|Home | About | Journals | Submit | Contact Us | Français|
Silicon (Si) is considered to be a “quasiessential” element for most living organisms. However, silicate uptake in bacteria and its physiological functions have remained obscure. We observed that Si is deposited in a spore coat layer of nanometer-sized particles in Bacillus cereus and that the Si layer enhances acid resistance. The novel acid resistance of the spore mediated by Si encapsulation was also observed in other Bacillus strains, representing a general adaptation enhancing survival under acidic conditions.
Silicon (Si), the second-most-abundant element in the earth's crust, is an important mineral for living organisms; it acts as a component of the outer skeleton of diatomaceous protozoans (1), as a trace element to help animal bone and tooth development (5), and as an element in plants that enhances their tissue strength and disease resistance (8, 9). These organisms take up silicate from the environment and accumulate it as silica that is formed from highly concentrated silicate (27). In 1980, relatively high concentrations of Si were observed at the spore coat region of Bacillus cereus and Bacillus megaterium spores by an analysis using scanning transmission electron microscopy (STEM) (14, 23). However, due to the low resolution and relatively weak signal, the precise localization of Si was not determined. On the other hand, the Si contents of Bacillus coagulans and Bacillus subtilis spores were reported to be almost absent or under the detection limit (4, 24). Some bacteriologists familiar with these data consider the presence of Si an anomaly (17). The presence of Si in bacterial spores (specifically, the spores of Bacillus anthracis) again became the focus of attention when anthrax spores were mailed to U.S. senators in the fall of 2001 (17). The Senate anthrax spores could be easily dispersed as single spores when the container was opened. The investigators considered that coating spores with silica might be involved in preventing spores from sticking to each other (17). Thus, if silica is normally absent from spores, its presence in B. anthracis spores suggested that they had been weaponized (17). Subsequent analysis convinced the investigators that the Si was a natural occurrence (3). However, since silica-rich and -poor spores of the same bacterial strain have never been compared, any relationship between naturally accumulated silica and spore dispersion remained hypothetical.
In the present study, we screened for the bacterium that takes up the largest amount of silicate from among a number of strains isolated from paddy field soil in order to study Si uptake, clarify the localization of Si, and reveal the roles of Si in bacteria. The effect of silica on spore dispersion was also discussed.
Half of a gram of soil that was collected in a paddy field (Higashi-Hiroshima, Japan) was suspended in 10 ml of sterilized water and mixed well. After serial dilution, the suspensions were spread on an mR2A agar medium (an R2A agar medium  supplemented with 0.2 mM CaCl2, 0.01 mM MnCl2, 0.05 mM ZnCl2, 0.05 mM FeSO4) and incubated at 28°C for 48 h. Then, each of the colonies was transferred to a liquid mR2A medium containing 10 μg/ml silicate, and the silicate concentrations after 24 and 48 h of incubation were measured by the molybdenum blue assay method (7).
Universal bacterial 16S rRNA gene PCR primers 27F (forward primer, 5′-AGAGTTTGATCCTGGCTCAG-3′) and 1510R (reverse primer, 5′-GTCCCGCAACGAGCGCAAC-3′) were used to amplify the 16S rRNA genes (28). DNA sequences were determined with an automated laser fluorescence sequencer (ABI310; Amersham Biosciences) by using the following primers: 27f, 5′-AGAGTTTGATCCTGGCTCAG-3′; rlL, 5′-GTATTACCGCGGCTGG-3′; r2L, 5′-CATCGTTTACGGCGTGGAC-3′; r3L, 5′-TTGCGCTCGTTGCGGGACT-3′; and r4L, 5′-ACGGGCGGTGTGTACAAG-3′. The 16S rRNA gene sequences were aligned by using the ClustalW program (25), using default parameters. The program TreeView (20) was used to generate the phylogenetic tree.
For scanning electron microscopy with energy dispersive X-ray spectrometry (SEM-EDX) analysis, spores were fixed with 2.5% glutaraldehyde for 6 h at 4°C and dehydrated with ethanol, followed by substitution of tert-butyl alcohol. After freeze-drying, the spores were analyzed with SEM-EDX (JSM-5900; JEOL, Japan). For both TEM and STEM analysis, spores were prepared by a procedure similar to that used for SEM-EDX, except that spores were fixed by using cold 2.5% glutaraldehyde, 0.1% MgSO4 in 0.1 M cacodylate buffer (pH 7.2) for 2 h at 4°C and stained by using 1% osmium tetraoxide in 0.2 M cacodylate buffer (pH 7.4) for 36 h at 4°C. These spores were suspended in molten 2% agarose. After solidification, part of the agarose containing spores was cut into small cubes, which were dehydrated in ethanol and embedded in epoxy resin. Serial thin sections were prepared and stained with uranyl acetate and lead citrate for structural analysis with TEM (H-9000UHR; Hitachi, Japan). Unstained ultrathin sections were used for STEM-EDX analysis (HD-2000; Hitachi, Japan).
Bacterial cells were cultured in 100 ml of the mR2A medium for 50 h at 28°C with or without 100 μg/ml silicate. The cultures were centrifuged at 6,000 × g for 15 min at 4°C, and each spore pellet was washed twice using an equal volume of cold, sterile distilled water. The spores were suspended in a final volume of 10-ml sterile water and kept at 4°C for 1 to 7 days. Microscopic observation indicated that more than 99.8% of cells were refractile spores. The spore solution was diluted to an optical density at 600 nm of 1.0 for the viability tests. The titer in this solution was approximately 108 CFU/ml. For testing wet-heat resistance, spores were incubated in water at 80°C. For testing UV resistance, spores were irradiated under the UV lamp of a FASIII system (Toyobo, Shiga, Japan), with a maximum output at 254 nm. For hydrogen peroxide resistance, spores were incubated at 28°C in 50 mM K2HPO4 buffer (pH 7.5) with 0.5% H2O2. For alkaline and acid resistance, spores were centrifuged and resuspended in equal volumes of 0.5 N sodium hydroxide (NaOH), 0.4 N hydrochloric acid (HCl), and 0.1 N nitric acid (HNO3), respectively. At the times indicated in figures, samples were diluted with cold water and spread on the R2A agar medium. Spore viability was determined by counting the colonies after a 24-h incubation at 28°C.
Hydrogen fluoride (HF) was used to dissolve silica of the spore layer. The spore solution, diluted to an optical density at 600 nm of 1.0, was centrifuged, and the pellet was suspended in equal volumes of 50 mM HF. Silicate released from the spore was measured as described above.
To investigate the physical properties of the spores, we prepared spore powder by grinding freeze-dried spores in a mortar. Spore density, which is required for the calculation of electrostatic charge value, was measured with a gas displacement pycnometer (Ultrapycnometer 1000; Quantachrome Instruments, FL). The electrostatic charge distribution and diameter of spores were determined by using an electrical single-particle aerodynamic relaxation time (E-SPART) analyzer (18) (Hosokawa Micron, Osaka, Japan). About 100 mg of the spore powder was shaken in a plastic bag, and then the charges and diameters of individual spores were measured with the E-SPART analyzer. The number of particles counted for each experiment was 3,000.
The 16S rRNA gene sequence data for YH64 and YH221 have been deposited in the GenBank database under the respective accession nos. GQ855295 and GQ855296.
A total of 240 bacterial colonies were obtained from paddy field soil and transferred to an mR2A medium (an R2A medium supplemented with trace amounts of CaCl2, MnCl2, ZnCl2, and FeSO4) in the presence or absence of 10 μg/ml silicate. Of the 240 isolates, 29 were capable of taking up more than 75% of the silicate from the medium after a 48-h incubation (data not shown). The 16S rRNA gene sequences of these 29 bacteria showed maximum homology to sequences of the genus Bacillus, and 21 of these 29 bacteria were homologous to the B. cereus group, a very homogenous cluster of six species: B. cereus, Bacillus thuringiensis, B. anthracis, Bacillus mycoides, Bacillus pseudomycoides, and Bacillus weihenstephanensis (26). The other eight bacterial isolates showed maximum homology to Bacillus shandongensis or Bacillus megaterium. The bacteria that take up silicate were all phylogenetically classified as belonging to the genus Bacillus. The strain that takes up the largest amount of silicate among the isolated strains was classified as B. cereus based on its 16S rRNA gene sequence (Fig. (Fig.1)1) and designated B. cereus strain YH64.
To analyze the silicate uptake along with the cell growth, the silicate concentration and cellular morphology of the YH64 strain were monitored during incubation on the mR2A medium containing 100 μg/ml silicate (Fig. (Fig.2A).2A). After the late-logarithmic growth phase (20 to 40 h), the silicate concentration in the medium decreased drastically. The silicate concentration dropped to ~0.05 μg/ml after 40 h, followed by a slight increase, indicating that a portion of the incorporated silicate was released (Fig. (Fig.2A).2A). After 48 h, almost all cells had formed matured spores just after the silicate concentration reached its minimum (Fig. (Fig.2B).2B). On the R2A medium, which does not contain trace minerals, YH64 did not take up silicate (Fig. (Fig.2A)2A) and showed impaired sporulation (approximately 10% of the sporulation efficiency observed in the mR2A medium) (Fig. (Fig.2B),2B), indicating that silicate uptake is related to spore formation.
To determine the relationship between silicate uptake and spore formation, we measured silicate concentration and the number of heat-resistant spores and confirmed the timing of the appearance of refractile spores during sporulation in the mR2A medium containing 100 μg/ml silicate from 12 to 36 h. Refractile spores appeared at around 16 h (data not shown). Heat-resistant spores appeared between 16 h and 18 h (Fig. (Fig.2C).2C). Silicate uptake started at around 22 h, and more than 90% of silicate was taken up by 36 h. Mother cells still remained at 36 h (data not shown). These results indicated that silicate uptake occurs after the spores acquire heat resistance in their maturing process.
We prepared YH64 spores from the culture using mR2A with or without silicate and then analyzed them by SEM-EDX. The EDX signal of Si was not observed in the YH64 spores harvested from the culture without silicate; these spores contained almost no or a very small amount of silicate (Fig. (Fig.2D)2D) and were denoted as low-Si spores. On the other hand, the Si signal was clearly observed in spores from the culture with silicate (Fig. (Fig.2D),2D), and these spores were denoted as high-Si spores. We added silicate to the low-Si spores. The low-Si spores could not take up silicate (data not shown), indicating that silicate was first incorporated in the mother cell and then accumulated in the spore during maturation.
Previous studies have revealed the presence of Si in the spore coat region of B. cereus and B. megaterium spores by using STEM (14, 23). Since YH64 took up 10-fold more silicate than did a type strain of B. cereus (data shown below), we expected that a relatively strong Si signal should enable us to precisely determine the location of Si. Ultrathin sections of the low- and high-Si YH64 spores were prepared and then analyzed by TEM and STEM-EDX. The spore layer structure (from the outside to the inside of the spore) consists of the exosporium (EX), coat (CT), undercoat (UC), cortex, and core (2, 6, 11, 12). The EX and UC in the two types of spores appeared the same. However, the CT was thicker in the high-Si spore, and novel nanometer sized-particles (SX) accumulated around the outer side of the CT in the high-Si spore (Fig. 3A and B). SX might correspond to the outer coat layer, using the B. subtilis analogy (12). Si signal mapping revealed that Si is present in the CT layer and SX particles but not in other regions (Fig. (Fig.3C).3C). EDX analysis of the CT gave an intense Si signal but no signals for other minerals, such as Mn, Zn, Ca, and Fe (data not shown).
To investigate the role of Si in spore dispersion, we prepared spore powder by grinding freeze-dried spores in a mortar. Then, we placed 10-mg samples of spore powder into clear 30-ml glass vials and shook them for a few seconds. However, unlike the Senate anthrax spores that floated freely (17), both low- and high-Si spores fell quickly to the bottom of the vials and stayed there (data not shown). This result indicated that Si accumulation alone did not make spores dispersible. The electrostatic charge of spores could make them repel one another and thus create self-dispersing spores (17). To test the electrostatic charges of spores, we shook the spores in a plastic bag and then applied them to an E-SPART analyzer (18). The average electrostatic charges of low- and high-Si spores were almost the same, and the individual spore charges showed similar distributions (data not shown). Furthermore, the zeta potentials of the low- and high-Si spores dispersed into water were not significantly different (data not shown). Therefore, the function of Si in bacterial spores had to be reconsidered.
Spores can survive under conditions unsuitable for growth and resist various kinds of stress. The spore coat is related to the impermeability to the spore's inner membrane; thus, the spore coat is thought to confer resistance to toxic chemicals (19). We compared the sensitivity of YH64 low- and high-Si spores to wet heat, UV irradiation, 5.0% H2O2, 0.5 N NaOH, and 0.4 N HCl. Only under the acidic condition was the viability of the high-Si spores increased compared to that of the low-Si spores (Fig. 4A to E). The viability of high-Si spores treated with a different acid solution (0.1 N HNO3) was also higher than that of low-Si spores (Fig. (Fig.4F),4F), indicating that the Si layer confers general acid resistance.
Next, we examined the acid resistance of another isolated strain, YH221, whose 16S rRNA gene sequence shared 99% identity with that of B. shandongensis (Fig. (Fig.1).1). YH221 took up 1/7 of the amount of silicate taken up by YH64 (Fig. (Fig.5).5). Surprisingly, high-Si YH221 spores had 1,000 times the survival rate of the low-Si spores after a 3-h incubation in 0.2 N HCl (Fig. (Fig.6A).6A). Then, we prepared YH221 spores in various silicate concentrations and examined their acid resistance. The acid resistance increased with increasing amounts of Si uptake (Fig. (Fig.6B).6B). We examined silicate uptake in Bacillus type strains deposited in the NITE Biological Resource Center (NBRC, Japan). B. cereus NBRC15305, B. thuringiensis NBRC101235, and B. megaterium NBRC15308 took up approximately 0.03, 0.10, and 0.05 pg of silicate per spore, respectively (Fig. (Fig.5).5). However, almost no silicate uptake was observed in B. subtilis 168 and B. mycoides NBRC101228 under these conditions (Fig. (Fig.5).5). The YH64 strain took up 15 times as much silicate (0.49 pg Si/spore, corresponding to approximately 6.3% dry weight) as its closest relative, B. cereus NBRC15305, in spite of more than 99% 16S rRNA gene identity between these two strains. Indeed, the morphological characteristics of the cells as observed under microscopy and the characteristics of the colony forms of B. cereus YH64 and NBRC15305 were different. We examined the HCl resistance of B. cereus NBRC15305 and B. thuringiensis NBRC101235 spores. As expected, the high-Si spores were more resistant to HCl than the low-Si spores (data not shown). Not all Bacillus strains take up silicate. However, it seems highly likely that the acid resistance conferred by Si encapsulation is a general phenomenon of Bacillus strains that take up silicate.
Strong mineral acids rupture Bacillus spores by breaking down spore permeability barriers (22). Most acidophilic organisms have evolved extremely efficient mechanisms to pump protons out of the intracellular space to maintain the cytoplasm at near neutral pH (15). However, the spore is metabolically inactive before germination. The Si encapsulation, probably as silica (discussed below), which is resistant to most acids, may decrease the proton permeability of the spore coat and confer acid resistance to the spores. Interestingly, Si encapsulation of an inorganic pigment, ultramarine blue, enhances acid resistance and overcomes the limitations on the use of the pigment (10).
As far as we know, diatoms, plants, and animals accumulate silicate as silica (13). Silica can be dissolved in HF (16). Accordingly, if the Si layer of spores contains silica, it could be removed from the high-Si spores with HF treatment. Approximately 75% of Si that was accumulated in the spores was released as silicate after treatment with 50 mM HF (data not shown). We compared the acid resistance of HF-treated high Si- and low-Si spores (Fig. (Fig.7).7). After HF treatment, the viability of the high-Si spores was no longer higher than that of the low-Si spores. These results indicated that the Si layer mainly contains silica and supports acid resistance.
Si is naturally available in soil and water. The silicate concentration in soil ranges from 0.1 to 0.6 mM (9.6 to 57.7 μg/ml) (8). Therefore, the acid resistance conferred by Si encapsulation may occur in nature. Spores may encounter strong acids in environments such as the digestive conditions in animal stomachs (around 0.1 N HCl), indicating that a physiological function of Si in bacteria may be to aid survival under these conditions. When the anthrax powder sent to the U.S. Senate in 2001 was found to be coated with unusual silica, it was discussed whether the silica was related to spore dispersion. We concluded that Si encapsulation is not sufficient to make spores dispersible but does contribute to survival under acidic conditions. Our findings also strongly indicate that the anthrax spores were harvested from culture on a silicate-containing medium.
This work was supported in part by the Solution Oriented Research for Science and Technology Program (PREST-SORST) of the Japan Science and Technology Agency.
Published ahead of print on 30 October 2009.