged every 48 to 72 h at the point of confluence.

Preparation of B. anthracis spores.

In order to prepare spores of B. anthracis, stationary phase cultures of B. anthracis Sterne 7702 and B. anthracis Sterne 7702 (ΔgerHA) (grown in the presence of 100 μg/ml kanamycin) were diluted into fresh brain heart infusion (BHI) broth and grown to mid-log phase (an optical density at 590 nm [OD₅₉₀] of 0.8). Next, 400 μl of each culture was subsequently inoculated into 40 ml of sporulation medium containing 0.6 mM CaCl₂·2H₂O, 0.8 mM MgSO₄·7H₂O, 0.3 mM MnSO₄·H₂O, 85.5 mM NaCl, and 8 g/liter nutrient broth (pH 6.0) (22). Sporulation cultures were grown at 30°C with constant shaking for 48 h, at which point the cultures were centrifuged at 10,000 × g, supernatant was removed, and the spore pellet was suspended in 10 ml of sterile double-distilled H₂O (ddH₂O). Following a 24-h incubation in ddH₂O at room temperature, the spore suspensions were passed through two in-line glass microfiber syringe filters (3.1 μm and 1.2 μm pore sizes, respectively) (VWR) to remove vegetative organisms from the spore suspension. Spores were pelleted by centrifugation at 18,000 × g, resuspended in 1 ml of sterile ddH₂O, and stored at −20°C. Prior to use, all spore suspensions were diluted to the working concentrations necessary for corresponding experiments and heated to 70°C for 30 min in order to kill any residual vegetative organism as well as to heat activate the spores. Spore counts were determined by dilution plating on BHI agar and counting the resulting CFU.

Monitoring germination using changes in the refractive index of the spore.

To follow spore germination, we took advantage of the changes in the refractive index (OD₆₀₀), which is known to occur during transition from spore to vegetative forms of the organism (17). Spores (3 × 10⁷) were resuspended in 200 μl of the corresponding medium (phosphate-buffered saline [PBS], germinant, or antigerminant) in the well of a 100-well Honeycomb 2 (Thermo Electron Corp.) microplate. Absorbance was monitored every 90 s for 30 min in a Bioscreen plate reader (Bioscreen C) with constant shaking at 37°C. Experiments were carried out in triplicate, and the mean absorbance was determined on three independent spore preparations. For all medium conditions used, standard deviation calculations were consistently less than 2% of the average value and therefore were not presented on the graphs.

Preparation of GIM.

To produce and collect germination inhibiting medium (GIM), spores were allowed to germinate for a 30-min incubation at 37°C in rich or defined medium, and the supernatant was collected by centrifugation at 18,000 × g for 5 min. The supernatants collected from these germinations were subsequently used as GIM in a series of experiments. Where indicated, 200 μl of GIM was supplemented with l-alanine sufficient to achieve a concentration of 0.5 mM l-alanine-1 mM inosine in PBS.

Quantification of d-alanine.

The concentration of d-alanine was determined using an Amplex Red hydrogen peroxide assay kit (Invitrogen). The manufacturer's protocol was performed with the following modification. The working solution of 100 μM Amplex Red reagent and 0.2 U/ml horseradish peroxidase also contained 20 μl of 0.025 U/ml d-amino acid oxidase (Sigma). Fluorescence was measured with a Fluostar Optima fluorescence microplate reader (BMG Labtechnologies) using excitation at 530 nm and fluorescence detection at 590 nm. Samples and standards were measured in triplicate, and the mean and standard deviation were determined. The mean d-alanine concentration for each sample was determined using a standard curve prepared simultaneously with known concentrations of d-alanine. The standard curve was plotted with fluorescence units versus fixed amounts of d-alanine, and the d-alanine concentrations were determined by identifying the x intercept from this standard curve.

Evaluation of rates and efficiency of germination at various spore densities in rich and defined media.

Two experimental approaches were used to determine the density-related effects on the efficiency of germination. In one series of experiments, four concentrations (ranging from 3.5 × 10⁸ to 3.0 × 10⁷ in 200 μl) of spores were examined for changes in OD over time in the presence of l-alanine-inosine with or without d-cycloserine (DCS). These data were then used to calculate the rates for initiation of germination for each condition. Maximal rates (as determined by the ΔOD₆₀₀/min) were identified for each of the conditions and the log 10 values of these rates were plotted against the spore densities in the presence or absence of DCS. In a second series of experiments, seven concentrations of spores, ranging from 1 × 10⁶/ml to 1 × 10⁹/ml, were prepared and suspended in 200 μl of BHI broth. Germination of spores in BHI broth was carried out at 37°C for 30 min. Following the incubation, 1:10 dilutions of each spore concentration were prepared in water, and aliquots were plated on BHI agar medium using the track dilution technique (13). The seven sets of dilutions were heated to 70°C for 30 min in order to heat-kill any spores that germinated, after which aliquots were plated (as described above). The percentage of spores resistant to heat treatment was then calculated by enumerating the CFU following an overnight incubation. These amounts were divided by the total CFU obtained without heating and multiplied by 100 in order to determine percent germination for each experimental condition. Three experiments were carried out on three different days, and the average and standard deviation were determined.

Evaluation of autoinhibition of germination during spore interactions with cultured murine macrophages.

Approximately 1.75 × 10⁶ RAW 264.7 murine macrophage cells were seeded into wells of Multiwell six-well tissue culture-treated plates (Becton Dickinson) in a total of 3 ml of Dulbecco's modified Eagle's medium-10% fetal calf serum. Cells were allowed to attach overnight at 37^°C in a 6% CO₂ atmosphere. Culture medium was removed from the wells and replaced with 2 ml of Dulbecco's modified Eagle's medium-10% fetal calf serum, with or without 10 mM DCS, containing B. anthracis spores sufficient to yield four different multiplicities of infection ([MOIs] 1, 5, 10, and 25). Centrifugation at 200 × g for 10 min was used to bring the spores into immediate contact with the murine macrophages. Medium supernatant was next removed, leaving 300 μl covering the tissue culture cell layer. Cells and spores were incubated then for 50 min at 37^°C in a 6% CO₂ atmosphere prior to being washed six times with PBS, after which 1 ml of water was added to each well. Cells were scraped off the well surface and triturated with a 1-ml pipette five times. A 200-μl aliquot was removed and used to make 1:10 dilutions in water. Again, the track dilution technique was used to enumerate CFU of B. anthracis. Spore dilutions were heated to 70^°C for 30 min, after which aliquots again were plated. The percentage of heat-resistant spores was calculated by enumerating the CFU following an overnight incubation, as described above. The average and standard deviation were calculated from three independent experiments performed in triplicate.

Genetic construction of a germination-deficient mutant in B. anthracis (ΔgerHA).

Construction of a ΔgerHA strain of B. anthracis has been previously described by Weiner and Hanna (26), and a similar mutant was generated in the current study. Overlap extension PCR was used to construct a DNA fragment consisting of apha-3 flanked by regions of DNA that were homologous to upstream and downstream regions of the gerHA gene. The apha-3 gene, which provides antibiotic resistance to kanamycin (25), was amplified using PCR from the Ωkm-2 cassette as the template DNA and the oligonucleotides 5′-TGCTCTAGAGAAGAGGATGAGGAGGCAGATTGCC-3′ and 5′-TGCTCTAGAGCTCGGGACCCCTATCTAGCGA-3′. A 1,500-bp region upstream of gerHA and a 1,500-bp region downstream of gerHA were PCR amplified using B. anthracis Sterne 7702 genomic DNA and the primer pair 5′-TGAATATGGAAATTCCATTTGTG-3′ and 5′-TCCTCATCCTCTTCTCTAGAGCATGCACTTCACCTACTTTCGTTTG-3′ (primer 1) or the pair 5′-TAGGGGTCCCGAGCTCTAGAGCATACATTCCACTTTCTTTCTAGAG-3′ and 5′-TGTGGATTTAATACAACTCCTGC-3′ (primer 6), respectively. Using primer 1 and primer 6 and the three aforementioned PCR products, a DNA fragment was generated using a PCR technique that combined all three products (9). A 4.2-kb DNA fragment produced by the overlap extension PCR was cloned into pGEM-T Easy (Promega) and subsequently subcloned into pUTE568 (a generous gift from Theresa Koehler), yielding pUTE568::ΔgerHA(Kan^r) (20). E. coli strain JM110 (Δdam Δdcm), was transformed with the pUTE568::ΔgerHA(Kan^r) to produce unmethylated DNA, which was electroporated into B. anthracis Sterne 7702. Transformants were selected on solid medium containing kanamycin (100 μg/ml) and chloramphenicol (10 μg/ml). Isolates in which a double crossover event had occurred were selected by serial passages in antibiotic-free medium, and clones exhibiting a Kan^r Cm^s phenotype were identified. The deletion mutant was confirmed by PCR using oligonucleotides specific for apha-3 and a region upstream of gerHA and were propagated in medium containing 100 μg/ml kanamycin.

Animals.

Mice were purchased and maintained as specific pathogen free. Female DBA/2 mice were obtained from Harlan (Indianapolis, IN) and were 8 to 9 weeks old at the time of challenge. The mice were housed five mice per cage in the vivarium located at the University of New Mexico. Mice had access to irradiated food and autoclaved, acidified (pH 2.6) water ad libitum. Animals were allowed at least 7 days to acclimate before being used in this study. All protocols were approved by the University of New Mexico Health Sciences Center Institutional Animal Care and Use Committee.

Preparation of B. anthracis spores for intratracheal infection.

B. anthracis spores (Sterne strain 7702) were thawed to room temperature and then diluted in either PBS alone, PBS containing 0.5 mM l-alanine plus 1 mM inosine, or PBS containing 1 M d-alanine. Two independent experiments were performed, one in which spores were prepared at the University of New Mexico laboratory in a similar manner as previously described (14) and another using spores prepared at The University of Oklahoma Health Sciences Center laboratory in order to account for any variation between spore preparations. As indicated in one experiment, spores were heat treated at 70°C for 20 min and compared to similar non-heat-treated spore preparations used in the animal studies. For each condition the inoculum was enumerated by plating serial dilutions on sheep blood agar plates.

Injections and inoculations.

For pulmonary challenge, the mice were inoculated intratracheally as previously described (15). Briefly, the mice were anesthetized, a small incision was made in the skin, and the trachea was exposed. The inoculum was then injected into the trachea in a volume of 50 μl using a bent 30-gauge needle inserted into and parallel to the trachea. The actual number of spores deposited in the lung was determined by sacrificing 2 mice/dose/group following infection, homogenizing the lungs, and culturing serial dilutions on sheep blood agar plates using a spiral plater (Spiral Biotech, Bethesda, MD), and the number of CFU was counted using a plate scanner (Spiral Biotech). Survival and clinical signs were evaluated twice daily for the duration of the experiment.

Statistical analysis.

The statistical module of Excel was used to perform a Student's t test on the appropriate data sets. The 50% lethal doses (LD_50s) were calculated according to the method of Reed and Muench (18). Further statistical comparisons between control and experimental groups in the mouse studies were performed using the log rank test integrated into the Kaplan and Meier test of GraphPad Prism (version 4) software.

RESULTS

Germination of B. anthracis spores results in accumulation of inhibitors of germination.

In the first series of experiments, we sought to determine if soluble antigermination factors are released during B. anthracis germination. Germination of B. anthracis spores was determined by monitoring changes in OD over a 30-min time frame. A time-dependent decrease in OD reflects a lower refractive index of the spores in culture, which is a result of germination (17). Figure 1 summarizes a typical profile of changes in OD during germination of B. anthracis spores under various known germination conditions. As shown in Fig. 1, no change in OD was detected when spores were suspended in PBS, a condition not known to promote germination. In contrast, and in line with previous reports (11), the addition of l-alanine-inosine (Fig. 1A) or l-serine-inosine (Fig. 1B) triggered a rapid decrease in optical density, indicative of germination.

FIG. 1.

Production of endogenous inhibitors of germination by B. anthracis. Experiments were performed to confirm the production of germination inhibitors by B. anthracis. Supernatants were collected from germination and subsequently added to fresh spores with (+) or without (−) supplementation with fresh germinants. (A) The impact of l-alanine-inosine on the production of germination inhibitors. Spores were incubated with PBS, 0.5 mM l-alanine-1 mM inosine, GIM_l-Ala/Ino (GIM), and GIM_l-Ala/Ino supplemented with 0.5 mM l-alanine-1 mM inosine (GIM +), and germination was assessed by changes in the OD of the culture. (B) The influence of l-serine-inosine on germination and the production of germination inhibitors. Spores were incubated with PBS, 1 mM serine-1 mM inosine, GIM_l-Ser/Ino (GIM), and GIM_l-Ser/Ino supplemented with 1 mM l-serine-1 mM inosine (GIM+), and germination was assessed by changes in the OD of the culture. The OD₆₀₀ was monitored for 30 min at 37°C using a BioScreen C microplate reader. Results are the average of three experiments, and standard deviations did not exceed 2% from the mean.

Next, supernatants were collected from germinations that occurred under the l-alanine-inosine or l-serine-inosine conditions. Fresh B. anthracis spores were tested for germination in each of these supernatants, again by following changes in OD over a 30-min time period. As shown in Fig. 1B, a small decrease in OD was observed in spores incubated with the supernatant collected from the l-serine-inosine germination. In contrast, supernatants collected from the l-alanine-inosine germination did not cause a detectable change in the OD of the culture when added to fresh B. anthracis spores (Fig. 1A). Supernatants collected from germination medium following incubation with spores are annotated as GIM_l-Ser/Ino and GIM_l-Ala/Ino for GIM containing l-serine-inosine and GIM containing l-alanine-inosine, respectively.

Experiments were next performed to determine if the effect of GIM was due to depletion of germinants or, conversely, production of metabolites that inhibit germination. To differentiate between these two explanations, GIM was supplemented with fresh germinants equivalent to concentrations present in the original incubations. Fresh spores were then added to the supplemented GIM and monitored for changes in OD. As shown in Fig. 1A, even after supplementing GIM_l-Ala/Ino, (Fig. 1A, GIM+) the change in absorbance did not reach that found under non-GIM conditions (i.e., medium that was not collected from a previous germination). In contrast, GIM_l-Ser/Ino supplemented with fresh germinants (Fig. 1B, GIM+) supported a change in OD similar to that observed in the original incubations. Because germination efficiencies can vary among different batches of spores, similar experiments were performed on three other independent preparations of spores, and each was evaluated for GIM-related effects. Regardless of the spore preparation, a consistent profile of inhibitory effects similar to that shown in Fig. 1A and B was detected (data not shown). Overall, these results supported the idea that germination inhibitors accumulate in the supernatants of spores incubated with l-alanine-inosine but not in the presence of l-serine-inosine.

Exogenous l-alanine is required for d-alanine synthesis and production of GIM.

Results from the first experiments support the idea that l-alanine is critical to the production of antigerminants during germination. A reasonable explanation for this observation is that l-alanine is converted to d-alanine by an endogenous racemase, and this inhibits further germination in the spore population. Such a process has previously been characterized during studies in other Bacillus species (4) and has been suggested for B. anthracis (23). Therefore, candidate GIMs collected from germinations in l-alanine-inosine, l-serine-inosine, or l-phenylalanine-inosine were analyzed for levels of d-alanine. Concentrations of d-alanine were approximately 15-fold higher in GIM_l-Ala/Ino (199 ± 6 μM) than concentrations detected in GIM_l-Ser/Ino or GIM_l-Phe/Ino (13 ± 5 μM). These results indicated that l-alanine is necessary to generate d-alanine, but whether this concentration of d-alanine (199 μM) was sufficient to inhibit germination was unclear. Previous work by Ireland and Hanna found that d-alanine inhibited l-alanine-mediated germination at a 1:1 ratio (11). Yet our results suggested that a low concentration of d-alanine, relative to l-alanine, might also prevent germination or, alternatively, that factors other than d-alanine are involved in inhibition of germination. To address this issue, we examined the inhibitory of effects of a range of d-alanine concentrations (0.1 mM to 0.01 mM) below the concentration of l-alanine (fixed at 0.5 mM). As shown in Fig. 2, a concentration of d-alanine up to 10-fold less than that of l-alanine was sufficient to block germination of B. anthracis spores. Collectively, these findings suggested that the levels of d-alanine detected in GIM are sufficient to inhibit germination.

FIG. 2.

Concentration-dependent effects of d-alanine on germination of B. anthracis spores. Spores of B. anthracis were incubated in PBS alone or in 0.5 mM l-alanine-1 mM inosine, plus the indicated concentrations of d-alanine. Germination was assessed by changes in the OD₆₀₀ monitored for 30 min. Results are the average of three experiments, and standard deviations did not exceed 2% from the mean.

B. anthracis spore germination is inhibited in a density-dependent manner.

The data from the experiments shown in Fig. 1 and 2 indicated that B. anthracis releases a soluble factor, primarily d-alanine, which prevents further germination of fresh spores under static conditions. However, the characteristics of this event under more dynamic conditions, such as those actively occurring during germination, were unknown. To begin to assess d-alanine-mediated endogenous inhibition of germination as a dynamic event, the effects of spore density and the contribution of alanine racemase to this process were assessed in real time. As shown in the Fig. 3, at spore densities greater than 7.5 × 10⁷ spores, there was a dramatic reduction in germination, which was not detected when spore density was equal to or below 3 × 10⁷. One possible explanation for this result is that germinant concentrations are insufficient to support germination at high spore densities. Alternatively, germination may be inhibited by the presence of d-alanine, in line with the results described in the previous sections. To differentiate between these two possibilities, DCS was added to spores in the presence of germinants to repress the racemase-mediated conversion of l-alanine to d-alanine. As shown in Fig. 3, addition of DCS to the germination reversed the density-dependent effects, leading to a rapid increase in germination regardless of spore density. Controls of spore samples incubated with DCS alone had no impact on germination.

FIG. 3.

The influence of culture density and alanine racemase activity on B. anthracis spore germination. B. anthracis spores were prepared and analyzed for differences in germination efficiency across a range of spore densities in a fixed volume. (A to D) Germination profiles at four spore densities were determined by changes in the OD₆₀₀ values across a 30-min time period. DCS was added to germinations to account for the endogenous production of d-alanine and autoinhibition of germination under each of the culture conditions. Controls of DCS or PBS alone were included for each of the germination profiles. The spore number per 200 μl is indicated within each corresponding panel. (E) Bar graph of calculated maximal rates of germination from each of the spore densities and the change in rates corresponding to the presence of DCS. Results are the average of three independent spore preparations examined in triplicate.

These data corresponded to the results described by Titball and Manchee (23) but covered a more narrow range of spore densities. Because of this we were able to more closely examine the efficiency of germination as it varies along with small changes in spore density. The data from experiments performed under these conditions suggest that a critical threshold exists for autoinhibition of germination. Comparison between the twofold change in spore concentrations at 1.5 × 10⁸ spores and 3.0 × 10⁸ spores revealed only negligible differences in germination efficiency (Fig. 3, compare A and B). However, comparisons between 1.5 × 10⁸ spores and 7.5 × 10⁷ spores, a twofold difference, revealed a dramatic change in germination efficiency (Fig. 3, compare C and D). These results suggested that the correlation between germination rates and spore density is nonlinear. A first approximation of the rates of initiation of germination was performed by identifying the optimal change in OD from 1-min intervals under each condition (ΔOD₆₀₀/min). To account for possible variations due to spore preparation, the mean ΔOD₆₀₀/min was determined from three independent spore preparations, each analyzed in triplicate. As shown in the Fig. 3E bar graph, the changes in germination efficiency relative to spore density were neither incremental nor linear, further suggesting a possible enzymatic threshold associated with this event. A similar analysis of germination efficiency in the presence of DCS (Fig. 3E) did not reveal a threshold effect, supporting the notion that endogenous inhibition of germination is due to racemase-mediated production of d-alanine.

Production of inhibitory amounts of d-alanine does not require germination.

Interestingly, results shown in Fig. 3 indicated that germination was repressed at high spore numbers without a detectable change in the OD of the culture. This observation supports a model where d-alanine-mediated inhibition of germination could occur in the absence of germination. Indeed, such a process has been described for B. subtilis (29). Furthermore, results from the DCS experiments suggest that germination efficiency is modulated by alanine racemase, which would not necessarily depend on germination. Hence, experiments were next performed to determine whether racemase-mediated conversion of l-alanine to d-alanine requires germination or if this particular racemase activity occurs in spores.

First, production of d-alanine was assessed under conditions that did not favor germination. l-Alanine in the absence of inosine, which has previously been shown not to support germination (11), was selected for this purpose. Second, production of d-alanine was assessed in spores that had been modified by genetic disruption of genes known to be critical for germination. Both experiments allowed an assessment of the production of d-alanine in the absence of germination. Despite the absence of germination with incubations of l-alanine, levels of d-alanine similar to those found under germinating conditions were detected. Although germination levels were less than 10% in l-alanine and greater than 90% in l-alanine-inosine, both conditions yielded d-alanine levels of approximately 200 μM. In the absence of l-alanine, amounts of d-alanine were below levels of detection. These results suggested that germination is not an absolute requirement for d-alanine production in B. anthracis.

Further evidence supporting the notion that germination is not required for d-alanine production was acquired using a strain of B. anthracis defective in germination. In a manner previously described by Weiner et al. (27), the first gene (gerHA) of the gerH operon was disrupted, yielding a germination-defective mutant, ΔgerHA. Similar to the results from treatment with l-alanine alone, germination efficiency was less than 10% in B. anthracis ΔgerHA, yet detectable d-alanine levels were found to be on average 166 μΜ when spores of this mutant were incubated in l-alanine-inosine. GIM collected using ΔgerHA spores (GIM_ΔgerHA) was assayed for germination-inhibiting activity in freshly prepared spores of B. anthracis Sterne supplemented with l-alanine-inosine. As shown in Fig. 4, germination of spores of B. anthracis in supplemented GIM_ΔgerHA (GIM_{ΔgerHA/l-Ala/Ino+} was reduced (final OD₆₀₀ of 0.319) from that of germination in l-alanine-inosine medium alone (final OD₆₀₀ of 0.212). Collectively, these data suggest that racemase conversion of l-alanine to d-alanine does not require germination of B. anthracis spores and is sufficient to generate inhibitory levels of d-alanine.

FIG. 4.

Production of GIM does not require germination of B. anthracis spores. The decrease in the OD₆₀₀ was used to follow B. anthracis spore germination in GIM established under germinating and nongerminating conditions. As controls, B. anthracis Sterne spores were incubated in 0.5 mM l-alanine-1 mM inosine (l-Ala/Ino) or 0.5 mM l-alanine (l-Ala), which are germinating and nongerminating conditions, respectively. GIM was collected from B. anthracis Sterne spore germinations in the presence of 0.5 mM l-alanine-1 mM inosine and then supplemented with 0.5 mM l-alanine-1 mM inosine (GIM_l-Ala/Ino+); alternatively, GIM was collected from B. anthracis Sterne spores incubated in only 0.5 mM l-alanine and then supplemented with 0.5 mM l-alanine-1 mM inosine (GIM_l-Ala+), or GIM was collected from the ΔgerHA mutant incubated in 0.5 mM l-alanine-1 mM inosine and then supplemented with 0.5 mM l-alanine-1 mM inosine (GIM_{ΔgerHA/l-Ala/Ino+}). Results are the average of three independent experiments, and standard deviations did not exceed 2% from the mean.

Endogenous d-alanine production alters germination of B. anthracis during interactions with cultured murine macrophages.

The results described above indicate that populations of B. anthracis spores are subject to autoinhibition of germination in BHI broth and defined medium. However, whether autoinhibition of germination can influence steps in pathogenesis is not known. Interestingly Hu et al. recently found that addition of d-alanine to B. anthracis spores increased survival of the organism in primary murine macrophages (10). This observation suggested that d-alanine, generated endogenously by B. anthracis, could alter spore-macrophage interactions. To address this issue experimentally, cultured murine macrophages were incubated with increasing amounts of spores, and the percent germination was determined for each condition. As shown in Fig. 5, as the MOI increased, there was a corresponding increase in the percentage of spores that remained resistant to heat inactivation, indicative of spores that had not germinated.

FIG. 5.

The Impact of racemase inhibition on spore-macrophage interactions. Spores were incubated across a range of MOIs from 1 to 25 with murine macrophages in the absence (solid bars) or presence (hatch bars) of 10 mM DCS. Following a 1-h incubation, spores attached to or taken up by cells were collected and enumerated as CFU prior to and following a 70^°C incubation for 30 min. Data are presented as percent heat-resistant CFU, which is representative of the percentage of ungerminated spores. The raw numbers for the mean values at MOIs of 1, 5, 10, and 25 were 2.83 × 10³, 2.63 × 10⁴, 4.97 × 10⁴, and 1.77 × 10⁵ with DCS, respectively, and 4.4 × 10³, 1.5 × 10⁵, 4.1 × 10⁵, and 1.93 × 10⁶ without DCS, respectively. Experiments were performed in triplicate, and error bars represent the standard deviation from the mean.

It was possible that the decrease in percent germination was due to limited germinants in the medium or within the murine macrophage and was not due to the endogenous production of d-alanine. In order to confirm that depletion of germinants was not responsible for the observed results, identical experiments were performed in the presence of DCS. Pilot experiments found that DCS reduced the efficiency of murine macrophage binding or uptake of B. anthracis spores by approximately 2- to 2.5-fold, making absolute comparisons between results from medium with DCS and medium without DCS impossible. However, when the data are compared in terms of total heat resistance as a percentage of the population on or within the macrophage, the contribution of racemase activity to this interaction becomes apparent. Unlike the DCS-negative condition, addition of DCS to the murine macrophage infection at increasingly higher MOIs significantly reduced the percentage of CFU exhibiting heat resistance. Hence, similar to the findings from germination in rich medium, inhibiting racemase activity results in a marked increase in germination during interactions with murine macrophages (Fig. 5). These results indicate that autoinhibition of germination could influence the number of spores that survive encounters with macrophages during early stages of anthrax disease.

Alanine stereoisomers influence the outcome of B. anthracis infection.

In toto, the results shown in Fig. 1 to 5 indicate that levels of racemase-generated d-alanine alter the germination efficiency of B. anthracis spores. We predicted that this event might be one of several important processes that lead to successful establishment of infection by B. anthracis because slowing the rate of germination might allow the organism to localize at an optimal anatomical site without destruction by the macrophage. To experimentally address this issue, mice were infected by intratracheal administration with spores of B. anthracis in the presence of PBS, l-alanine-inosine, or d-alanine, and the LD₅₀ was determined in two independent experiments. Importantly, each experiment involved distinct batches of spores prepared in two different laboratories in order to ensure that the findings were not due to anomalies unique to a particular preparation of spores. As shown in Fig. 6, addition of l-alanine-inosine to the spores just prior to infection increased the LD₅₀ of B. anthracis relative to spores administered in PBS alone. Although the trend of an increased LD₅₀ was consistently observed in the presence of l-alanine, the resulting P values were 0.1464 and 0.1327 as calculated by a log rank test analysis of the data. (The calculated LD₅₀ is inset in each of the panels of Fig. 6.) In contrast, addition of d-alanine, which was predicted to slow germination, reduced the LD₅₀ of B. anthracis by approximately twofold relative to spores administered in PBS alone, and these differences were found to be of statistical significance (P values of 0.0012 and 0.0294). The changes in the LD₅₀s ranged from two- to threefold, suggesting a subtle yet reproducible effect of the germinants on establishment of anthrax disease.

FIG. 6.

The impact of germinants and antigerminants on the survival of mice infected with B. anthracis. Spores of B. anthracis were prepared and mixed with either PBS, l-alanine-inosine, or d-alanine just prior to administering the spores to mice by intratracheal injection. As indicated in the panels, spores of three densities (in the range of 10⁴ to 10⁶) were used in these experiments, and mice were infected 10 per group. The LD₅₀ was calculated from the survival profiles under each of the experimental conditions. (A to C) Survival profiles of mice infected with B. anthracis Sterne spores under germination or antigermination conditions. Each condition and the corresponding infectious unit are indicated along with the calculated LD₅₀. (D to F) Replicate experiment using B. anthracis Sterne spores prepared from a different spore stock at a separate laboratory. Log rank test of the data at 1 × 10⁴ spores from experiments one and two revealed P values of 0.1464 and 0.1327, respectively, between PBS and l-alanine and P values of 0.0012 and 0.0294, respectively, between PBS and d-alanine.

DISCUSSION

Our work reveals autoinhibition of B. anthracis spore germination by a density-dependent mechanism involving spore-mediated stereoisomerization of l-alanine to d-alanine. Moreover, our results demonstrate the impact of d-alanine on B. anthracis interactions with RAW 264.7 murine macrophages, as well as the progression of anthrax disease in vivo.

In previous work Titball and Manchee reported that spores maintained at a density of 1.4 × 10⁸ spores/ml germinated less efficiently than spores at 1.4 × 10⁵/ml (23). We extended this observation in a more highly defined system by demonstrating that while 0.5 mM l-alanine-1 mM inosine promotes B. anthracis germination at densities of less than 3.0 × 10⁷ spores/ml, germination was inhibited at densities greater than 3.0 × 10⁷ spores/ml and was completely blocked 3.0 × 10⁸ spores/ml (Fig. 3). These results indicate that B. anthracis germination induced by l-alanine-inosine is relatively insensitive to increases in spore density until a threshold density is reached. When spore density reaches this threshold, germination efficiency becomes highly sensitive to small changes in spore number, resembling an enzymatic mechanism associated with autoinhibition of germination. In line with this, Titball and Manchee reported that the addition of a racemase inhibitor improved the germination efficiency of their denser spore preparations (23), suggesting that racemase-mediated conversion of l-alanine to d-alanine is a key step underlying the mechanism of density-dependent autoinhibition. Consistent with these earlier findings (23), we found that the racemase inhibitor DCS prevents autoinhibition of germination (Fig. 3), further supporting the idea that stereoisomerization of l-alanine into d-alanine is responsible for the inhibition of germination. We validated this idea by demonstrating that d-alanine alone is capable of inhibiting germination, even at 10-fold lower concentrations than l-alanine within the medium (Fig. 2). These experiments established that B. anthracis spore germination is sensitive to density-dependent autoinhibition by a mechanism involving the l- and d-enantiomers of alanine.

The contribution of alanine racemase activity to the autoinhibition of B. anthracis germination is apparent from the effectiveness of DCS (a racemase inhibitor) at reducing inhibition normally observed at higher spore densities. Notably, one of the two B. anthracis alanine racemases, Alr, is an exosporium protein (24) that would presumably, even for dormant spores, convert available l-alanine substrate to the inhibitory d-alanine. Our finding that GIM collected from the germination-deficient (gerHA) B. anthracis mutant is sufficient to inhibit germination of wild-type spores further supports the notion that germination is not a prerequisite for the production d-alanine (Fig. 4). The second B. anthracis racemase, Dal, is expressed primarily during vegetative growth and would be more likely involved in the generation of peptidoglycan (21). Alr is conserved in other closely related species such as Bacillus cereus and Bacillus thuringiensis, and recent studies indicated that when l-alanine was provided as a germinant, germination of a B. thuringiensis mutant strain lacking Alr increased relative to the wild-type parent strain, suggesting that the presence of exosporium alanine racemase was inhibitory for germination (28). While further work will be required to understand the relative contributions of Alr or Dal to density-dependent autoinhibition of germination, it is likely that the exosporium-localized Alr may more readily contribute to the conversion of exogenous l-alanine to the d-enantiomer in the presence of dormant spores.

While the mechanism underlying density-dependent autoinhibition of B. anthracis germination is poorly understood, an important factor may be the role of l-alanine as both a germinant and a substrate for alanine racemase. Notably, Yasuda et al. reported that, for B. subtilis, the germination receptor exhibits 1,000-fold higher affinity than alanine racemase for l-alanine (29), suggesting that, at least for this organism, l-alanine may preferentially bind receptors as a germinant rather than be consumed as a racemase substrate. However, germination receptors demonstrated a 10-fold higher affinity for d-alanine than l-alanine, indicating that low concentrations of d-alanine may be sufficient, at least partially, to inhibit germination, even in the presence of the l-enantiomer. These observations point to several competing processes—including germination receptor binding and racemization—that likely influence the extent to which spores germinate in the presence of l-alanine. We propose that autoinhibition of germination occurs when racemase activity is sufficient to raise d-alanine concentrations to inhibitory levels. Notably, high racemase activity would not only raise concentrations d-alanine but at the same time would also lower the concentration of available germinant l-alanine. In the context of germination, the ratio of available l-alanine to d-alanine thus appears to function as a rheostat, which is adjusted by the number of spores present in a fixed volume.

The importance or role of autoinhibition of germination in the establishment of B. anthracis infection has not been previously studied. The current model of inhalational anthrax predicts that B. anthracis may exploit uptake of spores by macrophages as one way to egress from the lungs (8). Yet while spores are resistant to macrophage killing, germination renders B. anthracis susceptible to macrophage-mediated destruction (10). Our results suggest that autoinhibition of B. anthracis germination may occur during macrophage interactions. The relative number of germinated B. anthracis spores decreased with increasing MOIs (Fig. 5), suggesting a relationship between germination and the density of spores within the infection. A role for alanine racemase in this phenomenon was supported by the capacity of the racemase inhibitor to diminish the relationship between MOI and spore germination (Fig. 5). It is important to note that, as these experiments are designed, we cannot draw conclusions about whether autoinhibition of germination occurs primarily outside or inside the macrophage. It is also possible that inhibited spores adhere differently to macrophages or may also be trafficked into a compartment that prevents germination. For this reason a broad view of these data is still necessary, and future experiments will be important for specifically defining the influence d-alanine production has on the overall spore-macrophage interaction. Most importantly at this point, it appears that alanine racemase-dependent autoinhibition of germination influences spore-macrophage interactions.

This work provides the first direct evidence that the presence of defined germinants (l-alanine-inosine) reduces B. anthracis-mediated lethality in an in vivo intratracheal murine model of inhalational anthrax. Interestingly, Rhian et al. reported 40 years ago that B. anthracis virulence increased when spores were preincubated with egg yolk agar prior to intraperitoneal administration to rodents (19). In contrast, aerosol administration of spores in the presence of egg yolk agar led to a slower time course to death and an overall reduction in the virulence of B. anthracis. Although little was known at the time about the effects of egg yolk agar on germination, the authors exhibited considerable insights by suggesting that the increased B. anthracis virulence following intraperitoneal delivery was likely due to the fact that germination occurred rapidly and allowed the organism to grow systemically while, for spores delivered as aerosols, rapid premature germination may have resulted in increased killing of B. anthracis by lung phagocytes.

While the current data provide fundamental insight into the role of d-alanine and the timing of germination during early steps in anthrax disease, these findings also raise concerns about inhibition of germination as a therapy for inhalational anthrax. Indeed, our data showing that d-alanine reduces the LD₅₀ of B. anthracis are counterintuitive to the idea that disrupting germination benefits the host. Recently, Akoachere et al. described the capacity of 6-thio-guanosine (6-TG), an inosine analog, to prevent germination of B. anthracis spores in cultured macrophages (1). In their study, 6-TG blocked germination for up to 7 h after spores were engulfed by cultured macrophages. Based on this observation, the authors posit that germination inhibitors such as 6-TG could be useful therapies for inhalational anthrax because of their capacity to attenuate a critical step in establishment of disease. Although there may be merit to this idea, our results suggest that such a treatment approach alone could have unfortunate consequences and should be considered in the fullest context of disease. It is reasonable to suggest that slowing germination protects spores from macrophage destruction and exacerbates instead of alleviating disease. Clearly, more detailed in vivo studies analyzing the outcome of treatment with germination inhibitors on inhalational anthrax are needed to determine if compounds such 6-TG have antagonistic or agonistic effects on disease.

In summary, we have presented evidence for autoinhibition of B. anthracis spore germination and have described one mechanism by which autoinhibition may occur. Our results suggest that autoregulation of spore germination may be important for B. anthracis virulence. An increased understanding of the biological processes that regulate the timing of germination may ultimately lead to a more comprehensive model of the initial stages of inhalational anthrax.