INTRODUCTION

Top Abstract Introduction Materials and Methods Results and Discussion References

Dormant bacterial endospores are highly resistant to a number of physical and chemical treatments which are normally considered germicidal (reviewed in reference 37). Spores owe some of their resistance to the presence of an outer proteinaceous layer termed the spore coat (reviewed in references 1, 6, 9, and 28). The question of how the spore coat contributes to spore resistance is currently a subject of considerable investigation (reviewed in reference 6). Experimental evidence indicates that the coat protects a dormant spore from enzymes, such as lysozyme (11), and from mechanical disruption (1, 9). The spore coat also protects the spore from some chemicals, such as hydrogen peroxide (H₂O₂) (11), but not others, such as organic solvents (19). Spore resistance to organic solvents and heat seems to be a function of the peptidoglycan cortex which underlies the coat (19, 29, 30), and protection of spore DNA from 254-nm UV radiation and from free radical damage is associated with binding of spore DNA by small, acid-soluble spore proteins in the spore core (reviewed in reference 37). Much of the information regarding the role of the spore coat in spore resistance is derived from studies in which workers examine spore resistance after the coat protein layers are chemically removed by treatment with reducing and protein-denaturing agents (24, 36, 41). Although it is not known how such harsh chemical treatments alter other spore components, such as the cortex, membranes, or core, there is some evidence which suggests that chemical decoating may also affect these protective structures (3, 46).

The synthesis and substructure of the spore coat were examined initially by electron microscopy and biochemical characterization of spore coat proteins (1) and more recently by molecular biological techniques (6, 49). Electron microscope studies have revealed that the spore coat of Bacillus subtilis is actually an ordered structure consisting of the following three morphologically distinct layers: an electron-dense outer coat, a thinner lamellalike inner coat, and an electron-diffuse undercoat (1, 6, 33a). The molecular events that underlie the control of coat protein synthesis and morphogenesis of the coat layers during B. subtilis sporulation have been the subject of recent intensive studies. All coat structural and morphogenetic proteins are synthesized in the mother cell compartment in a defined temporal sequence and are assembled in an ordered fashion on the surface of the developing spore (6). Two proteins produced in the mother cell, GerE and CotE, play major roles in the synthesis of spore coat proteins and in spore coat morphogenesis, respectively. GerE is a small DNA-binding protein (15) which appears to either positively or negatively regulate expression of several of the cot genes encoding spore coat structural proteins (55, 56). The gerE36 mutation, which was originally isolated on the basis of an impaired germination phenotype (22), was later shown to be a nonsense mutation (4a) which behaves like a null allele (57). Spores of gerE36 strains appear to be completely devoid of the lamellalike inner coat layers, but there is a somewhat misassembled outer coat that is loosely associated with each spore (6, 21). It has been shown that the CotE protein forms a shell around the developing forespore and that internal and external to this shell the inner and outer coat layers, respectively, assemble (7). cotE::cat insertion mutants produce mature spores which lack an outer coat, and the inner coat of each mature spore appears to be loosely associated with the cortex (6). A mutant strain carrying both gerE36 and cotE::cat produces mature spores that lack both the outer and inner coat layers (7).

Of recent interest in our laboratory has been the exploration of the factors which confer B. subtilis spore resistance to environmentally relevant extreme conditions, particularly solar UV radiation (reviewed in reference 23). The current molecular models of spore UV resistance mechanisms have been developed mainly from laboratory experiments performed with monochromatic 254-nm (UV-C) radiation, which is absent from the solar UV spectrum on the earth's surface (45a). We have found that B. subtilis spores exposed to solar UV radiation, which consists of UV-B and UV-A wavelengths ranging from 290 to 400 nm, exhibit quite different DNA photochemistry (40) and DNA repair responses (52) than do spores exposed to monochromatic 254-nm UV-C radiation in the laboratory. It has been reported that the spore coat layers do not contribute to spore resistance to 254-nm UV-C radiation (41). However, do the spore coats contribute to spore resistance to environmentally relevant solar UV wavelengths? Using strains carrying the gerE36 and/or cotE::cat mutation that affects spore coat synthesis and/or assembly, in this study we investigated the contributions of the inner and outer spore coat layers to the resistance of B. subtilis spores to sporicidal conditions both in the laboratory and in the environment.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results and Discussion References

Bacterial strains, plasmids, and growth conditions. All of the B. subtilis and Escherichia coli strains and plasmids used in this study are listed in Table 1. All of the B. subtilis strains used are isogenic derivatives of wild-type strain PY79 (54). Luria-Bertani medium (20) and Spizizen's minimal medium (42) containing the appropriate antibiotic(s) and/or growth requirements were used for routine maintenance and cultivation of B. subtilis and E. coli strains. When appropriate, selective antibiotics were added to media at the following final concentrations: chloramphenicol, 3 µg/ml; ampicillin, 50 µg/ml; and a combination of erythromycin (1 µg/ml) and lincomycin (25 µg/ml) (MLS). Liquid cultures were incubated with vigorous aeration, and all cultivations were performed at 37°C. Spores of B. subtilis strains were produced by cultivating the strains for 3 days at 37°C on solid agar plates containing nutrient broth sporulation medium (NSM) (32); this procedure resulted in more free spores available for purification than did cultivation in liquid media. Due to the lysozyme-sensitive nature of mutant spores lacking coats, all spore preparations were purified by repeated centrifugation and water washing as previously described (24) until more than 99% free spores were obtained, as assessed by phase-contrast microscopy. Purified spores were suspended in phosphate-buffered saline (PBS) (pH 7.4) (24), heat shocked (80°C, 10 min), and stored in the dark at 4°C until they were used. In all experiments in which spore resistance was measured, the plates were incubated at 37°C for at least 48 h prior to scoring to ensure that all survivors had formed visible colonies.

Chemical decoating of spores. Spores were suspended in decoating solution (50 mM Tris base [pH 10], 8 M urea, 50 mM dithiothreitol, 1% [wt/vol] sodium dodecyl sulfate) and incubated at 60°C for 90 min with vigorous vortexing at 10-min intervals. The treated spores were washed three times by resuspending them in STE buffer (150 mM NaCl, 10 mM Tris-HCl [pH 8], 1 mM EDTA) and centrifugating them (10,000 × g, 10 min), and then they were resuspended in PBS.

Assay for spore lysozyme resistance. Spores were diluted to a concentration of ca. 10⁸ CFU/ml in PBS and were titrated on NSM before and after treatment with lysozyme (final concentration, 0.5 mg/ml; Sigma) at 37°C for 10 min.

Assay for spore H₂O₂ resistance. Spores were diluted to a concentration of approximately 10⁸ CFU/ml in PBS, and 933 µl of the spore suspension was placed in a 1.7-ml microcentrifuge tube. After 100 µl was removed in order to determine the titer of viable spores in the remaining preparation (833 µl), 167 µl of 30% H₂O₂ (Mallinkrodt) was added to the spore suspension. The final H₂O₂ concentration was 5%. The suspension was incubated at room temperature (~25°C) with continuous gentle mixing, and 100-µl samples were removed at various times and immediately diluted 1:10 with a solution of bovine catalase (100 µg/ml in PBS; Sigma) that previously had been filter sterilized with a 0.45-µm-pore-size filter. Serial 1:10 dilutions of the catalase-treated spore suspension were then plated onto NSM and incubated in order to determine the number of viable colonies.

Assays for spore UV resistance. Spores were exposed to artificial UV-C, UV-B, and solar radiation and levels of spore survival were determined as described in detail previously (52). Artificial UV-C radiation was provided by a commercial low-pressure mercury arc lamp (model UVGL-25; UV Products, San Gabriel, Calif.) which emitted essentially monochromatic 254-nm UV radiation. Artificial UV-B radiation was provided by a commercial medium-pressure mercury arc lamp (model UVM-57; UV Products) which emitted a spectrum of UV wavelengths from 280 to 320 nm, with peak emission at 302 nm. The UV-B lamp was further modified to block UV wavelengths of less than 290 nm; this was done by using a polystyrene filter fashioned from a petri dish lid (LifeLINE Dishes; product no. LS-6601; Life Science Products, Denver, Colo.). For each artificial UV-B trial, the UV-B lamp was fitted with a fresh polystyrene filter. For solar UV radiation experiments, spore samples were exposed to sunlight on the roof of Building 90 at the University of Arizona during the daily period of maximal solar intensity, from 2 h before local noon to 2 h after local noon; local noon was calculated for the longitude of Tucson, Ariz. (111° 02' W), by using the Voyager II computer program (Carina Software, San Leandro, Calif.). Spore samples were exposed to full-spectrum sunlight by using a single layer of Saran Wrap (Dow Brands, Indianapolis, Ind.) as a UV-transparent covering (52). Spore samples were exposed to sunlight in which the UV-B portion had been filtered out (referred to as solar UV-A radiation below) by utilizing a 1.25-cm (0.5-in.)-thick glass plate as a solar UV-B filter which blocked UV wavelengths shorter than 325 nm (52). The UV doses produced by the artificial UV and solar UV radiation sources were measured by using a model UVX radiometer (UV Products) and the appropriate calibrated probes for UV-C radiation (model UVX-25), UV-B radiation (model UVX-31), and UV-A radiation (model UVX-36). UV doses are reported below in Joules per square meter.

Statistical analysis. Below, 90% lethal doses (LD₉₀) determined in experiments in which we examined spore resistance to hydrogen peroxide and to artificial UV treatments are expressed as averages ± standard deviations. LD₉₀ determined in experiments in which we examined spore resistance to solar UV radiation are expressed as averages ± standard errors because triplicate samples were studied at each time point in each independent trial. The significance of differences in LD₉₀ was determined by analysis of variance (ANOVA) by using Minitab software, version 10.5. Values were analyzed in multigroup pairwise combinations, and differences with P values of 0.05 were considered statistically significant.

	RESULTS AND DISCUSSION

Top Abstract Introduction Materials and Methods Results and Discussion References

It has been suggested that the spore coat plays a major role as a barrier to harmful sporicidal conditions that a spore might encounter during dormancy (reviewed in reference 6). In previous studies in which this issue was examined, researchers used a combination of chemical decoating procedures and spore coat-defective mutants to elucidate how the spore coat contributes to a spore's resistance to various sporicidal agents (1, 3, 10-12, 17, 21, 25, 41, 47, 50, 51). In this study, we used both chemical removal of the spore coat and mutants which produce incomplete and aberrant coats to examine the contributions of the inner and outer spore coat layers to resistance of spores to the common oxidant H₂O₂, to laboratory-generated UV-C and UV-B radiation, and to solar UV-B and UV-A radiation.

Spore resistance to lysozyme. Resistance to lysozyme is the traditional method used to assess the integrity of the spore coat layers (6, 21, 56); therefore, we tested the lysozyme resistance of spores of the strains used in this study. Under our lysozyme treatment conditions (0.5 mg of lysozyme per ml, 37°C, 10 min), 95% of wild-type PY79 spores survived; in contrast, chemically decoated PY79 spores became sensitized to lysozyme, and 4% of these spores survived. Spores of strains AD17, AD28, and AD142, which lacked the inner coat, the outer coat, and both coat layers, respectively, all were extremely sensitive to lysozyme (levels of survival, <0.006%), indicating that these strains contained severe defects in spore coat integrity.

Spore resistance to hydrogen peroxide. The levels of resistance of spores of wild-type strain PY79 and coat-defective derivatives to 5% H₂O₂ were determined. Semilogarithmic plots of the percentage of spores that survived versus time yielded the dose (in minutes) required to kill 90% of the spore population (LD₉₀), and the average LD₉₀ after at least three separate treatments were compared (Fig. 1). The LD₉₀ for PY79 (wild-type) spores was 49 ± 3 min; spores of PY79 whose spore coats had been chemically removed became sensitized to 5% H₂O₂, and the LD₉₀ was 12 ± 2 min (Fig. 1). Mutant derivatives of PY79 which made defective spore coats also were very sensitive to 5% H₂O₂. The LD₉₀ for strains AD28 (cotE::cat), AD17 (gerE36), and AD142 (cotE::cat gerE36) were 12 ± 4, 10 ± 2, and 7 ± 2 min, respectively (Fig. 1). The differences between the LD₉₀ for PY79 and the LD₉₀ for all of the other strains were highly significant, as determined by ANOVA (P < 0.001), while the LD₉₀ for strain AD28, AD17, and AD142 spores and chemically decoated PY79 spores were not significantly different, as determined by ANOVA. The results indicated that alterations in the spore coat resulting from either chemical removal or mutation increased the sensitivity of the endospores to 5% H₂O₂ approximately four- to sevenfold.

View larger version (41K): [in this window] [in a new window]   FIG. 1.   Spore resistance to 5% H2O2. The strains were assayed for H2O2 resistance as described in the text. LD90 are expressed as averages ± standard deviations (n  3). The asterisks indicate LD90 that were significantly different than the LD90 for wild-type PY79 spores, as determined by ANOVA (P  0.05). w.t., wild type.

Spore resistance to artificial UV-C radiation. Early experiments which indicated that the spore coat layers are not important determinants of spore UV resistance (41, 44, 45) were based on studies in which monochromatic 254-nm UV-C radiation was used. Resistance of spores to 254-nm UV-C radiation was assayed by determining the average LD₉₀ for each strain in three (strains AD17 and AD28 and chemically decoated strain PY79) or seven (strains PY79 and AD142) independent trials (Fig. 2). The LD₉₀ for spores of PY79, chemically decoated PY79, AD17 (gerE36), and coatless mutant AD142 (cotE::cat gerE36) were 102 ± 14, 118 ± 6, 101 ± 8, and 100 ± 18 J/m², respectively (Fig. 2). The LD₉₀ for these four strains were not significantly different, as determined by ANOVA, which strongly implied that removal of both spore coat layers by chemical or mutational methods did not affect spore resistance to 254-nm UV-C radiation; this conclusion is consistent with historical observations that chemical coat removal did not affect spore UV-C resistance (41, 44, 45). Interestingly, spores of mutant strain AD28 (cotE::cat), which lacks the outer spore coat, were more resistant to UV-C than were spores of all of the other strains; the LD₉₀ for AD28 spores was 140 ± 5 J/m² (Fig. 2). The difference, while not dramatic, was nonetheless statistically significant, as determined by ANOVA (P = 0.004).

View larger version (56K): [in this window] [in a new window]   FIG. 2.   Spore resistance to 254-nm UV-C radiation. The strains were irradiated and levels of survival were determined as described in the text. LD90 are expressed as averages ± standard deviations (n = 3 for AD17, AD28, and chemically decoated PY79; n = 7 for PY79 and AD142). The asterisk indicates an LD90 that was significantly different than the LD90 for wild-type spores, as determined by ANOVA (P  0.05). w.t., wild type.

Spore resistance to artificial UV-B radiation. While UV-C radiation is a convenient germicidal treatment and relevant to sterilization procedures, results obtained by using 254-nm UV-C radiation are not truly representative of the results obtained by using the UV wavelengths that endospores encounter in their natural environments (45a). Therefore, we examined the contribution which the spore coats made to spore resistance to artificial UV-B radiation (wavelengths, 290 to 320 nm), which includes wavelengths found in the UV-B portion of sunlight. Surprisingly, we found that although chemical decoating of PY79 spores did not affect the UV-B resistance of the spores, spores of coatless mutant strain AD142 (gerE36 cotE::cat) and gerE36 strain AD17 were significantly more UV-B sensitive than wild-type spores. Spore resistance to artificial UV-B radiation (290 to 320 nm) was assayed by determining the average LD₉₀ for each strain based on a minimum of three trials (Fig. 3). The LD₉₀ obtained for spores of wild-type strain PY79, chemically decoated strain PY79, AD28 (cotE::cat), AD17 (gerE36), and AD142 (gerE36 cotE::cat) were 39.8 ± 0.8, 41.9 ± 0.3, 52.1 ± 1.3, 31.3 ± 1.3, and 28.0 ± 2.3 kJ/m², respectively. The LD₉₀ for decoated PY79 spores was not significantly different than the LD₉₀ for PY79 spores (Fig. 3), indicating that chemical spore coat removal did not affect spore resistance to UV-B radiation. However, the LD₉₀ for spores of coat mutant strains AD17 (gerE36) and AD142 (gerE36 cotE::cat) were significantly lower than the LD₉₀ for wild-type PY79 spores (Fig. 3), indicating that, in contrast to the results obtained with artificial UV-C radiation, spores that had defects in their coats resulting from the gerE36 mutation were more sensitive to UV-B wavelengths. Again, as observed with UV-C radiation (Fig. 2), spores of strain AD28 (cotE::cat) were significantly more resistant to artificial UV-B radiation than spores of wild-type strain PY79 (Fig. 3).

View larger version (53K): [in this window] [in a new window]   FIG. 3.   Spore resistance to UV-B radiation. The strains were irradiated and levels of survival were determined as described in the text. LD90 are expressed as averages ± standard deviations (n  3). The asterisk indicate LD90 that were significantly different than the LD90 for wild-type PY79, as determined by ANOVA (P  0.05). w.t., wild type.

Spore resistance to full-spectrum solar radiation. In order to assess the contribution of the spore coat to solar radiation resistance in the field, we utilized an exposure system described in detail previously (52). In field trials we noted that exposure to solar radiation resulted in considerable heating of spore samples (temperatures greater than 70°C were not uncommon) (40) and that chemically decoated spores of PY79 survived field exposure poorly. As a result of further investigation of this phenomenon, we observed that our chemical decoating procedure did not affect PY79 spore viability but resulted in sensitization of PY79 spores to 90°C wet heat; the LD₉₀ for these spores decreased from 11 ± 3 to 2 ± 0.3 min. While sensitization of chemically decoated PY79 spores to lysozyme and H₂O₂ (Fig. 1) is consistent with results obtained previously with spores of several different Bacillus and Clostridium species (2, 11, 12, 16), the reduction in the heat resistance of PY79 spores exposed to our decoating procedure may have been due to additional weakening of the spore integument. In support of this hypothesis, it has been observed that chemical decoating causes the release of dipicolinic acid (DPA) and hexosamine from spores, which indicates that the cortex integrity may also be compromised (3, 46). Therefore, we were not able to assess the effect of solar UV radiation on chemically decoated spores. In contrast to chemically decoated spores, spores of mutant strains carrying the gerE36 and/or cotE::cat mutation were very sensitive to 5% H₂O₂ (Fig. 1) but were as heat resistant as wild-type spores (data not shown), indicating that the absence of coat layers due to the gerE36 and cotE::cat mutations probably does not affect cortex integrity and/or core dehydration.

View larger version (44K): [in this window] [in a new window]   FIG. 4.   Spore resistance to full-spectrum solar radiation. The strains were irradiated and levels of survival were determined as described in the text. LD90 are expressed as averages ± standard errors (n  3). The asterisks indicate LD90 that were significantly different than the LD90 for wild-type WN515 spores, as determined by ANOVA (P  0.05). w.t., wild type.

Spore resistance to solar UV-A radiation. Spores were exposed to solar radiation from which the UV-B portion was filtered by plate glass, and we constructed semilogarithmic plots of the percentage of survival versus dose, as measured with the UV-A radiation probe. The LD₉₀ was calculated for each strain in at least three independent trials. Killing of spores by UV-A sunlight was much less efficient and required longer exposure times. The LD₉₀ of UV-A sunlight for wild-type strain WN515 spores was 164.9 ± 3.2 kJ/m², whereas spores of coat mutant strains WN512 (gerE36) and AD142 (gerE36 cotE::cat) were significantly more sensitive to UV-A solar radiation; the LD₉₀ for the latter two strains were 101.0 ± 5.0 and 100.0 ± 7.5 kJ/m², respectively (Fig. 5). Again, as observed with artificial UV-C radiation (Fig. 2), UV-B radiation (Fig. 3), and full-spectrum solar radiation (Fig. 4), spores of strain AD28 (cotE::cat) were significantly more resistant to full-spectrum solar radiation than wild-type spores, and the LD₉₀ for AD28 spores was 240.0 ± 20.2 kJ/m² (Fig. 5).

View larger version (47K): [in this window] [in a new window]   FIG. 5.   Spore resistance to solar UV-A radiation. The strains were exposed and levels of survival were determined as described in the text. LD90 are expressed as averages ± standard errors (n  3). The asterisks indicate LD90 that were significantly different than the LD90 for wild-type WN515 spores, as determined by ANOVA (P  0.05). w.t., wild type.

View larger version (62K): [in this window] [in a new window]   FIG. 6.   Summary of the UV resistance of spores of coat mutant strains normalized to the resistance of the wild-type strain. The asterisks indicate LD90 that were significantly different than the LD90 for wild-type spores within a treatment group, as determined by ANOVA (P  0.05). w.t., wild type.