Accumulation and Release of Rare Earth Ions by Spores of Bacillus Species and the Location of These Ions in Spores

Biosorption of rare earth ions by growing cells of Bacillus species has been well studied and has attracted attention for possible hydrometallurgy applications. However, the interaction of spores from Bacillus species with rare earth ions has not been well studied. We investigated here the adsorption and/or desorption of two rare earth ions, Tb³⁺ and Dy³⁺, by Bacillus spores, the location of the adsorbed ions, and the spore properties after ion accumulation. The significant adsorption of rare earth ions on the surfaces of Bacillus spores and the ions’ rapid release by a chelator could allow the development of these spores as a biosorbent to recover rare earth ions from the environment.

ABSTRACT

Two rare earth ions, Tb³⁺ and Dy³⁺, were incorporated into spores of Bacillus species in ≤5 min at neutral pH to 100 to 200 nmol per mg of dry spores, which is equivalent to 2 to 3% of the spore dry weight. The uptake of these ions had, at most, minimal effects on spore wet heat resistance or germination, and the ions were all released upon germination, probably by complex formation with the huge depot of dipicolinic acid (DPA) released when spores germinate. Adsorbed Tb³⁺/Dy³⁺ were also released by exogenous DPA within a few minutes and faster than in spore germination. The accumulation of Tb³⁺/Dy³⁺ was not reduced in Bacillus subtilis spores by several types of coat defects, significant modification of the spore cortex peptidoglycan structure, specific loss of components of the outer spore crust layer, or the absence of DPA in the spore core. All of these findings are consistent with Tb³⁺/Dy³⁺ being accumulated in spores’ outer layers, and this was confirmed by transmission electron microscopy. However, the identity of the outer spore components binding the Tb³⁺/Dy³⁺ is not clear. These findings provide new information on the adsorption of rare earth ions by Bacillus spores and suggest this adsorption might have applications in capturing rare earth ions from the environment.

IMPORTANCE Biosorption of rare earth ions by growing cells of Bacillus species has been well studied and has attracted attention for possible hydrometallurgy applications. However, the interaction of spores from Bacillus species with rare earth ions has not been well studied. We investigated here the adsorption and/or desorption of two rare earth ions, Tb³⁺ and Dy³⁺, by Bacillus spores, the location of the adsorbed ions, and the spore properties after ion accumulation. The significant adsorption of rare earth ions on the surfaces of Bacillus spores and the ions’ rapid release by a chelator could allow the development of these spores as a biosorbent to recover rare earth ions from the environment.

INTRODUCTION

Rare earth ions, including 17 elements from La through Lu, are critical constituents in many advanced technologies. Worldwide, most rare earths are supplied by China, but large-scale mining and refining activities have caused serious environmental problems there due to the release of large amounts of rare earths and other chemicals into the environment around mining areas (1). Areas in southern China, Ganzhou in particular, are rich in rarer and more valuable heavy rare earths, including europium (Eu), gadolinium (Gd), terbium (Tb), dysprosium (Dy), holmium (Ho), erbium (Er), thulium (Tm), ytterbium (Yb), lutetium (Lu), and yttrium (Y) (2).

Rare earth ions are difficult to separate from each other due to their similar chemistries, ionic radii, and trivalent positive charge (3). This constraint and the need for environmentally friendly processing technologies have promoted the development of biotechnological hydrometallurgy, with microbial biosorption as a potentially effective and low-cost method for the recovery of rare earths (2, 4). Many methods have been suggested to take advantage of the adsorption of rare earth ions by different bacterial species, including cells of Roseobacter sp. strain AzwK-3b (5), Escherichia coli, and Caulobacter crescentus (3, 6). Cells of various Bacillus species are also able to absorb various rare earth ions. Both Bacillus thuringiensis and Bacillus subtilis cells absorb Eu³⁺, and B. subtilis cells exhibit some selectivity in rare earth ion adsorption (7–10). It appears likely that it is the Bacillus cell wall, in particular the wall teichoic acid with its multiple phosphate groups, that plays a major role in the adsorption of rare earth ions onto the cell surface (4).

In contrast to information about the adsorption of rare earth ions by growing cells of Bacillus species, much less is known about the adsorption of rare earth ions by spores of these organisms. These spores have very different structures than cells and lack teichoic acids (11). The outermost surface layer of Bacillus spores is either the crust or the exosporium and not a peptidoglycan (PG) cell wall. Beneath the outermost layer from the outside in are, in order, the outer membrane, the PG cortex and germ cell wall, the inner membrane, and finally the central core. The spore core has a very low water content, and ∼25% of its dry weight is a 1:1 chelate of Ca²⁺ with dipicolinic acid (CaDPA). Notably, DPA has a much higher affinity for rare earth ions than for Ca²⁺ or Mg²⁺ (12).

As a consequence of a unique structure and relatively dehydrated core, spores are metabolically dormant and extremely resistant to many environmental stresses, including desiccation, wet and dry heat, UV and gamma radiation, and most toxic chemicals. Despite their extreme dormancy, spores, if given the proper signals, can return to life in germination when the environment becomes conducive to cell growth. Interestingly, rare earth ions such as Tb³⁺ and even Dy³⁺ have been used in monitoring spore germination because their combination with the DPA released from the spore core in an early germination step gives highly fluorescent complexes with these and other rare earth ions (13). Tb³⁺ and Dy³⁺ (Tb³⁺/Dy³⁺) have also been found to strongly inhibit the germination of spores that have been chemically decoated or are genetically coat defective; although the mechanism of this inhibition of germination is not clear, one possible explanation is that rare earth ions directly block the inner membrane channel for CaDPA release in spore germination (14).

Given that the spore outer layer likely has both carboxylic and phosphate groups (15–17), we hypothesized that spores of Bacillus species have binding sites for rare earth ions on the spore surface. To test this hypothesis, spores of wild-type B. subtilis and mutant strains lacking various outer layer components or unable to synthesize DPA, as well as wild-type Bacillus cereus, were examined for their adsorption of the rare earth ions Tb³⁺ and Dy³⁺. The kinetics of the uptake and release of these ions with spores were measured by a fluorescence assay, and the effects of accumulated ions on spore properties, including spore germination, viability, and heat resistance, were also investigated. We also used transmission electron microscopy (TEM) to directly examine the absorption sites of rare earth ions on spores.

RESULTS

Kinetics and levels of Tb³⁺/Dy³⁺ absorption by spores.

Previous studies have found that the cells of some bacteria can absorb rare earth ions (6), but there are no reports examining these ions’ absorption by bacterial spores. To determine whether Bacillus spores can accumulate Tb³⁺/Dy³⁺ and the time course and pH optimum for this adsorption, we initially utilized spores of a number of B. subtilis strains and B. cereus strain T. The B. subtilis strains used were the wild-type (WT) PS832 and isogenic strains: (i) PS3738 and PS4150, which produce spores with coat defects, with PS4150 spores lacking both the inner and outer coat layers and retaining only most likely the thin insoluble crust protein layer (18, 19); (ii) FB122, which produces CaDPA-less spores (20, 21); and (iii) PS2066, PS2307, PS2421, and PS2422, all of which produce spores with a variety of defects in the structure of the spores’ cortex PG (22–24). To initially measure Tb³⁺/Dy³⁺ adsorption by B. subtilis WT spores, the spores were incubated at 23°C in one concentration of TbCl₃ or DyCl₃ at pH 4.5 to 8.0 for 1 to 30 min (Fig. 1). After incubation, the spores were centrifuged and rinsed to remove external solute, and the pellets were suspended in water, boiled for 30 min to extract all endogenous spore DPA, and centrifuged as described in Materials and Methods. The supernatant fluids’ Tb³⁺- DPA or Dy³⁺-DPA fluorescence was then measured, showing that (i) incubation at pH 7.4 gave optimal Tb³⁺ or Dy³⁺ accumulation and (ii) there was maximum accumulation of Tb³⁺/Dy³⁺ within 5 min. Consequently, all further incubations with TbCl₃ or DyCl₃ were for 5 min and at pH 7.4. Notably, the levels of DPA in the wild-type spores were sufficient to bind to all Tb³⁺ or Dy³⁺ adsorbed by spores, since the addition of 1 mM DPA to the boiled supernatant fluids from wild-type spores yielded no further increase in the levels of Tb³⁺ or Dy³⁺ fluorescence (data not shown).

FIG 1.

Tb³⁺ and Dy³⁺ uptake by B. subtilis PS832 spores at different pHs and times. Spores of B. subtilis PS832 were incubated at 23°C with 200 μM Tb³⁺ or Dy³⁺ at different pHs (4.5, 5.5, 6.5, 7.4, and 8.0) for 5 min (A) or at pH 7.4 for different times (0, 1, 5, 15, and 30 min) (B) and then washed. Tb³⁺/Dy³⁺-loaded spores at an OD₆₀₀ of 0.5 were boiled for 30 min and centrifuged, and the fluorescence levels of 90-μl portions of the supernatant fluids were measured in K-HEPES buffer (25 mM; pH 7.4). These values are expressed as the ratio of fluorescence at time t to maximum fluorescence × 100%, as described in Materials and Methods.

The amounts of Tb³⁺/Dy³⁺ accumulated by B. subtilis wild-type spores increased with higher TbCl₃ or DyCl₃ concentrations in incubations, but only up to ∼100 μM, and yielded 100 to 200 nmol/mg dry spores, equivalent to 2 to 3% of the spore dry weight (Fig. 2, Table 1). Since DPA comprises ∼10% of the spore dry weight, there is a significant excess of spore DPA over maximum amounts of Tb³⁺/Dy³⁺ adsorbed. The greater adsorption of Dy³⁺ than of Tb³⁺ indicates that spores may have some selectivity in the adsorption of rare earth ions, although there was no major difference in the adsorption capacity between B. subtilis and B. cereus wild-type spores. Although chemically decoated and untreated B. subtilis wild-type spores exhibited similar adsorption capacities for Tb³⁺/Dy³⁺, spores lacking the coat assembly mutant SafA, the more severely coat-defective B. subtilis PS4150 spores, and the CaDPA-less FB122 spores adsorbed more Tb³⁺ than did wild-type spores (Table 1). This finding with the CaDPA-less spores, in particular, suggests that the Tb³⁺/Dy³⁺ were not accumulated in the spore core and complexed with DPA, to which Tb³⁺/Dy³⁺ bind extremely tightly and much more tightly than to the Ca²⁺ that normally chelates almost all DPA in the spore core (12, 25). A variety of changes in the spore cortex PG structure, including altered cross-linking and lack of the spore-specific modification muramic acid–δ-lactam, also had no effects on Tb³⁺/Dy³⁺ adsorption (Table 1).

FIG 2.

Tb³⁺ and Dy³⁺ uptake by B. subtilis PS832 spores from solutions of different Tb³⁺ or Dy³⁺ concentrations. B. subtilis PS832 spores were incubated with various concentrations of Tb³⁺ (□) or Dy³⁺ (■) at 23°C for 5 min in K-HEPES buffer (100 mM; pH 7.4), and then the spores were washed. Tb³⁺/Dy³⁺-loaded spores at an OD₆₀₀ of 0.5 were boiled for 30 min and centrifuged, and the fluorescence levels of 90-μl portions of the supernatant fluids were measured in K-HEPES buffer (25 mM; pH 7.4), as described in Materials and Methods.

TABLE 1.

Tb³⁺ and Dy³⁺ uptake by spores of Bacillus species and strains^a

Species and strain	Mean uptake (nmol/mg [dry spores]) ± SD
	Incubation at 20 μM		Incubation at 200 μM
	Tb3+	Dy3+	Tb3+	Dy3+
B. subtilis WT	46 ± 4.5	52 ± 4.7	100 ± 9.7	191 ± 13
B. subtilis WT (dc)b	45 ± 6.1	49 ± 6.4	112 ± 12	187 ± 16
B. subtilis PS2066	50 ± 5.8	56 ± 4.6	107 ± 11	200 ± 18
B. subtilis PS2307	49 ± 4.8	54 ± 4.5	105 ± 10	195 ± 17
B. subtilis PS2421	43 ± 5.2	47 ± 4.3	95 ± 8.9	181 ± 15
B. subtilis PS2422	48 ± 4.7	53 ± 4.8	105 ± 11	189 ± 14
B. subtilis PS3738	58	66	182	317
B. subtilis PS4150	76 ± 7.7	88 ± 10	166 ± 17	186 ± 24
B. subtilis FB122c	77 ± 9.2	85 ± 9.3	157 ± 18	182 ± 26
B. cereus T	51 ± 4.6	71 ± 6.2	89 ± 6.5	102 ± 12

^aSpores of various strains of B. subtilis or B. cereus in 1 ml at an OD₆₀₀ of 2.0 were incubated with 200 or 20 μM TbCl₃ or DyCl₃ for 5 min at pH 7.4 and 23°C. The spores were washed and then boiled for 30 min, and the fluorescence levels of supernatants (90 μl) after centrifugation were measured as described in Materials and Methods. The Tb³⁺ and Dy³⁺ levels were calculated from the plotted standard curves for the fluorescence of Tb³⁺-DPA and Dy³⁺-DPA in Fig. 6. Data represent average values ± the standard deviations from two independent experiments, except for strain PS3738, for which only one spore preparation was analyzed. With strains PS4150 and FB122, the values were corrected for the larger numbers of spores needed to achieve an OD₆₀₀ of 2.0 with these spores compared to wild-type B. subtilis spores with or without decoating.

^bdc, these spores were chemically decoated.

^cThese spores had <1% of the CaDPA level of wild-type spores, and the Tb³⁺ and Dy³⁺ levels in these spores were determined from the levels of fluorescence when these spores were incubated with added DPA (Fig. 4D).

To examine whether other divalent metal ions affect the adsorption of Tb³⁺/Dy³⁺ by spores, wild-type B. subtilis spores were incubated with Tb³⁺/Dy³⁺ alone or with a 10-fold excess of Ca²⁺ or Mg²⁺ ions, and the Tb³⁺/Dy³⁺ uptake was measured (Table 2). The results showed that spores incubated with a 10-fold excess of Ca²⁺ took up ∼25% less of the rare earth ions than spores incubated with Tb³⁺/Dy³⁺ alone, but there was almost no change in Tb³⁺/Dy³⁺ uptake in the presence of a 10-fold excess of Mg²⁺ (Table 2). These results further indicated that spores adsorb some divalent cations much more tightly than others.

TABLE 2.

Effect of Ca²⁺ or Mg²⁺ on Tb³⁺ and Dy³⁺ uptake by B. subtilis spores^a

Incubation	Mean uptake (nmol/mg dry spores) ± SD
	Tb3+	Dy3+
TbCl3 or DyCl3	46 ± 4.5	52 ± 4.7
Plus 200 μM CaCl2	34 ± 2.8	39 ± 3.3
Plus 200 μM MgCl2	48 ± 4.3	51 ± 4.6

^aSpores of B. subtilis PS832 in 1 ml at an OD₆₀₀ of 2.0 were incubated with TbCl₃ or DyCl₃ (20 μM) alone or plus CaCl₂ or MgCl₂ (200 μM) for 5 min at pH 7.4 and 23°C; the spores washed and then boiled for 30 min, and the fluorescence levels of supernatants after centrifugation were measured as described in Materials and Methods. Tb³⁺ and Dy³⁺ levels were calculated from the plotted standard curves for fluorescence of Tb³⁺-DPA and Dy³⁺-DPA in Fig. 6, and values represent means ± the standard deviations from two independent experiments.

Viability, heat resistance, and germination of Tb³⁺/Dy³⁺-loaded spores.

To test whether Tb³⁺/Dy³⁺ accumulation affected spore properties, B. subtilis wild-type spores were incubated with or without 20 μM TbCl₃ or DyCl₃ for 5 min and washed as described in Materials and Methods. The levels of viability of the Tb³⁺/Dy³⁺-loaded spores and untreated spores on L broth agar plates were identical (±15%), and their levels of wet heat resistance at 90°C were also almost identical (data not shown). However, spores that had absorbed Tb³⁺/Dy³⁺ germinated slightly more slowly than control spores in the first 20 min of l-valine germination (Fig. 3 and data not shown). Even more striking, and consistent with previous work (14), was that the germination of coat-defective spores that had absorbed Tb³⁺/Dy³⁺ was inhibited, but this inhibition was eliminated if DPA (50 μM) was added to Tb³⁺/Dy³⁺-loaded spores before or at the time of addition of the germinant l-valine (see below).

FIG 3.

Germination of Tb³⁺-loaded spores of B. subtilis. Spores of B. subtilis PS832 were preincubated with Tb³⁺ (20 μM) in K-HEPES buffer (100 mM; pH 7.4) at 23°C for 5 min and then washed. The Tb³⁺-loaded and untreated spores were incubated at 37°C with 10 mM l-valine in K-HEPES buffer (25 mM; pH 7.4) and examined by phase-contrast microscopy every 20 min to distinguish between phase dark (fully germinated spores) and phase bright (dormant spores). These values are expressed as the ratio of the percentage of phase-dark spores at time t to total spores, as described in Materials and Methods.

Release of Tb³⁺/Dy³⁺ from spores.

Given that spores can accumulate large amounts of Tb³⁺/Dy³⁺, an obvious question is what is needed to remove these ions? We tested DPA alone or together with the germinant l-valine to elute these ions from Tb³⁺/Dy³⁺-loaded wild-type, severely coat defective, and CaDPA-less B. subtilis spores (Fig. 4). As expected, based on the extremely tight binding of Tb³⁺/Dy³⁺ to DPA (12, 25), when DPA was added to Tb³⁺/Dy³⁺-loaded spores, all of the Tb³⁺/Dy³⁺ was released almost immediately from the spores at either 23°C (data not shown) or 37°C (Fig. 4). There were no differences in the efficiency of the release of Tb³⁺/Dy³⁺ in 20 min using DPA alone or DPA plus l-valine, indicating that DPA may have promise for Tb³⁺/Dy³⁺ desorption from bacterial spores in applied settings. Although these rare earth ions bind DPA very tightly (12, 25), it seems likely that these ions would dissociate at low pH when DPA’s carboxyl groups and nitrogen become protonated. It was also notable that the desorption of Tb³⁺/Dy³⁺ from coat-defective spores was faster than from spores with intact coats (Fig. 4, compare panels A and C), suggesting that DPA more readily accesses Tb³⁺/Dy³⁺ in spores with coat defects.

FIG 4.

Tb³⁺ and Dy³⁺ release from Tb³⁺/Dy³⁺-loaded spores of B. subtilis strains. Spores of B. subtilis PS832 (A) and PS4150 (C) were initially incubated with Tb³⁺ (20 μM) and spores of PS832 (B) and FB122 (D) were initially incubated with Dy³⁺ (20 μM) in K-HEPES buffer (100 mM; pH 7.4) at 23°C for 5 min and washed. The fluorescence levels of the Tb³⁺/Dy³⁺-loaded spore suspensions at an OD₆₀₀ of 0.5 incubated at 37°C with 10 mM l-valine or 100 μM DPA alone or together in K-HEPES buffer (25 mM; pH 7.4) were measured, and these values are expressed as the ratio of fluorescence at time t to maximum fluorescence × 100%, as described in Materials and Methods.

l-Valine is an effective germinant for B. subtilis spores and has been extensively used to trigger germination receptor-dependent germination (26). Therefore, we also used l-valine to examine whether the DPA release triggered in spore germination also triggered the release of accumulated Tb³⁺/Dy³⁺ from spores (Fig. 4B and D). Our results showed that l-valine addition triggered germination and DPA release, which led to Tb³⁺/Dy³⁺ release from spores, although the rate of the Tb³⁺/Dy³⁺ release was lower than that obtained upon the addition of DPA alone (Fig. 4). However, l-valine addition alone did not cause Tb³⁺/Dy³⁺ release from coat-defective spores since these spores’ germination was very strongly inhibited, presumably by Tb³⁺/Dy³⁺-DPA complexes, as noted above (Fig. 4C). Obviously, due to the lack of CaDPA in FB122 spores, there was no DPA release from the spores with l-valine alone and thus no Tb³⁺/Dy³⁺ release (Fig. 4D).

Location of adsorbed Tb³⁺/Dy³⁺ in spores.

Given the results noted above, in particular the rapid Tb³⁺/Dy³⁺ uptake and release, it appeared likely that adsorbed Tb³⁺/Dy³⁺ were not in the spore core, but it was not clear where these ions were in spores’ more outer layers. To precisely determine the location of the adsorbed rare earth ions in spores, we used transmission electron microscopy (TEM) to localize the electron-dense Tb³⁺ ions. We used FB122 spores for this analysis, because preliminary TEM work visualized no Tb³⁺ in Tb³⁺-loaded wild-type B. subtilis spores (data not shown), most likely because of DPA release and thus the loss of Tb³⁺ during sectioning and sample processing for TEM (data not shown). TEM micrographs of sections of Tb³⁺-free FB122 spores clearly showed spores’ major layers, even without OsO₄ postfixation (Fig. 5A and C). However, compared to untreated spores, TEM micrographs of spores loaded with Tb³⁺ showed highly electron-dense areas all around the spores’ outer surfaces, which appeared to form an extended matrix (Fig. 5B and D, black arrows). The presence of this electron-dense matrix area plus some slightly increased electron density in the coat of the Tb³⁺-loaded spores (Fig. 5B and D, white arrows) indicated that Tb³⁺ was absorbed almost exclusively on the spore surface, consistent with the finding that adsorbed Tb³⁺ was rapidly and completely eluted by added DPA. As a further test of what outer spore components might be involved in Tb³⁺/Dy³⁺ binding, we also examined adsorption of these ions by B. subtilis spores lacking specific components of these spores’ outermost layer, the crust (27) (Table 3). The mutations removing crust components were in the PY79 background, so Tb³⁺/Dy³⁺ adsorption by the mutant spores was compared to that of wild-type PY79 spores, as well as PS533 spores. Except for slightly lower Dy³⁺ adsorption by the cge mutant spores, there was no significant decrease in Tb³⁺/Dy³⁺ adsorption by the crust mutants compared to the adsorption by the wild-type spores of either the PY79 or the PS533 background.

FIG 5.

TEM micrographs of spores of B. subtilis FB122 incubated with or without Tb³⁺. Untreated spores of B. subtilis FB122 (A and C) and B. subtilis FB122 spores incubated with TbCl₃ (20 μM) in K-HEPES buffer (100 mM; pH 7.4) at 23°C for 5 min (B and D) were washed and examined by TEM, as described in Materials and Methods. Panels C and D show higher magnifications of sections from the samples in panels A and B, respectively. Abbreviations for spore features: Ct, spore coat; Sm, spore surface matrix. White arrows indicate the spore coat, and black arrows denote the electron-dense areas where Tb³⁺ is adsorbed. Bars denote the scale, and micrographs in panels A and B are at the same magnification, as are those in panels C and D.

TABLE 3.

Tb³⁺ and Dy³⁺ uptake by spores of Bacillus strains with spore crust defects^a

Strain	Genotype-background	Avg uptake (%) ± SD
		Tb3+	Dy3+
B. subtilis PS533	WT-168	100	100
B. subtilis PS3483	WT-PY79	108 ± 21	92 ± 18
B. subtilis PE670	cotXYZ-PY79	82 ± 31	98 ± 15
B. subtilis PE2763	spsI-PY79	87 ± 22	58 ± 9
B. subtilis PE2916	cgeB-PY79	118 ± 21	88 ± 16

^aSpores of various strains of B. subtilis in 1 ml at an OD₆₀₀ of 2.0 were incubated at 23°C with 20 μM Tb³⁺ or Dy³⁺ in 25 mM K-HEPES buffer (pH 7.4) for 15 min, and the spores were centrifuged and washed as described in Materials and Methods. The Tb³⁺ and Dy³⁺ levels were determined from the levels of fluorescence when the washed spores were incubated with added DPA (Fig. 4) and calculated from the standard curves for fluorescence of Tb³⁺-DPA and Dy³⁺-DPA in Fig. 6. Data are expressed as the percentages of the PS533 level, which was set at 100%, and average values ± the standard deviations from two independent experiments are shown.

DISCUSSION

The finding in the current work that spores of several Bacillus species can adsorb Tb³⁺ or Dy³⁺ at up to ∼3% of the spore dry weight is an extension of work on rare earth ion adsorption by bacteria. TEM analysis of the location of Tb³⁺/Dy³⁺ ions adsorbed by B. subtilis spores clearly indicated that the adsorbed ions were in the spores’ outermost layers, and much evidence indicated that adsorbed Tb³⁺ and Dy³⁺ were not in the spores’ inner layers, in particular the spore core. This evidence included the following: (i) adsorption of maximum Tb³⁺/Dy³⁺ levels required ≤5 min at 23°C, whereas maximum uptake of molecules such as Li⁺, methylamine, and even H₂O into the spore core requires many hours (28–30); (ii) adsorbed Tb³⁺/Dy³⁺ ions were released almost instantaneously from B. subtilis spores upon addition of DPA and, as with uptake of molecules into the spore core, the release of core small molecules in the absence of spore germination is also extremely slow (20, 21, 31); (iii) a variety of alterations in the spore cortex peptidoglycan structure had no effects on Tb³⁺/Dy³⁺ adsorption by spores; and (iv) while Tb³⁺/Dy³⁺ ions bind extremely tightly to DPA, which makes up ∼20% of the spore core dry weight, and ∼10⁴ more tightly than does Ca²⁺, the cation normally bound to most if not all DPA in the spore core (12, 25, 32). Tb³⁺/Dy³⁺ uptake by spores that lack CaDPA was as high as, and perhaps even higher than, that observed with wild-type spores. Thus, all available data indicate that Tb³⁺/Dy³⁺ ions, as well as presumably other rare earth ions that spores might adsorb, are not in the spore core but on the spore’s outermost layer. It is also clear that there is some selectivity in divalent cation binding at the spore outer surface, with Dy³⁺ binding slightly better than Tb³⁺ and 10-fold excesses of Ca²⁺ or Mg⁺ decreasing Tb³⁺/Dy³⁺ binding only minimally (Ca²⁺) or not at all (Mg²⁺).

While binding of Tb³⁺/Dy³⁺ ions takes place in spore outer layers, events further inside the spore can influence this binding. In particular, spore germination, in which release of the spore core’s huge CaDPA depot is an essential early event, resulted in the release of all bound Tb³⁺/Dy³⁺. The reason for this seems most likely to be that the binding of Ca²⁺ with which DPA is associated in the spore core is much weaker than is binding of Tb³⁺/Dy³⁺ to DPA (12, 25). Thus, adsorbed spore Tb³⁺/Dy³⁺ displaces Ca²⁺ from CaDPA released in spore germination, and this results in release of the Tb³⁺/Dy³⁺ from the spore. Presumably, DPA association with Tb³⁺/Dy³⁺ is not only much stronger than with Ca²⁺, as noted above, but also much stronger than the association of Tb³⁺/Dy³⁺ to the spore component to which these ions bind. This scenario for the fate of Tb³⁺/Dy³⁺ adsorbed to spores is supported by the lack of release of Tb³⁺/Dy³⁺ from FB122 spores which contain no CaDPA, since these spores’ adsorbed Tb³⁺/Dy³⁺ was only released when DPA was added exogenously.

It is notable that spores made coat defective, either chemically or genetically, that had adsorbed Tb³⁺/Dy³⁺ did not germinate with l-valine, as assessed by the minimal Tb³⁺/Dy³⁺ release upon l-valine addition. This also indicated that there was no CaDPA release after l-valine addition, since if there had been significant CaDPA release, there would have been significant Tb³⁺/Dy³⁺-DPA fluorescence. Notably, previous work showed that a ≤10 μM concentration of either Tb³⁺ or Dy³⁺ strongly inhibits the l-valine germination of coat-defective spores (14). It was suggested that this inhibition was due to these Tb³⁺/Dy³⁺ ions’ permeation up to the spore inner membrane, binding to DPA as it passed through the CaDPA release channel in this membrane and thus plugging this CaDPA channel. However, the similar strong inhibition of germination of coat-defective spores to which Tb³⁺/Dy³⁺ ions were bound to spores’ outermost layer, with concentrations of free Tb³⁺/Dy³⁺ ions being minimal, suggests that there must be another explanation for the inhibition of the germination of coat-defective spores by Tb³⁺/Dy³⁺. More work will be required to determine the mechanism of the inhibition.

The final unknown about Tb³⁺/Dy³⁺ binding is the specific outer spore layer that is interacting with Tb³⁺/Dy³⁺. It is known that the outermost B. subtilis spore layer, termed the crust, contains both proteins and carbohydrate, giving a matrix outside the coat protein layer (27). However, the precise structure and composition of the crust layer are not known, and the spore crust would be expected to be greatly reduced in severely coat-defective spores such as PS4150, which lacks the coat morphogenic protein CotE, as well as the GerE transcription factor for many coat protein genes (19, 27). In principle, Tb³⁺/Dy³⁺ could bind to either carboxylate groups or phosphate groups on either a sugar or a protein backbone. Various types of teichoic acids containing phosphate groups in a sugar backbone are present in growing Bacillus cells, and in some cases these teichoic acids have been shown to be important in cation binding, including rare earth cations, by these cells (4, 7, 8, 33). However, teichoic acids are not present in Bacillus spores (11), although phosphoproteins in spores have been reported (17, 34). Notably, ∼50% of the acid-insoluble phosphate in B. subtilis spores is not in DNA or RNA, but in some unknown macromolecule (35). Perhaps this large amount of excess phosphate is in spores’ outer layers where, like the phosphate in teichoic acids, it can bind Tb³⁺/Dy³⁺ and perhaps other ions as well. Going forward, the identification of the outer spore component binding Tb³⁺/Dy³⁺ also seems an area for fruitful investigation, especially as one could envisage the use of spores for the biological mining of materials such as rare earth ions.

MATERIALS AND METHODS

Bacillus strains used and spore preparation.

The B. subtilis strains used were PS832, a wild-type laboratory derivative of B. subtilis 168 (36), and eight isogenic strains: (i) PS533, identical to PS832 but carrying plasmid pUB110 encoding resistance to kanamycin (10 μg/ml) (37); (ii) PS4150, in which the cotE and gerE genes are replaced with tetracycline and spectinomycin resistance cassettes, respectively (19), resulting in spores that lack both the inner and most of the outer coat layer, although a thin layer of insoluble coat material remains and is even present after these spores are digested with lysozyme, DNase, and pronase and boiled in sodium dodecyl sulfate (SDS) (19, 38); (iii) FB122, in which the spoVF operon and sleB gene have been deleted and replaced with spectinomycin and tetracycline resistance cassettes (this strain cannot make DPA in sporulation and its spores lack CaDPA, and although CaDPA-less spores normally rapidly germinate spontaneously, the sleB mutation stabilizes CaDPA-less spores) (20, 21); (iv) 2066 (22), with a deletion of the dacB gene, which encodes an enzyme that modifies the structure of spore cortex PG; (v) PS2421 (22), with deletions of the dacB and dacF genes, which both encode enzymes that modify cortex PG structure; (vi) PS2307 (23), which lacks the cwlD gene, which encodes an enzyme essential for generation of the spore cortex PG-specific modification, muramic acid–δ-lactam; (vii) PS2422 (24), with deletions in both cwlD and dacB genes; and (viii) PS3738 (18), with a deletion of the safA gene, which encodes a protein important in spore coat assembly (16, 38). In addition, four isogenic B. subtilis strains were used with the PY79 genetic background, including PS3483 (the wild-type strain), PE620 lacking the cotXYZ operon, PE2763 lacking spsI, and PE2916 lacking cgeB. The latter three strains were obtained from Patrick Eichenberger, and each lacks various components of the spores’ outermost crust layer (27). The B. cereus strain T was originally obtained from H. O. Halvorson.

B. subtilis spores were prepared on 2× Schaeffer’s glucose medium agar plates that were incubated at 37°C for 5 days, and spores were harvested and purified as described previously (31, 32). B. cereus spores were prepared in liquid defined sporulation medium at 30°C for 2 days and purified as described previously (39). All purified spores were >98% free of sporulating cells, germinated spores, and debris, as observed by phase-contrast microscopy, and were stored protected from light in deionized double-distilled water at 4°C. Spores were chemically decoated by incubation for 2 h at 70°C in decoating solution (100 mM NaCl, 100 mM NaOH, 10 g/liter SDS, 100 mM dithiothreitol), washed extensively with cold water, and used soon after washing (40). This decoating procedure removes all spore outer membrane proteins and almost certainly all outer membrane phospholipids as well (41, 42).

Measurements of rare earth ion accumulation by spores.

Purified spores were incubated at 23°C in terbium chloride hexahydrate (99.9% TbCl₃·6H₂O; Aldrich) or dysprosium chloride hexahydrate (99.9% DyCl₃·6H₂O; Aldrich) in 25 mM buffers to measure Tb³⁺ or Dy³⁺ accumulation as a function of exposure pH, time, and Tb³⁺/Dy³⁺ concentration. The buffers used at each pH were as described previously (43): homopiperazine−N,N-bis-2(ethanesulfonic acid) (Homopipes), pH 4.5; 2-(N-morpholino)ethanesulfonic acid (MES), pH 5.5 and 6.5; and HEPES, pH 7.4 and 8.0. Aliquots of 1 ml of purified spores at an optical density at 600 nm (OD₆₀₀) of 2.0 (∼0.25 mg [dry weight]/ml) were centrifuged at 14,000 rpm for 1 min in a microcentrifuge, and the pellets were suspended in 1 ml of various concentrations of TbCl₃ or DyCl₃ plus 100 mM buffers. After various incubations, spores were washed three or four times with water by centrifugation in a microcentrifuge, and the washed spore pellet was suspended in 1 ml of deionized double-distilled water. Control experiments showed that the final supernatant fluid contained ≤0.1 or 0.4 μM Tb³⁺ or Dy³⁺, respectively (data not shown). Then, 0.5-ml portions of washed spore suspensions at an OD₆₀₀ of 0.5 were boiled for 30 min, incubated on ice for 15 min, and centrifuged for 1 min at 14,000 rpm in a microcentrifuge, and the supernatant fluid was collected. The levels of Tb³⁺ and Dy³⁺ in the supernatant fluid were determined by mixing 90 μl of supernatant fluid with 10 μl of distilled water or 1 mM DPA plus 100 μl of 50 mM K-HEPES buffer (pH 7.4) in a well of a multiwell fluorescence plate (Thermo Fisher Scientific, Waltham, MA). The fluorescence of Tb³⁺-DPA or Dy³⁺-DPA was measured in a Gemini EM multiwell fluorescence plate reader (Molecular Devices, Sunnyvale, CA) with emission at either 545 nm (Tb³⁺-DPA) or 480 nm (Dy³⁺-DPA) and excitation at 270 nm. These were appropriate wavelengths for maximum fluorescence, as described for Tb³⁺-DPA (44) and measured for Dy³⁺-DPA. The concentrations of the two rare earth ions were calculated from measured fluorescence intensities and calibration curves for Tb³⁺-DPA and Dy³⁺-DPA (Fig. 6). To examine the effects of other metal ions on Tb³⁺/Dy³⁺ uptake by spores, wild-type B. subtilis PS832 spores in 1 ml at an OD₆₀₀ of 2.0 with TbCl₃ or DyCl₃ (20 μM) alone or plus CaCl₂ or MgCl₂ (200 μM) were incubated in 25 mM K-HEPES buffer (pH 7.4) for 5 min at 23°C. The spores were washed, boiled for 30 min, and centrifuged, and the levels of Tb³⁺ and Dy³⁺ in supernatant fluids were determined as described above. Unless indicated otherwise, duplicate analyses were carried out in all experiments, and data were assessed by analysis of variance.

FIG 6.

Standard curves of the fluorescence of Tb³⁺-DPA and Dy³⁺-DPA complexes with different amounts of Tb³⁺ and Dy³⁺. Tb³⁺ or Dy³⁺ at various concentrations was added in 90-μl volumes to 10 μl of DPA (1 mM) plus 100 μl of K-HEPES buffer (50 mM; pH 7.4), and fluorescence was measured as described in Materials and Methods.

Properties of spores loaded with Tb³⁺ or Dy³⁺.

For assays of the effects of Tb³⁺ or Dy³⁺ accumulation on spore heat resistance, spores were first incubated in 20 μM TbCl₃ or DyCl₃ (pH 7.4) for 5 min and then washed, as described above. The washed spores and spores that were treated similarly, but with no Tb³⁺/Dy³⁺ exposure, were incubated in sterile water at an OD₆₀₀ of 0.5 at 90°C. At various times, aliquots were serially diluted in water, and 10-μl portions of dilutions were spotted in duplicate on L broth agar plates (tryptone, 10 g/liter; yeast extract, 5 g/liter; NaCl, 10 g/liter; agar, 15 g/liter). The plates were incubated at 37°C overnight or longer, and the colonies were counted until there were no further increases in colony numbers.

To test the effect of Tb³⁺/Dy³⁺ accumulation on spore germination, untreated spores and spores loaded with either ion as described above were incubated at an OD₆₀₀ of 0.5 in 200 μl of 50 mM K-HEPES (pH 7.4) at 37°C. Germination was initiated by the addition of l-valine to 10 mM, and spore germination was measured by examining 100 to 200 individual spores by phase-contrast microscopy. The extent of complete spore germination at various times was expressed as the ratio of numbers of phase-dark spores that have completed germination to the total numbers of spores × 100%. l-Valine was used in the germination experiments instead of the more commonly used l-alanine to eliminate complications in assessing spore germination due to the conversion of l-alanine to d-alanine by a racemase in spores’ outer layers, since d-alanine is an extremely strong inhibitor of germination by the germinant receptor that recognizes either l-alanine or l-valine (26).

Kinetics of Tb³⁺/Dy³⁺ release from spores.

For analysis of release of Tb³⁺/Dy³⁺ from spores, unless indicated otherwise, spores were preloaded with 20 μM TbCl₃ or DyCl₃ at pH 7.4 for 5 min and then washed and centrifuged as described above. Spore pellets were suspended at an OD₆₀₀ of 0.5 in 25 mM K-HEPES buffer (pH 7.4) and then incubated with 50 μM DPA or 10 mM l-valine at 37°C. Measurement of Tb³⁺/Dy³⁺ release from spores was again carried out in a 96-well plate in a fluorescence plate reader, and Tb³⁺-DPA or Dy³⁺-DPA fluorescence was measured in real time as described above.

Localization of adsorbed Tb³⁺/Dy³⁺ in spores.

In order to determine where adsorbed rare earth ions were localized in spores, TEM was carried out using a previously described method (45) with minor modifications. Spores of PS832 and FB122, either with or without preloading with 20 μM TbCl₃ (as described above) and washing, were fixed with 2.5% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.2) for 1 h at 23°C and then rinsed with 0.1 M cacodylate buffer. After the rinse, the samples were dehydrated through a graded ethanol series and infiltrated with 1:1 resin (Poly/Bed 812; Polysciences, Inc., Warrington, PA)/ethanol for 30 min, 3:1 resin/ethanol for 1 h, and then in resin overnight. The resin was polymerized for 48 h at 60°C, and 70- to 80-nm sections through the embedded spores were cut with a diamond knife in a Leica Ultracut UCT ultramicrotome under a Leica S6E stereomicroscope. Individual sections were examined in a Hitachi H-7650 transmission electron microscope operated at 80 kV under standard conditions.