Characterization of Wet-Heat Inactivation of Single Spores of Bacillus Species by Dual-Trap Raman Spectroscopy and Elastic Light Scattering

Abstract

Dual-trap laser tweezers Raman spectroscopy (LTRS) and elastic light scattering (ELS) were used to investigate dynamic processes during high-temperature treatment of individual spores of Bacillus cereus, Bacillus megaterium, and Bacillus subtilis in water. Major conclusions from these studies included the following. (i) After spores of all three species were added to water at 80 to 90°C, the level of the 1:1 complex of Ca²⁺ and dipicolinic acid (CaDPA; ∼25% of the dry weight of the spore core) in individual spores remained relatively constant during a highly variable lag time (T_lag), and then CaDPA was released within 1 to 2 min. (ii) The T_lag values prior to rapid CaDPA release and thus the times for wet-heat killing of individual spores of all three species were very heterogeneous. (iii) The heterogeneity in kinetics of wet-heat killing of individual spores was not due to differences in the microscopic physical environments during heat treatment. (iv) During the wet-heat treatment of spores of all three species, spore protein denaturation largely but not completely accompanied rapid CaDPA release, as some changes in protein structure preceded rapid CaDPA release. (v) Changes in the ELS from individual spores of all three species were strongly correlated with the release of CaDPA. The ELS intensities of B. cereus and B. megaterium spores decreased gradually and reached minima at T₁ when ∼80% of spore CaDPA was released, then increased rapidly until T₂ when full CaDPA release was complete, and then remained nearly constant. The ELS intensity of B. subtilis spores showed similar features, although the intensity changed minimally, if at all, prior to T₁. (vi) Carotenoids in B. megaterium spores' inner membranes exhibited two changes during heat treatment. First, the carotenoid's two Raman bands at 1,155 and 1,516 cm⁻¹ decreased rapidly to a low value and to zero, respectively, well before T_lag, and then the residual 1,155-cm⁻¹ band disappeared, in parallel with the rapid CaDPA release beginning at T_lag.

Bacterial spores of Bacillus species are formed in sporulation and are metabolically dormant and extremely resistant to a variety of harsh conditions, including heat, radiation, and many toxic chemicals (37). Since spores of these species are generally present in foodstuffs and cause food spoilage and food-borne disease (37, 38), there has long been interest in the mechanisms of both spore resistance and spore killing, especially for wet heat, the agent most commonly used to kill spores. The killing of dormant spores by wet heat generally requires temperatures about 40°C higher than those for the killing of growing cells of the same strain (37, 43). A number of factors influence spore wet-heat resistance, with a major factor being the spore core's water content, as spores with higher core water content are less wet-heat resistant than are spores with lower core water (15, 25). The high level of pyridine-2,6-dicarboxylic acid (dipicolinic acid [DPA]) and the types of its associated divalent cations, predominantly Ca²⁺, that comprise ∼25% of the dry weight of the core also contribute to spore wet-heat resistance, although how low core water content and CaDPA protect spores against wet heat is not known. The protection of spore DNA against depurination by its saturation with a group of α/β-type small, acid-soluble spore proteins also contributes to spore wet-heat resistance (14, 23, 33, 37).

Despite knowledge of a number of factors important in spore wet-heat resistance, the mechanism for wet-heat killing of spores is not known. Wet heat does not kill spores by DNA damage or oxidative damage (35, 37). Instead, spore killing by this agent is associated with protein denaturation and enzyme inactivation (2, 7, 44), although specific proteins for which damage causes spore death have not been identified. Wet-heat treatment also often results in the release of the spore core's large depot of CaDPA. The mechanism for this CaDPA release is not known but is presumably associated with the rupture of the spore's inner membrane (7). In addition, the relationship between protein denaturation and CaDPA release is not clear, although recent work suggests that significant protein denaturation can occur prior to CaDPA release (7). Almost all information on spore killing by moist heat has been obtained with spore populations, and essentially nothing is known about the behavior of individual spores exposed to potentially lethal temperatures in water. Given the likely heterogeneity of spores in populations, in particular in their wet-heat resistances (16, 18, 39, 40), it could be most informative to analyze the behavior of individual spores exposed to high temperatures in water.

Raman spectroscopy is widely used in biochemical studies, as this technique has high sensitivity and responds rapidly to subtle changes in molecule structure (1, 22, 31). In addition, when Raman spectroscopy is combined with confocal microscopy and optical tweezers, the resultant laser tweezers Raman spectroscopy (LTRS) allows the nondestructive, noninvasive detection of biochemical processes at the single-cell level (9, 10, 19, 46). Indeed, LTRS has been used to analyze the DPA level and the germination of individual Bacillus spores (5, 19, 30). In order to obtain information more rapidly, dual- and multitrap laser tweezers have been developed to allow multiple individual cells or particles to be analyzed simultaneously (11, 13, 24, 27), and the dual trap has been used to measure the hydrodynamic cross-correlations of two particles (24). In addition to Raman scattering, the elastic light scattering (ELS) from trapped individual cells also provides valuable information on cell shape, orientation, refractive index, and morphology (12, 45) and has been used to monitor spore germination dynamics as well (30).

In this work, we report studies of wet-heat treatment of individual spores of three different Bacillus species by dual-trap LTRS and ELS. A number of important processes related to wet-heat inactivation of spores, including CaDPA release and protein denaturation, and the correlation between these processes were investigated by monitoring changes in Raman scattering at CaDPA-, protein structure-, and phenylalanine-specific bands and changes in ELS intensity.

MATERIALS AND METHODS

Bacillus strains, spore preparation, and storage.

The Bacillus strains used in this work were B. cereus T (originally obtained from H. O. Halvorson), B. megaterium QM B1551 (originally obtained from H. S. Levinson), and two isogenic B. subtilis strains, PS832 (originally obtained from D. J. Tipper) and PS533 (34). Both are prototrophic derivatives of strain 168. B. subtilis PS533 is PS832 that carries plasmid pUB110, which encodes resistance to kanamycin (10 μg/ml). Spores of these strains were prepared as follows: B. cereus spores at 30°C in liquid-defined sporulation medium (6); B. megaterium spores at 30°C in liquid-supplemented nutrient broth medium (17); and B. subtilis spores at 37°C on 2× SG medium agar plates without antibiotics (29). After harvesting, the spores were suspended in cold water, purified by repeated centrifugation and resuspension in cold water, and stored at 4°C in water protected from light. All spore preparations were free (>98%) of growing or sporulating cells and germinated spores as determined by phase-contrast microscopy.

Dual-trap optical tweezers and Raman spectroscopy.

The dual-trap optical tweezers system used in this work was constructed from a previous LTRS system (47). The laser beam emitted from a laser diode at 785 nm was divided into two orthogonally polarized beams by a polarized beam splitter (PBS). The two beams were recombined by another PBS, reflected by a Raman edge long-wave pass LP01-808U-25 filter (Semrock Corporation, Rochester, NY), converted into circular polarized beams by a quarter-wave plate, and then introduced into an inverted Nikon TE2000-S microscope (Nikon Instruments, Melville, NY). The two laser beams were introduced into a Nikon 100× (numerical aperture [NA] of 1.3) objective with different angles at the rear aperture and formed two optical traps separated by ∼3 μm on the focal plane, as shown schematically in Fig. 1a. Two spores suspended in the microscope sample holder were confined in the dual traps about 10 μm above the coverslip (Fig. 1b). The backscattering Raman light excited by the trapping lasers was collected by the same objective, returned, passed through the long-wave pass filter, and then focused onto the entrance slit of a Jobin Yvon Triax 320 spectrograph (Horiba Jobin Yvon Inc., Edison, NJ) equipped with a nitrogen-cooled charge-coupled detector (CCD). Inside the spectrograph, the Raman scattering light from the two trapped spores was dispersed by a grating and imaged onto different rows of the nitrogen-cooled CCD so that the spectra could be acquired simultaneously. The Raman spectra were recorded from 600 to 1,800 cm⁻¹, with a spectral resolution of ∼6 cm⁻¹. A background spectrum without spores in the traps was recorded under the same conditions and subtracted from the spectra of individual spores.

FIG. 1.

Schematic diagram of dual-trap optical tweezers (a) and the bright-field images of two trapped spores separated by ∼3 μm (b).

Elastic light scattering.

The backward ELS from the two trapped spores passed though a beam splitter and a long-wave pass filter and reflected partially toward the side port of the microscope, to which a CCD camera (QSI 520; Quantum Scientific Imaging, Dana Point, CA) was mounted. This scattered light was imaged onto two different regions (50 pixels by 50 pixels) on the CCD chip that were separated by 50 pixels from center to center, and the interference between the scattering light was very small. The counts of image data then were summed over all pixels in the corresponding region giving the ELS intensity from each spore. An optical attenuator was necessary to avoid saturation of the CCD. A background image was taken under the same conditions without spores in the traps and subtracted from the images of individual spores to reduce the influence of scattering light from the interface between the water and the coverslip.

Wet-heat treatment of bacterial spores.

Spore wet-heat treatment was in distilled water at 90°C (B. subtilis) or 80°C (B. cereus and B. megaterium) for 20 to 30 min. The sample holder for the dual-trap LTRS system was filled with water (∼600 μl), preheated, and kept at 90°C or 80°C using a temperature-stabilized controller, and an aliquot of spores (∼20 μl of 4°C water; ∼2 × 10⁶ spores/ml) was injected into the sample chamber. The volume of the sample chamber was big enough (∼600 μl) such that the addition of spore samples had only a minimal influence on the temperature of the water in the sample holder.

Measurement of dynamics of individual spores during wet-heat treatment.

After the addition of spore samples to the sample holder as described above, a pair of spores was captured in the dual traps in ∼1.5 min (defined as T₀), and measurements were begun at T₀. The laser power of each trap used in this work was 7.5 mW, and Raman spectra of the trapped spores were recorded, with an acquisition time of 30 s. The ELS intensities were acquired simultaneously by the CCD camera, with an exposure time of 20 ms and an acquisition interval (resolution) of 1 s. The dynamics during spore wet-heat treatment were monitored by Raman spectroscopy and ELS for 20 to 30 min. After this procedure was completed, a new spore sample from the same preparation was loaded, and the measurement was repeated. For each Bacillus strain analyzed, 20 individual spores were randomly trapped and investigated.

The CaDPA level in individual spores was determined from the Raman scattering intensities at 1,017 cm⁻¹ (averaged over 7 adjacent spectral data points), as described previously (19). The status of proteins in spores was monitored by the intensities of Raman spectral bands at 1,004 and 1,655 cm⁻¹ (averaged over 7 adjacent data points), again as described previously (7). For B. megaterium spores that had significant carotenoid levels (26, 32), the intensities of peaks at 1,155 and 1,516 cm⁻¹ were also calculated (averaged over 11 adjacent data points) to analyze carotenoid status during wet-heat treatment. For convenience of description, the Raman scattering intensities at all these bands were normalized to the initial values at T₀. The ESL data were smoothed by five-point averaging and also normalized to the initial value for the same reason.

RESULTS

Wet-heat treatment of B. cereus spore populations.

Preliminary work with our spore populations showed that incubation of B. subtilis spores in 90°C water gave 90 to 98% killing in ∼25 min; similar results were obtained when B. cereus and B. megaterium spore preparations were incubated in 80°C water (data not shown). Figure 2 shows the average Raman spectra of 20 B. cereus spores from an untreated population (Fig. 2, curve a) and a population heat treated in water at 80°C for 20 min (Fig. 2, curve b). The bands at 658, 824, 1,017, 1,395, and 1,572 cm⁻¹ are from CaDPA (19, 21); the band at 1,004 cm⁻¹ is due to phenylalanine in spore protein; and the bands at 1,655 and 1,668 cm⁻¹ are associated with α-helical and irregular structures of protein amide I (peptide bond C=O stretch), respectively (20). The Raman spectra also had a weak band at 1,155 cm⁻¹ that corresponds to the C-C stretch vibration of carotenoids that is a prominent band in B. megaterium spores (3, 26, 32). The results shown in Fig. 2 indicated that the CaDPA-specific Raman bands disappeared after the B. cereus spores were incubated for 20 min at 80°C. Although spores killed by wet heat can retain CaDPA, the release of CaDPA would almost certainly result in rapid spore inactivation via the replacement of CaDPA in the spore core by water and the resultant rapid core protein denaturation (7). The weak band at 1,155 cm⁻¹ also disappeared, and the intensity of the band at 1,004 cm⁻¹ was reduced, largely in parallel with CaDPA release. In addition, the intensity of the Raman band at 1,655 cm⁻¹ was reduced, and the band at 1,668 cm⁻¹ became more prominent after heat treatment, indicating that much of the spore proteins' structure had changed from an α-helical to an irregular and presumably denatured state. Similar results have been obtained for individual B. subtilis spores in populations (7).

FIG. 2.

Average Raman spectra of 20 individual B. cereus spores. Curve a is for untreated spores, and curve b is for spores that were incubated in 80°C water for 20 min. The baselines of the two curves were shifted for display. The inset shows the magnified spectra in the range of the protein amide I bands. Note that the positions of the bands assigned to the amide I bands of proteins in α-helical and irregular structures in the inset (and in Fig. 3 and 10) are designated 1,655 and 1,668 cm⁻¹, respectively, in agreement with nomenclature in previous work (20), even though the precise positions of these two bands varied somewhat between experiments. a.u., arbitrary units.

Wet-heat treatment of individual B. cereus spores.

Although the average spectral data discussed above provided some information on changes taking place during spore wet-heat treatment, these results were only an average of the behaviors of many different spores whose individual behaviors during heat treatment could differ significantly. Indeed, answers to a number of important questions about wet-heat inactivation of spores, including reasons for heterogeneity in the wet-heat resistance of individual spores in populations, and the relationships between CaDPA release, protein denaturation, and spore killing, will require information on these phenomena for individual spores. Figure 3 shows the Raman spectra of a typical B. cereus spore during incubation at 80°C. The curve (Fig. 3, curve b) indicates that after 7 min of incubation the spore contained the same amount of CaDPA as that at the beginning of measurement, and the intensities of Raman bands at 1,004 and 1,155 cm⁻¹ also remained unchanged. However, the spectral intensity at 1,655 cm⁻¹ had decreased, and the peak at 1,668 cm⁻¹ had become more prominent (Fig. 3, inset), suggesting that significant spore protein denaturation had taken place after 7 min at 80°C. However, by 8.5 min, all bands due to CaDPA, as well as the 1,155-cm⁻¹ band, had disappeared; the intensity of the band at 1,004 cm⁻¹ was reduced; and there was a further decrease in the intensity of the 1,655-cm⁻¹ band (Fig. 3, curve c), although there were no further changes over the next 6.5 min (Fig. 3, curve d). Similar dynamic phenomena (e.g., protein denaturation prior to rapid CaDPA release) were observed with 20 other individual B. cereus spores analyzed (data not shown).

FIG. 3.

Raman spectra of a single B. cereus spore heat treated in water at 80°C and the corresponding images taken under the illumination of trapping laser. Curves a to d are for heat treatment for 2, 7, 8.5, and 15 min, respectively. The inset shows the magnified spectra in the range of the protein amide I bands. The baselines were shifted for display. The vertical dotted lines show the positions of the amide I Raman bands from protein in an α-helical (1,655-cm⁻¹) or irregular (1,668-cm⁻¹) structure.

The images of this same spore taken under the illumination of a trapping laser are also shown in Fig. 3. Interestingly, the bright ring in the center of the spore seen at 2 min became dim just before the CaDPA was released and became bright again following CaDPA release, and then a new bright ring appeared around the central ring. The bright rings possibly arose from interference between the backward scattering lights from the spore's front and rear interfaces. Changes in the intensities of these rings may also reflect changes in the shape, size, or refractive index of the spore or the position of the spore in the trap or changes in some or all of these parameters.

To quantitatively describe features of CaDPA release and protein denaturation during spore wet-heat treatment, the intensities of Raman spectral bands at 1,017, 1,004 and 1,655 cm⁻¹ for the single B. cereus spore analyzed in Fig. 3 were plotted as a function of heating time (Fig. 4). The intensities of bands at 1,017 and 1,004 cm⁻¹ remained nearly constant for 6.5 to 7 min and then rapidly decreased in ∼1 min (Fig. 4a and b). During rapid CaDPA release, the intensity of the 1,017-cm⁻¹ band decreased to zero, suggesting that the spore released all of its CaDPA into the medium, while the intensities of the phenylalanine- and protein structure-specific bands fell only 30 to 50%, suggestive of denaturation of some spore proteins. However, the decrease in the intensity of the protein structure-specific band at 1,655 cm⁻¹ took place a bit earlier than did changes in the other two bands, suggesting again that protein denaturation slightly preceded rapid CaDPA release. As shown in Fig. 4, the Raman spectral intensities of the three bands were fit well by Boltzmann functions, and the times at which the spore's CaDPA level fell by 8% and 92% were defined as the lag time, T_lag, and release time, T_release, respectively. The great majority (84%) of the spore's CaDPA was released in ∼1 min, between T_lag and T_release.

FIG. 4.

Raman scattering intensities of a typical B. cereus spore during wet-heat treatment at 80°C. The Raman scattering and ELS intensities for the spore described in the legend for Fig. 3 were determined for various times and plotted as a function of incubation time at 80°C. (a to c) Plots for bands at 1,017, 1,004, and 1,655 cm⁻¹, respectively, are shown, and the experimental data are fit with Boltzmann functions (solid lines). (d) The ELS intensity of this same spore is shown. The two vertical dotted lines mark the positions of T_lag and T_release.

As described above, the brightness and pattern of the B. cereus spore's image observed under the illumination of the trapping laser also changed during heat treatment. To describe this change quantitatively, the counts of the image data were summed over all pixels to give the ELS intensity of single spores. When the ELS intensity of a single B. cereus spore was plotted as a function of heating time (Fig. 4d), the intensity decreased gradually ∼30%; reached a minimum at T₁ shortly after T_lag but earlier than T_release; then began to increase rapidly ∼1.5-fold until T₂, shortly after T_release; and finally remained nearly constant. The decrease and increase in ELS intensity before, during, and after CaDPA release during spore heat treatment suggest that significant changes in the structures and/or states of spore components occur during this process.

Kinetic parameters for changes in multiple individual B. cereus spores during wet-heat treatment.

With knowledge of the kinetics of changes in an individual B. cereus spore during wet-heat treatment in hand, we then turned to analysis of the wet-heat treatment of multiple individual B. cereus spores, since the wet-heat resistance of individual spores from the same spore preparation can vary significantly, as noted above. The Raman spectral intensities of the 1,017-cm⁻¹ CaDPA-specific band and the ELS intensities of six individual B. cereus spores as a function of incubation time at 80°C are shown in Fig. 5. The mean values and standard deviations for T_lag, T_release, ΔT_release, T₁, and T₂ and the mean CaDPA levels at T₁ for 20 individual B. cereus spores incubated at 80°C are also given in Table 1. The values of T_lag and T_release varied significantly from spore to spore, although the time required for release of the majority of spores' CaDPA (ΔT_release = T_release − T_lag) was nearly the same (∼1.5 min) for all individual spores. The changes in ELS intensity during incubation at 80°C were also always associated with CaDPA release, as the ELS intensities of each spore decreased gradually and reached their minima at T₁, when ∼80% of the spore's CaDPA was released, and increased rapidly to constant values at T₂, when spores had released all of their CaDPA. It should be noted that results from a few spores that did not show CaDPA release in ∼30 min were discarded in this experiment, as well as in experiments with B. megaterium and B. subtilis spores (data not shown).

FIG. 5.

CaDPA release and ELS intensities for six individual B. cereus spores during wet-heat treatment at 80°C. (a) Relative intensities of the CaDPA-specific Raman band at 1,017 cm⁻¹ for six individual spores (spores a to f) were determined as described in Materials and Methods. (b) Relative ELS intensities of the six spores described in the legend for panel a were determined as described in Materials and Methods. The baselines of ELS intensity were shifted for display, and the dashed line below each ELS signal is the corresponding zero level. The downward and upward arrows mark T_lag and T_release, respectively, for each spore.

TABLE 1.

Mean values and standard deviations of T_lag, T_release, ΔT_release, T₁, and T₂ and mean CaDPA levels at T₁ of 20 individual spores incubated at high temperatures in water^a

Spore and temperature	Tlag (min)	Trelease (min)	ΔTrelease (min)	T1 (min)	CaDPA level at T1 (%)	T2 (min)
B. megaterium at 80°C	6.1 ± 3.4	7.1 ± 3.3	1.0 ± 0.3	6.8 ± 3.4	24.0	7.5 ± 3.3
B. cereus at 80°C	7.7 ± 4.1	8.8 ± 4.1	1.1 ± 0.2	8.6 ± 4.1	20.0	9.7 ± 4.3
PS832 at 90°C	11.2 ± 5.6	12.6 ± 6.4	1.4 ± 1.0	11.6 ± 5.7	53.0	12.8 ± 6.2
PS533 at 90°C	14.4 ± 6.8	16.3 ± 7.9	2.0 ± 1.4	14.9 ± 7.0	41.0	16.5 ± 8.1

^aSpores of various Bacillus species or strains were incubated in water at various temperatures, and Raman spectra and ELS intensities for 20 individual spores in a dual trap as well as various parameters were determined as described in Materials and Methods and other portions of the text.

Kinetic parameters for changes in multiple individual B. megaterium and B. subtilis spores during wet-heat treatment.

To compare the kinetics of changes during wet-heat treatment in individual B. cereus spores with those of spores of other Bacillus species, we also examined 20 individual B. megaterium and B. subtilis spores incubated at 80°C and 90°C, respectively. The Raman spectral intensities of the main CaDPA band at 1,017 cm⁻¹ as well as the ELS intensities of six individual B. megaterium and B. subtilis PS533 spores during high-temperature incubation are shown in Fig. 6 and 7, and kinetic parameters for 20 individual spores of both species are summarized in Table 1. As was seen with B. cereus spores, the T_lag values prior to initiation of rapid CaDPA release varied significantly between individual spores, although the ΔT_release values were relatively constant. Also as seen with B. cereus spores, there was little if any decrease in CaDPA content during heat treatment prior to T_lag. This was seen most dramatically when the averaged intensities of CaDPA-specific Raman bands at 1,017, 1,395, and 1,570 cm⁻¹ for 10 B. subtilis PS533 spores were plotted as a function of incubation time in 90°C water and the relative times for individual spores were normalized to the time, T_c, at which 50% of the spore's CaDPA was retained (Fig. 8a). Strikingly, all CaDPA-specific bands exhibited nearly identical behaviors, and the spore's CaDPA level remained constant prior to rapid CaDPA release beginning at T_lag. In addition, when CaDPA release as measured from the intensities of the 1,017-cm⁻¹ bands in four individual spores was plotted in the same way (Fig. 8b), the rapid CaDPA release for individual spores took place almost in parallel, even though individual spores had values for T_lag that varied ≥2-fold (Fig. 7 and data not shown). Essentially identical results were obtained when similar analyses were carried out for multiple individual B. cereus or B. megaterium spores incubated at 80°C (data not shown). In contrast to the lack of change in the intensities of spores' CaDPA-specific Raman bands prior to T_lag noted above, there are significant changes in the ratios of intensities of CaDPA-specific Raman bands upon wet-heat activation of individual B. cereus and B. subtilis spores (47). The reason the latter differences were not seen in the current work is likely because wet-heat inactivation was carried out at temperatures 15 to 20°C higher than those for heat activation, and thus heat activation-related changes in the spores' Raman spectra would have taken place in ≤1.5 min, before acquisition of Raman spectra (4, 47). Indeed, by the time the B. cereus and B. subtilis spores were trapped, the ratios of intensities of CaDPA-specific Raman bands had changed to close to the corresponding values for heat-activated spores (Table 2). Perhaps not surprisingly, this suggests that during incubation at wet-heat killing temperatures, the individual spores had undergone heat activation changes in the state of CaDPA long before the spores were killed.

FIG. 6.

CaDPA release and ELS intensities for six individual B. megaterium spores during wet-heat treatment at 80°C. (a) Relative intensities of the Raman spectral band at 1,017 cm⁻¹ for six individual spores (spores a to f) were determined as described in Materials and Methods. (b) Relative ELS intensities of the six spores described in the legend for panel a were determined as described in Materials and Methods. The baselines of ELS intensity were shifted for display, and the dashed line below each ELS signal is the corresponding zero level. The downward and upward arrows mark T_lag and T_release, respectively, for each spore.

FIG. 7.

CaDPA release and ELS intensities for six individual B. subtilis PS533 spores during wet-heat treatment at 90°C. (a) Relative intensities of the Raman spectral band at 1,017 cm⁻¹ for six individual spores (spores a to f) were determined as described in Materials and Methods. (b) Relative ELS intensities of the six spores described in the legend for panel a were determined as described in Materials and Methods. The baselines of ELS intensity were shifted for display, and the dashed line below each ELS signal is the corresponding zero level. The solid downward, solid upward, and dotted downward arrows mark T_lag, T_release, and T₁, respectively, for each spore.

FIG. 8.

Intensities of CaDPA-specific bands from spores of B. subtilis PS533 as a function of time at 90°C. (a) Averaged intensities of CaDPA-specific Raman bands from 10 individual spores were normalized to the T_c at which the spore retains half of its initial CaDPA level after the spores were added to 90°C water. (b) Intensities of the 1,017-cm⁻¹ Raman band of four individual spores as a function of time at 90°C, with time values for individual spores normalized to T_c as described above.

TABLE 2.

Ratios of the intensities of characteristic Raman bands of B. cereus and B. subtilis spores^a

Spores and conditions used	I1,017/I824	I1,395/I824	I1,570/I824
B. cereus at 25°Cb	3.03 ± 0.09	1.79 ± 0.06	1.24 ± 0.05
B. cereus at 65°C for 30 minb	2.61 ± 0.08	1.33 ± 0.06	1.01 ± 0.04
B. cereus at 80°Cc	2.62 ± 0.09	1.25 ± 0.08	1.11 ± 0.04
B. subtilis at 25°Cb	2.19 ± 0.08	0.97 ± 0.05	0.87 ± 0.05
B. subtilis at 70°C for 30 minb	1.92 ± 0.08	0.85 ± 0.04	0.81 ± 0.03
B. subtilis at 90°Cc	1.86 ± 0.19	0.79 ± 0.04	0.76 ± 0.12

^aB. subtilis strain PS832 was used. I, intensity of Raman band at the indicated wavelength (cm⁻¹).

^bData for untreated spores at 25°C and heat-activated spores held at heat activation temperatures (65 or 70°C) in water are from reference 47.

^cB. cereus and B. subtilis (strain PS832) spores were injected into water at heat-killing temperatures, and intensities of CaDPA-specific Raman bands were determined as described in Materials and Methods and other portions of the text. The intensity ratios were determined just after the spores were trapped and averaged over 10 individual spores.

Correlations between changes in DPA and ELS intensity during wet-heat treatment of the B. megaterium and B. subtilis spores were also largely similar to those seen with B. cereus spores, including the decrease in ELS intensity prior to T_lag with B. megaterium spores (Fig. 6b). However, the ELS intensity from individual B. subtilis spores, either strain PS533 (Fig. 7b) or PS832 (data not shown), did not decrease during T_lag but remained nearly constant and then increased toward the end of rapid CaDPA release, the latter as was the case with B. cereus and B. megaterium spores. Consequently, for B. subtilis spores, T₁ was defined as the time at which the ELS intensities began to rise rapidly (Fig. 7b).

Comparison of kinetic parameters of the two spores in the dual trap during wet-heat treatment.

The results noted above indicated that individual spores from the same spore preparation have marked differences in their dynamic responses to incubation at high temperatures in water, and it will be important to identify the reasons for these differences. It is not clear if differences in the microscopic physical environment during wet-heat treatment itself, such as the temperature, pressure, and medium's thermal convection, might be responsible for heterogeneity in the wet-heat resistance of individual spores in populations. This concern could, however, be addressed through the use of the dual-trap optical tweezers, since the minimal separation (∼3 μm) of the spores in the two traps makes it highly likely that the two trapped spores are exposed to essentially the same microenvironment. Figure 9 shows the relative Raman spectral intensities of the CaDPA-specific 1,017-cm⁻¹ band for a typical pair of B. subtilis PS533 spores confined in the dual trap. It is obvious that this pair of spores exhibited totally different wet-heat resistances, although their CaDPA contents as determined by the absolute Raman spectral intensities of the band at 1,017 cm⁻¹ were almost the same (data not shown). Twenty pairs of PS533 spores were compared in this fashion, and the correlation coefficient between T_lag1 and T_lag2 was calculated to be a ρ₁₂ (Pearson's correlation coefficient) of 2.4%, where subscripts 1 and 2 represent the spores in the two traps. This value for ρ₁₂ indicates that the difference in wet-heat resistances between the two spores in the dual trap is not due to different microenvironments during wet-heat treatment but rather stems from heterogeneity between different individual spores.

FIG. 9.

CaDPA release from a typical pair of B. subtilis PS533 spores during wet-heat treatment at 90°C. This pair of spores was confined in dual traps separated by ∼3 μm.

Status of carotenoids during wet-heat treatment of B. megaterium spores.

In addition to the prominent Raman bands seen with B. cereus and B. subtilis spores, B. megaterium spores also exhibit two additional major Raman bands at 1,155 and 1,516 cm⁻¹ due to the C-C and C=C stretch vibrations of carotenoids (3), as shown in Fig. 10a. This B. megaterium carotenoid is thought to be located in the spore's inner membrane (26, 32) but is largely if not completely absent from B. cereus and B. subtilis spores (Fig. 2; data not shown), although B. cereus and B. subtilis spores do exhibit a small Raman band at 1,155 cm⁻¹ (Fig. 2 and 3; data not shown). Figure 10a and b show the Raman spectra and the calculated intensities of bands at 1,017, 1,155, and 1,516 cm⁻¹ of a typical B. megaterium spore during incubation at 80°C. The intensities of the carotenoid-specific bands dropped rapidly during incubation at 80°C, and the band at 1,516 cm⁻¹ had disappeared by 6 min. The intensity of the 1,155-cm⁻¹ band exhibited two periods of change. Initially, a fast change paralleled the change in the 1,516-cm⁻¹ band and decreased the band's intensity to ≤10% of its initial value, although this decreased amplitude may be underestimated due to the delay in spectrum acquisition because of the time needed to fix two spores in the dual trap; indeed, some B. megaterium spores had largely lost these two bands by the time they were trapped (data not shown). Interestingly, the reduced height of the 1,155-cm⁻¹ band in B. megaterium spores prior to T_release was comparable to the height of this band in B. cereus and B. subtilis spores at either T₀ or T_lag (Fig. 2, 3, and 10; data not shown). However, after T_lag, the residual intensity of the 1,155-cm⁻¹ band in B. megaterium spores decreased rapidly to zero (Fig. 10), as was also the case during wet-heat treatment of B. cereus and B. subtilis spores (Fig. 3; data not shown).

FIG. 10.

Effect of wet-heat treatment at 80°C on carotenoids in B. megaterium spores. Raman spectra of a typical B. megaterium spore incubated in water at 80°C for different times at the 1st, 6th, 10th, and 13th min after the addition to heated water (a) and the spectral intensities of bands at 1,017, 1,155, and 1,516 cm⁻¹ as a function of time at 80°C (b) were determined as described in Materials and Methods. The zero levels of bands at 1,017 and 1,155 cm⁻¹ were shifted upward for display and are marked by the dashed lines that cross throughout the relevant spectra.

DISCUSSION

The results reported in this communication allow a number of new, significant conclusions about wet-heat inactivation of individual spores of Bacillus species, conclusions that could not be drawn from studies of the wet-heat treatment of spore populations. Most of these conclusions are the same for spores of three different Bacillus species, suggesting that these conclusions are general ones for spores of all Bacillus species.

The first conclusion is that during incubation in water at heat-killing temperatures, Bacillus spores exhibit no significant change in their CaDPA content until T_lag and then release the majority of their CaDPA in 1 to 2 min in ΔT_release. The mechanism that determines CaDPA release during ΔT_release is not clear. However, it seems likely that this rapid CaDPA release is due to a breakdown of the spore's major permeability barrier to small molecules, and this permeability barrier is most likely the spore's inner membrane (37, 38). In addition, it is not exactly clear how CaDPA leaves the spore core during heat treatment. During spore germination, CaDPA is also almost completely released in 1 to 2 min (5, 30), and while it is not known how this release is triggered, CaDPA is suggested to be released via inner membrane channels composed at least in part of SpoVA proteins (41, 42). Perhaps high-temperature treatment somehow triggers the rapid synchronous opening of these normal CaDPA channels, either by action on the channels themselves or indirectly by effects on the spore's inner membrane. Alternatively, there may be such a drastic change in inner membrane structure that all core small molecules, not just CaDPA, are rapidly released in parallel during wet-heat treatment. Indeed, all core small molecules are released during spore wet-heat treatment (21, 37, 38), although the rates of rapid release of different core small molecules are not known for individual spores. In addition, not all core small molecules are released during spore germination (36), so it may be possible to determine if small-molecule release during wet-heat treatment is similar in its selectivity to that seen with spore germination.

The second conclusion is that while values of ΔT_release during spore heat inactivation are relatively constant for different individual spores, either in one species or across species, the variation in T_lag is extremely large, most notably between individual spores in the same spore preparation. The reason for these differences between individual spores in populations is not some heterogeneity in the physical microenvironment during wet-heat treatment, but a likely cause is spore heterogeneity due to differences in the microenvironments of different cells in the same sporulating culture, perhaps because different cells in a culture sporulate at different times. While the overall physical and chemical environments during sporulation are the same for all sporulating cells in a population, the microenvironments may vary significantly, in particular because of differences in the kinetics of sporulation of different cells in the population (18, 39, 40). Such differences in spore formation may in turn result in slight differences in the levels of various components important in spore moist-heat resistance such as core water, divalent cations, and CaDPA, as well as slight differences in structures of important spore layers, such as the peptidoglycan cortex. A number of these slight differences might act synergistically to cause significant differences in the heat resistance of individual spores in populations, and a challenge for the future will be to identify those parameters responsible for the heterogeneity in spore wet-heat resistance and how this heterogeneity arises during sporulation.

The third conclusion is that during wet-heat treatment, some spore proteins are denatured or damaged in parallel with and even slightly prior to rapid CaDPA release during wet heat treatment, as some spore protein changes from an α-helical structure to one that is irregular and likely denatured. The state of phenylalanine in spore proteins also changes in parallel with and to some degree even before rapid CaDPA release, and this change may also be related to protein damage or denaturation. Given that protein damage can inactivate protein function, this suggests that protein damage is a major reason that wet heat kills spores of Bacillus species, as was suggested recently (7), although specific proteins to which damage causes spore death during wet-heat treatment have not been identified.

It was also notable with individual spores exposed to wet heat that there was significant change in overall spore protein structure prior to rapid CaDPA release. This suggests that during wet-heat treatment some spores should retain CaDPA for a short period, even if they are already dead from protein damage, and some B. subtilis spores in populations treated with wet heat have been shown to do just that (7). Presumably wet heat causes damage to some crucial protein or proteins such that while CaDPA is retained, the spores are not viable, and even if they can germinate, they cannot progress into outgrowth (7, 8). However, with further wet-heat treatment, more protein damage accumulates, including damage to proteins in the spore's inner membrane, and ultimately, this damaged membrane ruptures, leading to rapid CaDPA release. Unfortunately, we have no way at present to know exactly when individual spores being observed during wet-heat treatment actually die, although available evidence indicates that B. subtilis spores that have lost all CaDPA during wet-heat treatment are definitely dead (7). However, spores that retain CaDPA during wet-heat treatment may be dead, still alive, or only conditionally alive, depending on the conditions used for recovering viable spores (7, 8).

The fourth conclusion concerns correlations between the times for changes in spores' ELS intensities and CaDPA release during wet-heat treatment. For B. cereus and B. megaterium spores, ELS intensities decreased gradually during wet-heat treatment, reaching minimum values at T₁, when about 80% of the spores' CaDPA had been released, and then rose rapidly until T₂, when CaDPA release was complete. The ELS intensities of B. subtilis spores during wet-heat treatment exhibited similar features, except that ELS intensities remained constant prior to T₁. Increases in ELS intensity are possibly related to abrupt and/or significant changes in the spore core refractive index due to CaDPA release and its replacement by water. Inhomogeneity of the refractive index inside the spore may also generate more scattering centers that scatter more incident photons (30). Changes in the refractive index of spore core components also may result in minute transitions in spores' axial position in the optical trap, since changes in the ratio of the particle's refractive index to the refractive index of the medium leads to a reequilibration of scattering force and gradient force in the axial direction (28). When the scattering force decreases or the gradient force increases, the spore will be trapped more closely to the focus of the laser. However, the reasons that the ELS intensities of B. cereus and B. megaterium spores decrease gradually prior to T₁ and why this is not seen with B. subtilis spores are not clear.

The last novel conclusion concerns the carotenoid or carotenoids present in high levels in B. megaterium spores, supposedly in the spore inner membrane (26, 32). Carotenoid-specific Raman bands at 1,155 and 1,516 cm⁻¹ decreased rapidly beginning as soon as these spores were exposed to elevated temperatures, with the 1,516-cm⁻¹ peak disappearing completely and well before rapid CaDPA release. However, the intensity of the band at 1,155 cm⁻¹ changed in two stages, an initially large decrease well before T_lag and then a rapid decrease to zero accompanying rapid CaDPA release. The mechanism for the changes in the intensity of carotenoid-specific Raman bands is not clear but is possibly related to photo-oxidation of carotenoids induced by the trapping laser at elevated temperatures, although experiments carried out at room temperature indicated that this effect was not pronounced (data not shown). In addition, the residual intensity of the 1,155-cm⁻¹ band prior to T_lag may not be due to a carotenoid but to CaDPA (19). However, if the carotenoid is indeed in the spore's inner membrane (26, 32), then changes in spores' carotenoid Raman spectra during wet-heat treatment may be a reflection of striking changes in the environment and structure of the spore inner membrane during this treatment.