APPLIED AND ENVIRONMENTAL MICROBIOLOGY, May 2010, p. 3275–3282 0099-2240/10/$12.00 doi:10.1128/AEM.02485-09 Copyright © 2010, American Society for Microbiology. All Rights Reserved.
Vol. 76, No. 10
Spatially Resolved Characterization of Water and Ion Incorporation in Bacillus Spores䌤 Sutapa Ghosal,1 Terrance J. Leighton,2 Katherine E. Wheeler,2 Ian D. Hutcheon,1 and Peter K. Weber1* Chemical Sciences Division, Lawrence Livermore National Laboratory, Livermore, California 94550,1 and Children’s Hospital Oakland Research Institute, Oakland, California 946092 Received 13 October 2009/Accepted 18 March 2010
We present the first direct visualization and quantification of water and ion uptake into the core of individual dormant Bacillus thuringiensis subsp. israelensis (B. thuringiensis subsp. israelensis) endospores. Isotopic and elemental gradients in the B. thuringiensis subsp. israelensis spores show the permeation and incorporation of deuterium in deuterated water (D2O) and solvated ions throughout individual spores, including the spore core. Under hydrated conditions, incorporation into a spore occurs on a time scale of minutes, with subsequent uptake of the permeating species continuing over a period of days. The distribution of available adsorption sites is shown to vary with the permeating species. Adsorption sites for Liⴙ, Csⴙ, and Clⴚ are more abundant within the spore outer structures (exosporium, coat, and cortex) relative to the core, while Fⴚ adsorption sites are more abundant in the core. The results presented here demonstrate that elemental abundance and distribution in dormant spores are influenced by the ambient environment. As such, this study highlights the importance of understanding how microbial elemental and isotopic signatures can be altered postproduction, including during sample preparation for analysis, and therefore, this study is immediately relevant to the use of elemental and isotopic markers in environmental microbiology and microbial forensics. and subsequent studies led to the dismissal of the impermeability hypothesis, although structural limits on permeability, particularly into the spore core, remain an active area of interest (14, 52). Water uptake properties of Bacillus spores have been examined by various laboratories (6, 28, 29, 34, 38, 40, 45, 52, 54). The general picture to emerge from this collective work is that the entire spore volume, including the core, is permeable to water. However, in these previous studies, uptake of water into the spore core was not observed directly; rather, the permeability of spores to water was inferred from measurements of correlated, but indirect, phenomena such as the change in size of a spore in response to water adsorption. The distribution of the permeating species within individual spores after an exposure event has received less attention, primarily because of the technical challenge posed by direct interrogation of the spore interior. The permeation studies cited above used bulk methods or light/force microscopy methods to infer distribution of the permeating species. Scanning transmission electron microscopy (STEM) has the necessary spatial resolution to interrogate the spore interior (10, 49, 50) but is limited by sensitivity and sample preparation requirements and has not been used for such studies. We have applied imaging mass spectrometry with a NanoSIMS, a high-resolution secondary ion mass spectrometer (27, 33, 48), to directly probe the spatial distribution of permeating water vapor and ions in individual Bacillus thuringiensis subsp. israelensis (B. thuringiensis subsp. israelensis) spores following exposure and dehydration treatments. We used an imaging depth profile method, described in detail elsewhere, to characterize elemental distributions within individual spores (Fig. 1) (20). The results reported here confirm that the entire spore volume, including the core of a dormant B. thuringiensis subsp.
In recent years, the physical and chemical characteristics of spores have received increasing attention, particularly in response to the anthrax letter attacks in fall 2001 (4, 8, 11, 12, 37). A number of studies have suggested that nongenetic identifiers, such as elemental and isotopic signatures, can potentially be used to identify the means, geographic location, and approximate date of production of biological agents (13, 21, 25, 26, 41, 55). Such nongenetic markers can also extend the utility of bacterial spores as environmental tracers to study the fate and transport of biological materials in natural environments (23, 24, 46). In order to establish the utility of elemental and isotopic signatures for microbial forensics and environmental microbiology, it is necessary to know if and how the elemental and isotopic compositions of bacterial spores can be altered during and after production. This knowledge is essential for understanding the potential sources and robustness of a signature as well as the best way to extract it. Permeation by water and solvated ions is a candidate pathway for alteration of the concentration and distribution of potential signature species in bacterial spores. The permeability of bacterial spores has been extensively studied because of interest in spore dormancy and resistance to bactericidal treatments (5–7, 14, 19, 29, 30, 34, 38, 42, 44, 45, 52, 54). Historically, spores were assumed to be impermeable. However, pioneering studies by Lewis et al. (29) and others (5–7, 19) demonstrated that dormant spores are permeable to water and that permeability to various solutes is controlled by their molecular weight, charge, and lipophilicity (19). These
* Corresponding author. Mailing address: 7000 East Avenue, L-231, Livermore, CA 94550. Phone: (925) 422-3018. Fax: (925) 422-3160. E-mail:
[email protected]. 䌤 Published ahead of print on 26 March 2010. 3275
3276
GHOSAL ET AL.
APPL. ENVIRON. MICROBIOL.
FIG. 1. Representative distribution of C, F, P, and Cl in a single B. thuringiensis subsp. israelensis endospore, as determined by NanoSIMS imaging depth profile analysis using a 133Cs⫹ primary ion beam. (A) The ion images obtained from serial scans of the spore by the Cs beam represent the summed ion counts for each species. Lighter colors represent higher counts. Scale bar ⫽ 500 nm. (B) The depth profile data are extracted from serial images of the spore as it is eroded away by the Cs beam. The data are extracted from the ⬃200-nm-diameter circle, which is centered on the P image. Depth is estimated from the lateral dimension of the spore and the sputter rate of similar samples.
israelensis spore, is permeable to water and hydrated ions. Based on these findings, we explored the permeation and incorporation of metal and halogen ions by dormant B. thuringiensis subsp. israelensis spores suspended in salt solutions. We discuss the implications of our results in terms of mechanisms of permeation and incorporation. MATERIALS AND METHODS Spore samples. Bacillus thuringiensis subsp. israelensis (BGSC 4Q1) was obtained from the Bacillus Genetics Stock Center. Spore preparation and purification were performed as described previously (31, 35). For solid-phase sporulation, 1.5% agar was added to Schaeffer’s sporulation medium (35), and the plates were incubated at 37°C until confluent sporulation was achieved. Spores were harvested in double-distilled water, washed six times by centrifugation (15,000 rpm for 30 min), and resuspended in double-distilled water. Spore samples were stored in double-distilled water or stored dry, after lyophilization. For quality control, aliquots of the treated spore samples were examined by phase-contrast microscopy to determine spore integrity (phase bright versus phase dark). Only preparations that contained ⬎95% native phase-bright spores were used for chemical analysis. Intact spores were mapped by scanning electron microscopy (SEM) for NanoSIMS analysis. As a second line of quality control, Ca or P levels in spore cores were determined during NanoSIMS analysis to ensure that the abundance of one of these elements was consistent with intact native spores (20). Cells and germinated or damaged spores have significantly lower Ca and P abundances than intact spores. Permeability experiments. For water and ion uptake experiments, B. thuringiensis subsp. israelensis spores were exposed for predetermined intervals to one of the following reagents: deuterated water (D2O) vapor, lithium fluoride (LiF) solution (0.4, 10, and 40 mM) in deionized D2O (pH ⬃4.7), or cesium chloride (CsCl) solution (2 mM) in deionized D2O. The solution concentrations were measured by inductively coupled plasma mass spectrometry (ICP-MS) at Lawrence Livermore National Laboratory (LLNL). Individual experiments are described in detail below. Because the spore exosporium, coat, and cortex could not clearly be distinguished from one another in the NanoSIMS images, we refer to these three structures collectively as the “outer structures.” The data presented here are average results of multiple measurements of individual spores within a given population. Uptake of water. Water uptake was studied by exposing dry spores to saturated D2O vapor. The use of deuterium-labeled water enabled the permeating water to be distinguished from intrinsic water associated with the spores. One set of samples was exposed to D2O for times ranging from 1 min to 15 days in sealed glass containers with Teflon septa containing 10 l of D2O. Another sample was placed in a Parr bomb, with 10 l of D2O at 80°C for 2 days to simulate longer exposure conditions. The D2O exposures were staggered so that all samples were harvested within 15 min of each other, with the shortest duration exposures harvested last. For each treatment, dry B. thuringiensis subsp. israelensis spores deposited on 5- by 5-mm Si wafers were used. After being harvested, the samples were placed in a single sample holder which was introduced directly into the Nano-
SIMS air lock, which was evacuated to ⬍1 ⫻ 10⫺3 torr in ⬃1 min and to ⬍1 ⫻ 10⫺5 torr in ⬃5 min. A control sample was placed in the air lock with the other D2Oexposed samples (⬍1 min of exposure time). The samples were transferred to an evacuated holding chamber after being pumped down to a pressure of ⬃1 ⫻ 10⫺7 torr. After being pumped down to a pressure of ⬃1 ⫻ 10⫺9 torr in the holding chamber, the samples were subsequently moved to the analysis chamber where the pressure was ⬍1 ⫻ 10⫺9 torr. A second control sample was loaded into the holding chamber separately. Samples were analyzed on the day of harvesting and after 3 months of residence in the vacuum chamber. Ion exposure experiments. Cation and anion uptake was studied by exposing B. thuringiensis subsp. israelensis spores to a LiF or CsCl solution in D2O for exposure times ranging from 5 min to 4 days. The choice of cations was based on the observation that both Li and Cs have low abundances in native spores and also provide a range of ionic radii. Following exposure to either a LiF or CsCl solution, spores were washed several times in deionized water to remove residual solute. In the initial ion exposure experiments, small quantities of dry B. thuringiensis subsp. israelensis spores were suspended in 1 ml of LiF solution (either 10 mM or 40 mM) at the start of the experiment. The LiF-spore mixture was vortexed to ensure uniform mixing. At the end of the exposure period, 100 l of the suspension was removed and washed three times consecutively using the following procedure. The suspension was first centrifuged (13,000 rpm for 2 min) to isolate the spores from the solution, followed by removal of the supernatant; the spores were then resuspended in D2O, and the entire process repeated. Finally, 1 l of the washed spore suspension was deposited onto a 5- by 5-mm Si wafer and air dried in a laminar flow hood. Prior to NanoSIMS analysis, the samples were imaged with an optical microscope and/or SEM to identify and map regions for analysis. Prepared samples were stored in dry boxes under argon. Prior to analysis, samples were introduced into NanoSIMS and evacuated to ⬍1 ⫻ 10⫺9 torr. In subsequent ion exposure experiments, hydrated B. thuringiensis subsp. israelensis spores were added to salt solutions to ensure that diffusion into the spores was the dominant mechanism of ion movement relative to advection during hydration. These treatments used 0.4 mM LiF and 2 mM CsCl solutions; exposure and washing procedures were similar to those used with the initial experiments. Anion leaching. To test the mobility of intrinsic F and Cl in B. thuringiensis subsp. israelensis spores, small aliquots (⬍100 g) of dry spores were placed in 1.5 ml of deionized water and sampled by removing ⬃1 l for deposition on a Si wafer for NanoSIMS analysis. Secondary ion mass spectrometry (SIMS) analysis. SIMS was performed using the Lawrence Livermore National Laboratory NanoSIMS 50 (Cameca Instruments, Gennevilliers, France) (48). NanoSIMS 50 is a double-focusing magnetic sector mass spectrometer configured to enable simultaneous detection of up to 5 species with pulse counting on electron multipliers. For the spore analyses, NanoSIMS was tuned for ⬃3,000 mass resolving power to resolve isobaric interferences. A 16 kV 133Cs⫹ primary ion beam, which enhances the yield of electronegative elements, was used to analyze samples for H⫺, D⫺, 12C⫺, 12 CH⫺, 12CD⫺, 19F⫺, 28Si⫺, 31P⫺, and 35Cl⫺. Sixteen-kilovolt 16O⫺ and 16O2⫺ primary ion beams, which enhance the yield of electropositive elements, were
VOL. 76, 2010
WATER AND ION INCORPORATION IN BACILLUS SPORES
3277
FIG. 2. Deuterium uptake with depth and time data obtained from B. thuringiensis subsp. israelensis spores exposed to D2O. (A) Distribution of deuterium (D) and hydrogen (H) as a function of depth in a single spore exposed to D2O. The depth profile data are normalized to those of C⫺. Error bars represent 2 standard errors and are not visible where smaller than data points. (B) Summed H⫺ and D⫺ images of a single spore. Color scales are linear and indicate counts per pixel. Scale bar ⫽ 500 nm. (C) Uptake and exchange of D2O by B. thuringiensis subsp. israelensis spores as a function of time of exposure to D2O vapor and when the analysis took place relative to the end of the exposure experiment (day 1 and 3 months, respectively). D⫺/H⫺ ratios are whole-spore averages. Open symbols represent data obtained with Parr bomb treatment (2 days at 80°C). Untreated spores yielded the natural abundance D/H ratio of ⬃0.00016.
used to analyze samples for 12C⫹, 7Li⫹, 23Na⫹, 28Si⫹, 40Ca⫹, and 133Cs⫹. C and Si were detected in both modes. Compared to the 16O⫺ beam, the 16O2⫺ primary ion beam has a shallower penetration depth and higher sputter rate and, therefore, provides better depth resolution. The primary ion beam was typically stepped over the sample in a 256- by 256-pixel raster to generate secondary ions and produce element specific, quantitative digital images that reflect the composition of the sample. The lateral spatial resolution of the NanoSIMS secondary ion images equaled the diameter of the primary beam. The 133Cs⫹ primary beam was focused to a nominal spot size of ⬃100 to 200 nm, with a current of 1 to 6 pA. In the case of the 16O⫺ and 16O2⫺ primary beams, the spot size was typically less than 0.5 m, and the current was 10 to 100 pA for 16O⫺ and ⬃1 pA 16O2⫺. The dwell time for ion imaging was 1 ms/pixel, and the raster area ranged from 6 to 100 m2. The depth resolution was ⬃10 to 20 nm for both 133Cs⫹ and 16O⫺ (20). NanoSIMS is an ultrahigh vacuum (UHV) instrument, and samples were evacuated under UHV conditions (⬍1 ⫻10⫺9 torr) in NanoSIMS prior to analysis, which likely removed the majority of unbound water. We measured elemental concentrations within whole spores by imaging depth profile analysis, which is presented in detail in reference 20. In brief, consecutive scans of the primary ion beam eroded away a spore, during which a series of secondary ion images were collected. Depth-resolved data were extracted from the central ⬃200-nm diameter region of each imaged spore to generate a depth profile. An advantage of this method is that it does not require elaborate sample preparation techniques such as embedding, fixation, or sectioning and, therefore, minimizes the likelihood of sample preparation-induced artifacts. Figure 1 shows the representative distributions of C, F, P, and Cl in a single B. thuringiensis subsp. israelensis spore, determined by NanoSIMS imaging depth profile analysis using a 133Cs⫹ primary ion beam. The depth scale was estimated from the lateral dimensions of the spore and the sputter rates of similar samples. The rise in the “top hat” C trace resulted from the onset of sputtering of the spore by the Cs beam, the flat region was where sputtering equilibrium was attained, and the decline with increasing depth resulted from the spore being sputtered away. The behavior of the other elements can be visualized by reference to the C trace. The rise and fall of the P intensity reflect the high concentration of P-bearing nucleotides in the spore core. The data were corrected for detector dead time and image-to-image shift and were processed as quantitative elemental ratio images using custom software (LIMAGE; L. R. Nittler, Carnegie Institution of Washington). The secondary ion intensities were normalized to 12C⫺ or 40Ca⫹. Silicate glass standard reference material, produced by the National Institute of Standards and Technology (NIST SRM 610), was used as a reference standard for the quantification of Li⫹/Ca⫹ and Cs⫹/Ca⫹ ion ratios.
RESULTS Uptake, exchange, and distribution of water in dormant B. thuringiensis subsp. israelensis spores. We directly imaged the spatial distribution of deuterium (D) incorporation in individual B. thuringiensis subsp. israelensis spores in response to D2O vapor exposure (Fig. 2). Figure 2A shows the distribution of D
and H as a function of depth in a B. thuringiensis subsp. israelensis spore. D was found throughout the spore. These measurements provide direct evidence of the incorporation of D associated with the D2O vapor into the spore core. We postulate that the observed distribution of D in the spore was the result of D2O permeation and exchange within the spore volume, including the core. Our observations are in agreement with Black and Gerhardt (6) and Leuschner and Lillford (28), who proposed that the core of a dormant spore exists as an insoluble and heat-stable gel permeable to ambient water. The abundance of D (expressed as the D⫺/C⫺ ratio) was ⬃30% higher in the core relative to the outer structures. Both D and H showed similar distributions within the spore, indicating that the relative abundance of D adsorption sites was proportional to the abundance of intrinsic H-containing species (e.g., H2O and OH⫺) in the spore. Incorporation of D into the spore was likely the result of exchange with intrinsic water and exchangeable protons in the spore. The rate of water permeation into the spore was not determined in this study because of the time lapse between water exposure and sample analysis in NanoSIMS. However, we did examine the level of D incorporation with increasing D2O exposure time. The abundance of D in the spore increased with increasing D2O exposure time, approaching the level of D associated with the sample treated in a Parr bomb; we took this value to be representative of the asymptotic concentration of D in spores (Fig. 2C). The time required to reach half of the asymptotic value for D uptake was approximately 1 h. This time is characteristic of the exchange mechanism and not of the rate of permeation. Analysis of the samples after 3 months in vacuo showed that the D/H ratio decreased across the full range of D2O exposures. A likely explanation for this result is that weakly bound or unbound D2O originating from the D2O exposure was progressively lost via dehydration or exchange with vacuum chamber H2O. The relative difference in the D/H ratio was maintained across exposure times, supporting our conclusion that longer time exposures lead to higher levels of exchange. Metal uptake and distribution in spores. The permeability of B. thuringiensis subsp. israelensis spores to water vapor suggests that hydration pathways or channels may provide a mech-
3278
GHOSAL ET AL.
APPL. ENVIRON. MICROBIOL.
FIG. 3. Li uptake into B. thuringiensis subsp. israelensis spores from 10 mM LiF solution. (A) Depth profiles of 7Li⫹ in control spores (no exposure) and spores exposed to LiF solution for 3 days, showing uptake of Li from solution throughout the spore, with greater uptake in the outer structures. Li content in the treated spores (measured as a 7Li⫹/12C⫹ ion ratio) increased by greater than 4 orders of magnitude relative to the control sample. (B, C) Summed 44Ca⫹ and 7Li⫹ NanoSIMS ion images of spores with no exposure (B) and 3 days of exposure to 10 mM LiF solution (C). The Ca images show the locations of the spores, and the Li images show the low background in the control sample (B) and that Li is localized in the spores in the exposed samples (C).
anism for diffusion of ions within spores due to the increased mobility of hydrated ions. Leuschner and Lillford (28) used nuclear magnetic resonance (NMR) spectroscopy to show that hydration enhances the molecular mobility of phosphorus and carbon in Bacillus subtilis spores. The permeability of spores to a variety of solutes has been examined previously by Gerhardt and coworkers (5, 6, 19). On the basis of bulk spore measurements, they reported that solutes with sufficiently high molecular masses (⬎8 kDa) were excluded from spores and that the permeation rates were affected by the solute charge state. We examined the permeability of dormant B. thuringiensis subsp. israelensis spores by using hydrated monovalent cations, namely, Li⫹ and Cs⫹, in deionized water (pH ⬃4.7). These particular cations were selected for the permeability experiments based on the observation that both Li and Cs have low abundances in native spores, and hence, their uptake by spores should be readily identifiable. In these experiments, spores were exposed to dilute aqueous solutions of either lithium fluoride (LiF) or cesium chloride (CsCl) salts, wherein both Li and Cs were expected to be present as hydrated ionic species. The permeation of Li⫹ in B. thuringiensis subsp. israelensis spores was examined with two different concentrations of LiF solutions, 10 and 0.4 mM. LiF treatment did not alter the percentage of phase-bright spores in the samples. In the initial exposure experiments, dry spores were placed in a 10 mM LiF solution to test the method. Exposure to a 10 mM LiF solution resulted in permeation of Li throughout the spore, showing that the core of a dormant B. thuringiensis subsp. israelensis spore was permeable not only to water but also to small ions such as Li⫹ in solution. Following 3 days of exposure to a 10 mM LiF solution, the Li content of the spores increased by a factor of ⬃104 compared to that of the unexposed spore (Fig. 3). NanoSIMS Li depth profiles of exposed spores show that Li concentrations were higher in the spore outer structures (exosporium, coat, and cortex) relative to the core (Fig. 3A). In the second series of experiments, we examined Li uptake in hydrated B. thuringiensis subsp. israelensis spores as a function of exposure time. In this case, a lower-concentration (0.4 mM) LiF solution was used to minimize the unwanted precipitation of salt crystals during the exposure. As with the 10 mM
LiF solution, exposure to 0.4 mM LiF solution resulted in rapid permeation of Li throughout the spore. Average Li distribution profiles in spores for the experimental exposure times are shown in Fig. 4B. In all cases, Li content was higher in the spore outer structures than in the core. Also, the Li content increased steadily with increasing exposure times extending into days, as indicated by the increase in the measured Li⫹/C⫹ ion ratio over time (Fig. 4A). The kinetics of Li uptake into the core was examined by fitting the core Li⫹/C⫹ ion ratio as a function of exposure time to the following rate equation, R(t) ⫽ R0 ⫹ ⌬R[1 ⫺ e⫺(t/⌫)], where R(t) is the Li⫹/C⫹ ion ratio for the exposure time t, R0 is the initial ratio when t equals 0, ⌬R is the steady-state increment, and ⌫ is the time constant. The best-fit value for ⌫ is 18 h. Similar permeability experiments were performed with spores suspended in a 2 mM aqueous CsCl solution. Unlike LiF, exposure to the CsCl solution resulted in a significant portion (⬃40%) of the spores becoming phase dark. By SEM, these spores were observed to have lost a large portion of their contents and deflated, and their NanoSIMS P and Ca depth
FIG. 4. Average Li uptake by B. thuringiensis subsp. israelensis spores. (A) 7Li⫹/12C⫹ ion ratios in the spore outer structures and core as a function of exposure time to 0.4 mM LiF solution. The rate of Li incorporation into the spore core is modeled by fitting the core Li⫹/C⫹ ionic ratio as a function of exposure time to the rate equation, R(t) ⫽ R0 ⫹ ⌬R[1 ⫺ e⫺(t/⌫)]. Here, [7Li⫹/12C⫹] ⫽ 1.6 ⫺ 1.5et/⫺17.9. (B) Average 7Li⫹ depth profiles for spores exposed to 0.4 mM LiF for different exposure times. The spores in the depth profiles with exposure times of 7 h and 3 days are not fully consumed. The numbers of spore analyses performed for each parameters in panels A and B are 21 for 5 min, 12 for 15 min, 12 for 7 h, 13 for 1 day, and 13 for 3 days.
VOL. 76, 2010
WATER AND ION INCORPORATION IN BACILLUS SPORES
3279
FIG. 5. (A to C) Average F content (measured as 19F⫺/12C⫺) in the spore outer structures and core as a function of exposure time to 0.4 mM LiF solution (A), as a function of LiF solution concentration (B), and as a function of exposure time to deionized water (C). The data shown in panel B correspond to 3 days of exposure for each concentration. Open symbols correspond to the control samples. The control used for the 0.4 mM samples was stored hydrated, and the control used for the other treatments was stored dry. The number of spore analyses performed for each parameter is 27 for the control, 10 for 7 h, 11 for 1 day, and 33 for 3 days (A); 6 for the control, 33 for 0.4 mM, 5 for 10 mM, and 8 for 40 mM LiF (B); and 6 for the control, 5 for 10 min, and 5 for 3 days (C).
profiles differed significantly from those of viable spores (20). Previous studies involving Cs and Cl exposure in spores have reported germination (39) and toxicity (1, 36) effects. The exact nature of the CsCl effect was not pursued. For these permeation studies, only the intact spores, which yielded normal P or Ca depth profiles, were used. These depth profiles showed that CsCl exposure led to permeation of Cs⫹ throughout the spore, with more Cs accumulating in the spore outer structures. As with LiF exposure, the Cs content of intact spores increased with continued exposure over a period of several days. However, the temporal behavior of Cs uptake in spores was not pursued in detail due to the lysis effect of exposure to CsCl. Anion uptake and distribution in spores. In conjunction with the Li and Cs uptake studies, we also examined the uptake and distribution of the associated counterions, namely, fluoride (F⫺) and chloride (Cl⫺). These studies were affected by several factors which are discussed here. In contrast to the extremely low abundances of Li and Cs in native B. thuringiensis subsp. israelensis spores, F and Cl abundances are significant, typically with higher abundances of these species in the spore outer structures relative to the core (Fig. 1) (20). In our experiments, both F and Cl were detected in the spores prior to exposure to either LiF or CsCl solutions. F⫺ and Cl⫺ species are ubiquitous in the natural environment, and the medium used to produce the spore samples contained potassium chloride (KCl). In addition, Cl⫺ is a germinating agent and therefore potentially responsible for the observed lysis of a significant portion of the spore population in response to CsCl exposure (39). Figure 5 shows the average F (19F⫺/12C⫺) content in the spore core and outer structures following LiF and deionized water exposures. These data demonstrate the mobility and exchange of F⫺ into the core of spores in response to these treatments. Three concentrations of LiF (0.4, 10, and 40 mM) were used in these experiments. Also, both dry and hydrated spore preparations were utilized. Prior to exposure, spores had higher F content in the outer structures relative to the core. Exposure to 0.4 and 10 mM LiF solutions resulted in the loss of F from the spore outer structures, while the abundance of F in the core remained relatively unchanged (Fig. 5B). This led to a reversal in the spore F distribution pattern, with the core having higher F content relative to the outer structures. In the
case of 0.4 mM LiF exposure, the spore F content remained relatively unchanged over time following the initial F loss from the outer structures. In contrast, exposure to the 40 mM LiF solution resulted in a significant increase in the core F concentration (approximately four times that of the control spore sample). Exposure of dry spores to deionized water led to an initial removal of F from the spore as a whole, following which the overall F content and distribution within the spore remained unchanged (Fig. 5C). These data show the uptake, loss, and redistribution of F in the spores in response to the exposure treatments. The initial loss in the spore F content typically observed for all the treatments (Fig. 5) is likely due to the removal of postproduction F contaminants. However, the source of these contaminants is not known. For the 0.4 mM LiF treatment, the starting spore sample was stored in double-distilled water which showed no significant F⫺ contamination by ion chromatography. The other treatments utilized dry spores, which may have absorbed F from ambient sources. In all of the LiF treatments, the spores developed higher F abundances in the core relative to the outer structures (Fig. 5B), thereby reversing the initial F distribution observed prior to the exposures. We interpret the resulting F distribution to be an equilibrium state of the spore relative to the surrounding solution. Potentially, these equilibria may have been affected by the starting state of the spores (dry versus hydrated), but the data do not indicate a major effect. Time-resolved F data were collected for the 0.4 mM LiF and deionized water treatments (Fig. 5A and C). These data show that changes in spore F abundance occurred rapidly relative to changes in the spore Li concentration (Fig. 4). However, the F data are not directly comparable to the Li uptake results since the concentration of F either decreased or remained relatively constant in the spores over time instead of increasing as Li concentrations did. The Cl data obtained from the LiF and deionized water exposures show permeation of Cl⫺ into the spore (Fig. 6). However, while the spore Cl content was altered by these exposures (including exposure to a CsCl solution), the overall distribution of Cl remained unchanged, with the outer structures having higher Cl content relative to the core. Cl contents of both the outer structures and the core were reduced in the
3280
GHOSAL ET AL.
APPL. ENVIRON. MICROBIOL.
FIG. 6. (A to C) Change in spore core and outer structure Cl content (measured as 35Cl⫺/12C⫺) with exposure time to 0.4 mM LiF solution (A), LiF concentration (B), and exposure time to deionized water (C). Open symbols indicate data for the control samples. The control used for the 0.4 mM samples was stored hydrated, and the control used for the other treatments was stored dry. The number of spore analyses performed for each parameter is 27 for the control, 10 for 7 h, 11 for 1 day, and 33 for 3 days (A); 6 for the control, 5 for 10 mM, and 8 for 40 mM LiF (B); and 6 for the control, 5 for 10 min, and 5 for 3 days (C).
0.4 mM LiF treatments, with the exception of the spore core Cl content obtained with the 3-day treatment, which was similar to the starting value (Fig. 6A). In the 40 mM LiF, 10 mM LiF, and deionized water treatments, the Cl content unexpectedly increased (Fig. 6B and C). Ion chromatography analyses for Cl⫺ found no Cl contamination in the LiF solutions or deionized water. One potential explanation for the difference among the treatments is the starting spore material. The treatments involving dry spores resulted in increased spore Cl content, while those using hydrated spores did not. Potentially, the increase in spore Cl obtained by the treatments utilizing dry spores resulted from mobilization of Cl⫺ associated with precipitates or contaminants in the samples following hydration, with subsequent incorporation of Cl by the spores. Whatever the source of Cl in these treatments, the data demonstrate that the Cl contents of both the spore core and the outer structures can be modified relative to the starting composition and that the outer structures maintain higher Cl content relative to the core. DISCUSSION The results presented here demonstrate that water and monatomic ions in solution permeate B. thuringiensis subsp. israelensis spores into the core through the exosporium, coat, and cortex. Upon exposure to aqueous solutions of LiF and CsCl, measurable amounts of cations and, in some cases, anions are retained by the spore postexposure, resulting in time-dependent net ion uptake. We postulate that the uptake of solvated ions is the combination of the following two equilibrium processes: (i) movement of ions into the spore in response to a chemical gradient (diffusion) or hydration gradient (advection) and (ii) adsorption at available sites within the spore via an exchange mechanism. For cations, the equation, X⫹ ⫹ R-Y 7 R-X ⫹ Y⫹, applies, where Xⴙ is the permeating cation, R represents an organic ligand at the spore exchange site, and Yⴙ is another cation or the hydronium ion. With dehydration, the majority of unbound ions and water are presumed to leave the spores. We postulate that charge balance is maintained by the copropagation of cations and anions. Here, we limit the discussion to diffusion into hydrated spores, which is relevant to spore production. The data acquired showing the evolution of elemental gradients in spores
over time can be used to determine whether diffusion or exchange is the rate-limiting step. To the first order, under conditions where diffusion is the rate-limiting step, the elemental profile will form a gradient from edge to core. As elemental concentrations approach equilibrium, the profile will flatten. In the case where diffusion is rapid compared to exchange at active sites, there would be no gradient or solely a gradient determined by the distribution of exchange sites. Under these conditions, elemental abundances should increase as a function of exposure until all of the exchange sites are saturated. In our hydrated spore experiments, the Li uptake profiles showed gradients (Fig. 3B), but the gradient structure was independent of exposure time, and Li content increased with increasing exposure time. On the basis of these observations, we conclude that under hypotonic conditions, uptake of cations in B. thuringiensis subsp. israelensis spores is limited by the rate of exchange. This conclusion is not surprising considering our and previous results with respect to the permeability of spores to water (6, 28, 29, 34, 38, 40, 45, 54). The diffusion rates of ions into spores would have to be many orders of magnitude lower than the rates seen in water (51) to produce diffusion rates that approach the rate of uptake observed in this study. Previous studies on bacterial spores have led to similar conclusions (17). Our results confirm that diffusion is relatively rapid, and solvated cations permeate the entire spore, including the core, on a time scale of seconds to minutes (5 min being the shortest time point studied in our experiments). The observed cation gradients in spores indicate higher concentrations of monovalent cation exchange sites in the outer structures relative to the core. Bacterial cell walls are known to adsorb cations (2, 3, 9, 16, 18, 53). They contain carboxylate, phosphate, hydroxylate, and amino functional groups that serve as exchange sites for metal cations via deprotonation. Carboxylate and phosphate groups are negatively charged under hydrated conditions, which allow them to be potent scavengers of cations. In Bacillus spores, the spore structure is characterized by several proteinaceous spore coat layers and by cortex and core wall peptidoglycan enclosing the central core region. For species such as B. thuringiensis subsp. israelensis, the coat is covered by an additional outer layer, called the exosporium, which is primarily composed of proteins (35). All of these layers contain potential exchange sites for cation uptake. He and Tebo (22) studied the adsorption of Cu(II) by
VOL. 76, 2010
WATER AND ION INCORPORATION IN BACILLUS SPORES
spores of the marine Bacillus sp. strain SG-1 and concluded that spore surface functional groups are responsible for the adsorption of metal cations by the SG-1 spores. These observations were based on bulk measurements and therefore did not provide information regarding the distribution of cations within the spore interior. Potential exchange sites for cations in the core are phosphate and amino groups that are associated with the spore DNA, small-acid-soluble spore proteins, and dipicolinic acid (DPA). Based on the high abundance of DPA in the spore core and the relatively low retention of Li and Cs in the spore core seen in these experiments, spore core DPA does not readily bind monovalent cations under the conditions used in this study. What is surprising is how slowly ion exchange took place in these experiments. Based on our conclusion that exchange is the rate-limiting step, the estimate of ⌫ equaling 18 h for Li uptake by the spore core is representative of the exchange process. A similar value of ⌫ was also estimated for Li uptake by the spore outer structures (exosporium, coat, and cortex). The similarity in these values suggests that the uptake of Li at both of these locations in the spore involves similar exchange mechanisms. These rates are 2 orders of magnitude lower than those previously reported for spores (22) and bacterial cell walls (2, 3, 9, 16, 18, 53). He and Tebo (22) found that more than 60% of Cu(II) was adsorbed by the marine Bacillus sp. strain SG-1 within the first minute. In contrast, the time needed for 60% Li uptake in our experiments is approximately 1 day. A possible reason for this difference is that the measurements of adsorption are typically performed under hydrated conditions, whereas our measurements probe species remaining in spores after dehydration. The pH and ionic strength of the LiF solution in this study may also influence the nature and rate of ion exchange (15, 56). Our data indicate that anion incorporation is also controlled by equilibrium binding (Fig. 5 and 6). Potential anion binding sites include proteins and cations. The difference in F⫺ and Cl⫺ equilibrium distributions may be explained by the affinity of F⫺ for Ca2⫹, which is abundant in the spore core. The difference in the distribution of cations and anions in the spores demonstrates that elemental incorporation is independent of the counterion, even if copropagation maintains charge balance. The anion data (Fig. 5 and 6) indicate postproduction accumulation of contaminants in the spore outer structures, which can be removed with washing. The presence of these contaminants is notable because the starting samples were washed repeatedly in double-distilled water and stored in double-distilled water, in the case of the hydrated spores. This observation indicates that accumulation of contaminants could be used to reconstruct the postproduction history of a sample. More work would be necessary to understand the potential sources and the exact information that could be gained. Under standard growth conditions, calcium is the most abundant metal in Bacillus spores because it is chelated by DPA in the spore core (43, 50). Previous work has shown that in the absence of Ca, other elements will substitute for Ca (18, 47), demonstrating that the elemental composition of bacterial spores is strongly influenced by the elemental composition of the media used to produce the spores. It has also been shown
3281
that Ca can be removed from spores by acid treatments and replaced by other elements (32). This study demonstrates that elemental incorporation can occur after spore formation through solute permeation without special treatment and, therefore, must be considered when analyzing the elemental composition of bacterial spore samples. Notably, the monovalent cations are preferentially incorporated in the outer structures of spores but also penetrate to, and accumulate in, the core. Our data demonstrate that permeation is rapid relative to incorporation, and the primary control of the distribution of permeating elements in a spore is the availability of binding/exchange sites. Because the rates of incorporation found here are low relative to standard processes such as washes, the level of incorporation potentially could be used to determine the time of exposure under known conditions. Finally, our results demonstrate that elemental abundances and distributions in spores are influenced by the chemistry of the local environment and, hence, present novel opportunities for environmental microbiology and microbial forensics. These results also demonstrate the need to collect samples and prepare them for analysis using methods that minimize the alteration of the elemental or isotopic signatures of interest. ACKNOWLEDGMENTS We thank C. Ramon for laboratory assistance, J. Smith for LiF solutions, E. Ramon and R. Lindvall for water analysis, L. Nittler for software development, and the reviewers for helpful comments. This work was performed under the auspices of the U.S. Department of Energy at Lawrence Livermore National Laboratory under contract DE-AC52-07NA27344, with financial support from Laboratory Directed Research and Development, LLNL; the Federal Bureau of Investigation; and the Department of Homeland Security. REFERENCES 1. Avery, S. V. 1995. Cesium accumulation by microorganisms: uptake mechanisms, cation competition, compartmentalization and toxicity. J. Ind. Microbiol. 14:76–84. 2. Beveridge, T. J., and R. G. E. Murray. 1980. Sites of metal deposition in the cell wall of Bacillus subtilis. J. Bacteriol. 141:876–887. 3. Beveridge, T. J., and R. G. E. Murray. 1976. Uptake and retention of metals by cell walls of Bacillus subtilis. J. Bacteriol. 127:1502–1518. 4. Bhattacharjee, Y., and M. Enserink. 2008. Anthrax investigation: FBI discusses microbial forensics—but key questions remain unanswered. Science 321:1026–1027. 5. Black, S. H., and P. Gerhardt. 1961. Permeability of bacterial spores. I. Characterization of glucose uptake. J. Bacteriol. 82:743–749. 6. Black, S. H., and P. Gerhardt. 1962. Permeability of bacterial spores. IV. Water content, uptake and distribution. J. Bacteriol. 83:960–967. 7. Black, S. H., R. E. MacDonald, T. Hashimoto, and P. Gerhardt. 1960. Permeability of dormant bacterial spores. Nature 185:782–783. 8. Bohaty, R. 2008. FBI’s anthrax analysis: experts discuss experimental techniques used to identify suspect Ivins. Chem. Eng. News 86:6. 9. Borrok, D., J. B. Fein, M. Tischler, E. O. O’Loughlin, H. Meyer, M. Liss, and K. M. Kemmer. 2004. The effect of acidic solutions and growth conditions on the adsorptive properties of bacterial surfaces. Chem. Geol. 209:107–119. 10. Brewer, L. N., J. A. Ohlhausen, P. G. Kotula, and J. R. Michael. 2008. Forensic analysis of bioagents by X-ray and TOF-SIMS hyperspectral imaging. Forensic Sci. Int. 179:98–106. 11. Budowle, B., S. E. Schutzer, M. S. Ascher, R. M. Atlas, J. P. Burans, R. Chakraborty, J. J. Dunn, C. M. Fraser, D. R. Franz, T. J. Leighton, S. A. Morse, R. S. Murch, J. Ravel, D. L. Rock, T. R. Slezak, S. P. Velsko, A. C. Walsh, and R. A. Walters. 2005. Toward a system of microbial forensics: from sample collection to interpretation of evidence. Appl. Environ. Microbiol. 71:2209–2213. 12. Budowle, B., S. E. Schutzer, S. A. Morse, K. F. Martinez, R. Chakraborty, B. L. Marrone, S. L. Messenger, R. S. Murch, P. J. Jackson, P. Williamson, R. Harmon, and S. P. Velsko. 2008. Criteria for validation of methods in microbial forensics. Appl. Environ. Microbiol. 74:5599–5607. 13. Cliff, J. B., K. H. Jarman, N. B. Valentine, S. L. Golledge, D. J. Gaspar, D. S. Wunschel, and K. L. Wahl. 2005. Differentiation of spores of Bacillus subtilis
3282
14.
15.
16.
17.
18.
19. 20.
21.
22.
23.
24.
25. 26.
27.
28.
29. 30.
31. 32. 33.
34.
GHOSAL ET AL.
grown on different media by elemental characterization using time-of-flight secondary ion mass spectrometry. Appl. Environ. Microbiol. 71:6524–6530. Cowan, A. E., E. M. Olivastro, D. E. Koppel, C. A. Loshon, B. Setlow, and P. Setlow. 2004. Lipids in the inner membrane of dormant spores of Bacillus species are largely immobile. Proc. Natl. Acad. Sci. U. S. A. 101:7733–7738. Daughney, C. J., and J. B. Fein. 1998. The effect of ionic strength on the adsorption of H⫹, Cd2⫹, Pb2⫹, and Cu2⫹ by Bacillus subtilis and Bacillus licheniformis: a surface complexation model. J. Colloid Interface Sci. 198: 53–77. Doyle, R. J., T. H. Matthews, and U. N. Streips. 1980. Chemical basis for selectivity of metal ions by the Bacillus subtilis cell wall. J. Bacteriol. 143: 471–480. Fernando, W. J. N., and R. Othman. 2006. Relevance of diffusion through bacterial spore coats/membranes and the associated concentration boundary layers in the initial lag phase of inactivation: a case study for Bacillus subtilis with ozone and monochloramine. Math. Biosci. 199:175–187. Foerster, H., and J. Foster. 1966. Endotrophic calcium, strontium, and barium spores of Bacillus megaterium and Bacillus cereus. J. Bacteriol. 91:1333– 1345. Gerhardt, P., and S. H. Black. 1961. Permeability of bacterial spores. II. Molecular variables affecting solute permeation. J. Bacteriol. 82:750–760. Ghosal, S., S. J. Fallon, T. Leighton, K. E. Wheeler, I. D. Hutcheon, and P. K. Weber. 2008. Imaging and 3D elemental characterization of intact bacterial spores with high-resolution secondary ion mass spectrometry (NanoSIMS) depth profile analysis. Anal. Chem. 80:5986–5992. Gikunju, C. M., S. M. Lev, A. Birenzvige, and D. M. Schaefer. 2004. Detection and identification of bacteria using direct injection inductively coupled plasma mass spectrometry. Talanta 62:741–744. He, L. M., and B. M. Tebo. 1998. Surface charge properties of and Cu(II) adsorption by spores of the marine Bacillus sp. strain SG-1. Appl. Environ. Microbiol. 64:1123–1129. Hodgson, C. J., J. Perkins, and J. C. Labadz. 2003. Evaluation of biotracers to monitor effluent retention time in constructed wetlands. Lett. Appl. Microbiol. 36:362–371. Hodgson, C. J., J. Perkins, and J. C. Labadz. 2004. The use of microbial tracers to monitor seasonal variations in effluent retention in a constructed wetland. Water Res. 38:3833–3844. Horita, J., and A. A. Vass. 2003. Stable-isotope fingerprints of biological agents as forensic tools. J. Forensic Sci. 48:1–5. Kreuzer-Martin, H. W., M. J. Lott, J. V. Dorigan, and J. R. Ehleringer. 2003. Microbe forensics: oxygen and hydrogen stable isotope ratios in Bacillus subtilis cells and spores. Proc. Natl. Acad. Sci. U. S. A. 100:815–819. Lechene, C., F. Hillion, G. McMahon, D. Benson, A. Kleinfeld, J. Kampf, D. Distel, Y. Luyten, J. Bonventre, D. Hentschel, K. Park, S. Ito, M. Schwartz, G. Benichou, and G. Slodzian. 2006. High-resolution quantitative imaging of mammalian and bacterial cells using stable isotope mass spectrometry. J. Biol. 5:20. Leuschner, R. G. K., and P. J. Lillford. 2000. Effects of hydration on molecular mobility in phase-bright Bacillus subtilis spores. Microbiology 146: 49–55. Lewis, J. C., N. S. Snell, and H. K. Burr. 1960. Water permeability of bacterial spores and the concept of a contractile cortex. Science 132:544–545. Lindsay, J. A., T. C. Beaman, and P. Gerhardt. 1985. Protoplast water content of bacterial spores determined by buoyant density sedimentation. J. Bacteriol. 163:735–737. Longchamp, P., and T. Leighton. 2000. Molecular recognition specificity of Bacillus globigii spore antibodies. Lett. Appl. Microbiol. 31:242–246. Marquis, R. E., and G. R. Bender. 1985. Mineralization and heat resistance of bacterial spores. J. Bacteriol. 161:789–791. Moreau, J. W., P. K. Weber, M. C. Martin, B. Gilbert, I. D. Hutcheon, and J. F. Banfield. 2007. Extracellular proteins limit the dispersal of biogenic nanoparticles. Science 316:1600–1603. Neihof, R., J. K. Thompson, and V. R. Deitz. 1967. Sorption of water vapour and nitrogen gas by bacterial spores. Nature 216:1304–1306.
APPL. ENVIRON. MICROBIOL. 35. Nicholson, W., and P. Setlow. 1990. Sporulation, germination, and outgrowth, p. 391–450. In S. M. Cutting et al. (ed.), Molecular biological methods for Bacillus. John Wiley & Sons, West Sussex, United Kingdom. 36. Perkins, J., and G. M. Gadd. 1995. The influence of pH and external K⫹ concentration on caesium toxicity and accumulation in Escherichia coli and Bacillus subtilis. J. Ind. Microbiol. 14:218–225. 37. Piggee, C. 2008. Tracing killer spores: the science behind the anthrax investigation. Anal. Chem. Online News http://pubs3.acs.org/journals/ancham /news/2008/09/18/cp_anthrax.html. 38. Plomp, M., T. J. Leighton, K. E. Wheeler, and A. J. Malkin. 2005. The high-resolution architecture and structural dynamics of Bacillus spores. Biophys. J. 88:603–608. 39. Rode, L. J., and J. W. Foster. 1962. Ionic germination of spores of Bacillus megaterium QM B1551. Arch. Mikrobiol. 43:183–200. 40. Rubel, G. O. 1997. A non-intrusive method for the measurement of water vapour sorption by bacterial spores. J. Appl. Microbiol. 83:243–247. 41. Ryzhov, V., Y. Hathout, and C. Fenselau. 2000. Rapid characterization of spores of Bacillus cereus group bacteria by matrix-assisted laser desorptionionization time-of-flight mass spectrometry. Appl. Environ. Microbiol. 66: 3828–3834. 42. Scherrer, R., T. C. Beaman, and P. Gerhardt. 1971. Macromolecular sieving by the dormant spore of Bacillus cereus. J. Bacteriol. 108:868–873. 43. Scherrer, R., and P. Gerhardt. 1972. Location of calcium within Bacillus spores by electron probe X-ray microanalysis. J. Bacteriol. 112:559–568. 44. Scherrer, R., and P. Gerhardt. 1971. Molecular sieving by the Bacillus megaterium cell wall and protoplast. J. Bacteriol. 107:718–735. 45. Setlow, B., and P. Setlow. 1980. Measurements of the pH within dormant and germinated bacterial spores. Proc. Natl. Acad. Sci. U. S. A. 77:2474–2476. 46. Sinton, L. W., R. R. Braithwaite, C. H. Hall, L. Pang, M. E. Close, and M. J. Noonan. 2005. Tracing the movement of irrigated effluent into an alluvial gravel aquifer. Water Air Soil Pollut. 166:287–301. 47. Slepecky, R., and J. W. Foster. 1959. Alterations in metal content of spores of Bacillus megaterium and the effect on some spore properties. J. Bacteriol. 78:117–123. 48. Slodzian, G., B. Daigne, F. Girard, F. Boust, and F. Hillion. 1992. A highresolution scanning ion microscope with parallel detection of secondary ions, p. 169–178. In A. Benninghoven, K. T. F. Janssen, J. Tumpner, and H. W. Werner (ed.), Proceedings of the 8th International Conference for Secondary Ion Mass Spectrometry 1991, SIMS VIII. John Wiley & Sons, Chichester, United Kingdom. 49. Stewart, M., A. P. Somlyo, A. Somlyo, H. Suman, J. A. Lindsay, and W. G. Murrell. 1981. Scanning electron probe X-ray microanalysis of elemental distributions in freeze-dried cryosections of Bacillus coagulans spores. J. Bacteriol. 147:670–674. 50. Stewart, M., A. P. Somlyo, A. V. Somlyo, H. Suman, J. A. Lindsay, and W. G. Murrell. 1980. Distribution of calcium and other elements in cryosectioned Bacillus cereus T spores, determined by high-resolution scanning electron probe X-ray microanalysis. J. Bacteriol. 143:481–491. 51. Stokes, R. H. 1950. The diffusion coefficient of eight uni-univalent electrolytes in aqueous solution at 25C. J. Am. Chem. Soc. 72:2243–2247. 52. Sunde, E. P., P. Setlow, L. Hederstedt, and B. Halle. 2009. The physical state of water in bacterial spores. Proc. Natl. Acad. Sci. U. S. A. 106:19334–19339. 53. Urrutia, M. M., and T. J. Beveridge. 1993. Mechanism of silicate binding to the bacterial cell wall in Bacillus subtilis. J. Bacteriol. 175:1936–1945. 54. Westphal, A. J., P. B. Price, T. J. Leighton, and K. E. Wheeler. 2003. Kinetics of size changes of individual Bacillus thuringiensis spores in response to changes in relative humidity. Proc. Natl. Acad. Sci. U. S. A. 100:3461–3466. 55. Whiteaker, J. R., C. Fenselau, D. Fetterolf, D. Steele, and D. Wilson. 2004. Quantitative determination of heme for forensic characterization of Bacillus spores using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Anal. Chem. 76:2836–2841. 56. Yee, N., D. A. Fowle, and F. G. Ferris. 2004. A Donnan potential model for metal sorption onto Bacillus subtilis. Geochim. Cosmochim. Acta 68:3657– 3664.
