Efficient Inhibition of Germination of Coat-Deficient Bacterial Spores by Multivalent Metal Cations, Including Terbium (Tb³⁺)

Abstract

Release of dipicolinic acid (DPA) and its fluorescence with terbium (Tb³⁺) allow rapid measurement of the germination and viability of spores of Bacillus and Clostridium species. However, germination of coat-deficient Bacillus spores was strongly inhibited by Tb³⁺ and some other multivalent cations. Tb³⁺ also inhibited germination of coat-deficient Clostridium perfringens spores.

TEXT

Many Bacillus and Clostridium species can form dormant, resistant spores under conditions unfavorable for growth (24, 25). These spores can resist extreme conditions and survive for years in the absence of nutrients. However, when nutrients return, these spores can germinate and become growing cells (16, 23, 25). Some spore-forming bacteria can also cause food spoilage, food poisoning, and disease, and spores of Bacillus anthracis are a potential bioterrorism agent (9, 13, 25). Therefore, rapid, sensitive assays for bacterial spores have applied importance.

An assay has been developed to detect low spore concentrations, based on spores' high concentration (∼10% [dry weight]) of pyridine-2,6-dicarboxylic acid (dipicolinic acid [DPA]) (9). DPA forms a strong 1:1 highly fluorescent complex with terbium (Tb³⁺) ions, and spore quantification by Tb³⁺-DPA fluorescence takes only a few minutes (4, 12). DPA is also released in the first minutes of spore germination (23), and since germination is essential for spore viability, Tb³⁺-DPA fluorescence has been used to detect spore germination, and there is a good correlation between DPA release and spore viability (4, 26, 28, 29, 31, 32). However, we report a flaw in the latter assay of spore viability, as Tb³⁺ and some other multivalent metal cations strongly inhibit germination of coat-deficient spores. The coats comprise the outermost layer of spores of many species, including Bacillus subtilis, and protect spores against many agents, including enzymes and reactive chemicals (8, 10, 11, 24).

The B. subtilis strains used are isogenic derivatives of a laboratory 168 strain and are (i) PS533 (wild type), carrying plasmid pUB110, encoding resistance to kanamycin (22); (ii) PS3328 (ΔcotE mutant), in which much of the cotE coding sequence was replaced with a tetracycline resistance (Tc^r) cassette (15); and (iii) PS3738 (ΔsafA mutant), which carries a Tc^r cassette replacing much of the safA coding sequence (10). These strains were sporulated at 37°C on 2× Schaeffer's-glucose (SG) agar plates without antibiotics, and spores were harvested and cleaned as described previously (14). The Bacillus cereus strain used was strain T obtained from H. O. Halvorson, and its spores were prepared at 30°C in a defined liquid medium and were purified as described previously (3). The Bacillus megaterium strain used was strain ATCC 12782, and its spores were prepared at 30°C in liquid supplemented nutrient broth and were harvested and cleaned as described previously (5). The Clostridium perfringens strain used was MRS101 (Δcpe::catP mutant) in which the enterotoxin gene (cpe) in the C. perfringens food poisoning isolate SM101 has been inactivated by insertion of a chloramphenicol resistance gene (18). The C. perfringens spores were prepared and purified as described previously (17, 18). All spores used in this work were free (>98%) of growing or sporulating cells, germinated spores, and cell debris, as determined by phase-contrast microscopy. For chemical decoating, spores at an optical density at 600 nm (OD₆₀₀) of ∼25 were incubated at 70°C for 2 h in 1% SDS-0.1 M NaOH-0.1 M NaCl-0.1 M dithiothreitol, and decoated spores were washed and stored as described previously (1).

Prior to germination, spores at an OD₆₀₀ of ∼10 were heat activated at 75°C for 30 min (B. subtilis), 70°C for 20 min (B. cereus), or 60°C for 15 min (B. megaterium); activated spores were cooled on ice before germination. Spores were routinely germinated at an OD₆₀₀ of 0.5 at 37°C in a 96-well plate in 200 μl of 25 mM K-HEPES buffer (pH 7.4) (B. subtilis and C. perfringens), 25 mM Tris-HCl buffer (pH 8.8) (B. cereus), or 25 mM Tris-HCl buffer (pH 7.4) (B. megaterium), with or without 50 μM TbCl₃ (30). Spore germination was initiated by the addition of germinants to final concentrations of 2 mM l-valine (B. subtilis) (valine was used instead of l-alanine to preclude germination inhibition by d-alanine generated by spore-associated alanine racemase [23, 25]), 40 mM NH₄Cl plus 10 mM l-alanine (B. cereus), or 10 mM glucose (B. megaterium). Dodecylamine germination did not require heat activation and was carried out at 45°C in 40 mM dodecylamine with 25 mM K-HEPES buffer, pH 7.4 (19). For wild-type spores, germination was measured in the presence of 50 μM TbCl₃ by quantifying DPA release in a fluorescence plate reader by measuring the formation of Tb³⁺-DPA (30). To measure germination of coat-deficient spores, 50 μM TbCl₃ was added to germination incubations at various times and the fluorescence intensity was recorded. However, to measure whether Tb³⁺ inhibited DPA release from coat-deficient spores, 10 μM TbCl₃ was added to germination incubations unless otherwise specified. The percentages of germinated spores at the end of incubations were checked by examination of ≥100 spores by phase-contrast microscopy. All germination experiments were repeated at least twice with similar results (less than or equal to ±10% of all values reported).

To evaluate the effects of Tb³⁺ on germination of coat-deficient spores, the valine germination of intact and chemically decoated wild-type B. subtilis spores was measured with or without Tb³⁺. To measure germination plus Tb³⁺, TbCl₃ was added at the beginning of germination, and DPA release was determined by continuously monitoring Tb³⁺-DPA fluorescence. To measure spore germination without Tb³⁺, TbCl₃ was added at various times during germination, and the fluorescence immediately after TbCl₃ addition was considered the DPA released by that time point. Intact wild-type B. subtilis spores released DPA similarly with or without 50 μM Tb³⁺ (Fig. 1A), and ≥95% of these spores germinated after 2.5 h, as determined by phase-contrast microscopy. Without Tb³⁺, decoated B. subtilis spores released DPA slower with l-valine than did intact spores (Fig. 1A), but 85% of the decoated spores germinated in 2.5 h, as determined by phase-contrast microscopy. However, with 10 μM Tb³⁺ present continuously, no DPA release was detected from decoated spores incubated with l-valine (Fig. 1A), and ≥99% of these spores remained phase bright after 2.5 h (data not shown).

Fig. 1.

Effect of Tb³⁺ on DPA release during spore germination. DPA release during germination of intact (○, •) or chemically decoated (▿, ▾) spores in the presence (•, ▾) or absence (○, ▿) of TbCl₃ was measured as described in the text. In panels A, B, and C, the TbCl₃ concentration was either 50 μM (•) or 10 μM (▾), and in panel D TbCl₃ was at 50 μM (•, ▾). (A) B. subtilis PS533 (wild-type) spores germinating in 2 mM l-valine; (B) B. subtilis PS533 (wild-type) spores germinating in 40 mM dodecylamine; (C) B. cereus spores germinating in 40 mM NH₄Cl plus 10 mM l-alanine; and (D) B. megaterium spores germinating in 10 mM d-glucose. With intact spores germinating with nutrient germinants, the amounts of fluorescence in the TbCl₃ assay were similar with 10 μM and 50 μM TbCl₃ present throughout the assay (data not shown).

We also investigated if Tb³⁺ interfered with germination triggered by the cationic surfactant dodecylamine that triggers germination independent of spores' nutrient germinant receptors (GRs), in contrast to germination by l-valine that is GR dependent (19, 23). Intact wild-type B. subtilis spores germinated similarly with dodecylamine and with or without Tb³⁺ and released most of their DPA within 3 h (Fig. 1B). Without Tb³⁺ present continuously, dodecylamine germination of decoated wild-type B. subtilis spores was like that of intact spores (Fig. 1B), consistent with previous work (19). However, 10 μM Tb³⁺ present continuously inhibited the dodecylamine germination of decoated wild-type B. subtilis spores.

The effects of Tb³⁺ on the germination of spores of other species were also examined. Intact B. cereus and B. megaterium spores incubated with NH₄Cl plus l-alanine and glucose, respectively, released almost all DPA within 10 to 30 min with or without Tb³⁺ (Fig. 1C and D). Without Tb³⁺, decoated B. cereus and B. megaterium spores germinated normally (B. megaterium) or germinated somewhat slower (B. cereus) than intact spores. However, 10 and 50 μM Tb³⁺ completely inhibited DPA release from decoated B. cereus and B. megaterium spores, respectively (Fig. 1C and D); this inhibition of germination was confirmed by phase-contrast microscopy (data not shown). We also tested effects of Tb³⁺ on the dodecylamine germination of intact and decoated C. perfringens spores (Table 1). The effects of Tb³⁺ on germination by KCl, a particularly effective germinant for C. perfringens MRS101 spores (17), were not tested, because KCl germination of these decoated spores was extremely slow (data not shown); however, dodecylamine germination of these decoated spores was relatively normal. Tb³⁺ gave only minimal inhibition of the dodecylamine germination of intact C. perfringens spores but inhibited dodecylamine germination of the decoated spores, albeit not as strongly as decoated B. subtilis spores (Fig. 1B; Table 1).

Table 1.

Effect of various Tb³⁺ concentrations on the dodecylamine germination of intact and decoated C. perfringens spores^a

Tb3+ concn (μM)	% spore germination in 2.5 h
	Intact spores	Decoated spores
0	100	100
50	100	56
500	92	32
2,500	83	21

^aIntact and chemically decoated spores of C. perfringens were germinated with dodecylamine plus various TbCl₃ concentrations, and the extent of germination was assessed by monitoring DPA release fluorometrically as described in the text. Use of 2.5 mM TbCl₃ in the fluorometric DPA assay did not alter the fluorescence compared to that with the standard 50 μM TbCl₃ normally present.

Since Tb³⁺ strongly inhibited the germination of at least Bacillus spores that had been chemically decoated, we then asked if Tb³⁺ inhibited the germination of spores that were coat deficient due to specific mutations. Spores of B. subtilis were used that lacked either CotE or SafA, morphogenic proteins essential for the proper spore coat assembly and for coats' protection against enzymes and reactive chemicals (8, 10). As expected, there was no significant effect on valine germination of intact wild-type B. subtilis spores with ≤200 μM Tb³⁺ (Table 2). However, valine germination of chemically decoated wild-type spores and of intact ΔcotE and ΔsafA spores was completely inhibited by 10 μM Tb³⁺ (Table 2).

Table 2.

Effect of various Tb³⁺ concentrations on the valine germination of intact and coat-deficient B. subtilis spores^a

Spore type used	% spore germination in 2.5 h at a Tb3+ concentration of:
	0	500 nM	2 μM	10 μM	50 μM	200 μM
Intact wild type	96	92	96	94	96	95
Decoated wild type	80	80	84	0	0	0
ΔcotE mutant	76	68	78	0	0	0
ΔsafA mutant	78	86	78	0	0	0

^aSpores of various B. subtilis strains, including intact and chemically decoated wild-type (PS533) spores, were germinated for 2.5 h at 37°C with 2 mM l-valine plus various TbCl₃ concentrations, and the degree of spore germination was determined by phase-contrast microscopy as described in the text.

Other multivalent metal cations were also examined for their ability to inhibit germination of coat-deficient spores (Table 3). With intact wild-type B. subtilis spores, all cations tested had no significant effect on germination except for 5 mM Zn²⁺ (Table 3). However, with decoated spores, two lanthanide cations completely inhibited germination at all concentrations tested, and Zn²⁺ had a similar inhibitory effect, although Mg²⁺ and Ca²⁺ were relatively ineffective.

Table 3.

Effects of various multivalent cations on the valine germination of intact and decoated wild-type B. subtilis spores^a

Cation	% spore germination in:
	Intact spores at cation concn of:					Decoated spores at cation concn of:
	0	50 μM	500 μM	2.5 mM	5 mM	0	50 μM	500 μM	2.5 mM	5 mM
Gd3+	96	92	94	84		85	0	0	0
Dy3+	96	92	94	96		84	0	0	0
Ca2+	94	88	92		88	82	72	64		60
Mg2+	94	88	90		82	82	76	64		48
Zn2+	94	90	88		20	82	0	0		0

^aIntact and chemically decoated wild-type B. subtilis spores (PS533) were germinated for 2.5 h at 37°C with 2 mM l-valine plus various concentrations of divalent and trivalent cation chlorides, and the percentages of germinated spores were determined by phase-contrast microscopy as described in the text.

This communication shows that low concentrations of Tb³⁺ and some other multivalent metal ions strongly inhibit the germination of coat-deficient spores. This finding is notable, as these same low Tb³⁺ concentrations are used to assay viable spores by measuring DPA release in spore germination (26, 28, 29, 31, 32). Since coat-deficient spores have viability comparable to that of intact spores (8, 15), Tb³⁺-based assays of spore viability could yield false-negative results with coat-deficient spores. Consequently, the Tb³⁺ inhibition of spore germination may be significant when Tb³⁺-based assays are used to quantitate viable spores in either old environmental samples or by the food industry.

While the Tb³⁺ inhibition of germination of coat-deficient spores is of applied importance, the mechanism of this inhibition is not clear. One possibility is that spore coat defects facilitate access of cations such as Tb³⁺ to inner spore layers not accessible in an intact spore, and cation binding to some now accessible target inhibits germination. One such target could be the GRs in spores' inner membranes, and GR inhibition could abolish nutrient germination (16, 23). However, while Tb³⁺ and similar cations might inhibit GR function, this cannot explain these cations' effects on spore germination, since Tb³⁺ also inhibits dodecylamine germination, and dodecylamine germination is GR independent (19). Other possible targets of Tb³⁺ are the cortex-lytic enzymes (CLEs) that hydrolyze spores' peptidoglycan cortexes, a process essential to complete spore germination. However, while Tb³⁺ and similar multivalent cations may inhibit CLEs, such inhibition again cannot explain such cations' inhibition of DPA release during germination, because spores of Bacillus species lacking CLEs release DPA completely during germination, albeit slower than wild-type spores, and dodecylamine germination is completely CLE independent (7, 19, 20, 21). Since effects of Tb³⁺ and similar cations on spores' GRs or CLEs cannot explain these cations' inhibition of DPA release during spore germination, the only other likely target is the channel through which DPA is released during germination. This DPA channel has not been definitively identified but seems likely to contain proteins encoded by the spoVA operon, and if such channels were blocked, this would block not only DPA release during germination but also almost certainly subsequent germination events (16, 23, 27). How Tb³⁺ and similar cations might inhibit such DPA channels is not clear, but it is interesting that the efficacy of the various cations' inhibition of the germination of coat-deficient spores of Bacillus species parallels the values of the logarithms of the stability constants for M^2/3+-DPA complexes (where M is any metal), and the stability constants for the formation of multi-M^2/3+ ion-DPA complexes also have this same relative order (2, 6) (data not shown). These observations suggest that DPA channels may be blocked by formation of M^2/3+ complexes that act like a cork to plug the DPA channel. While this mechanism is certainly speculative, and Tb³⁺ was not as effective in blocking C. perfringens spore germination, this suggestion seems worth further study.