Role of DNA Repair by Nonhomologous-End Joining in Bacillus subtilis Spore Resistance to Extreme Dryness, Mono- and Polychromatic UV, and Ionizing Radiation

Abstract

The role of DNA repair by nonhomologous-end joining (NHEJ) in spore resistance to UV, ionizing radiation, and ultrahigh vacuum was studied in wild-type and DNA repair mutants (recA, splB, ykoU, ykoV, and ykoU ykoV mutants) of Bacillus subtilis. NHEJ-defective spores with mutations in ykoU, ykoV, and ykoU ykoV were significantly more sensitive to UV, ionizing radiation, and ultrahigh vacuum than wild-type spores, indicating that NHEJ provides an important pathway during spore germination for repair of DNA double-strand breaks.

It has been shown that endospores of gram-positive bacteria can remain viable for at least thousands of years (5, 44, 54; reviewed in reference 31). Bacterial spores persist in a metabolically inactive state, and environmental damage to spore cellular components accumulates unrepaired until germination and outgrowth (32). However, Bacillus subtilis spores are highly resistant to different environmental stresses, such as toxic chemicals and biocidal agents, desiccation, pressure and temperature extremes, and ionizing and UV radiation. The reason for this high resistance to environmental extremes lies partly in the spore structure itself: spores possess thick layers of highly cross-linked coat proteins (13), a modified peptidoglycan spore cortex, and abundant intracellular constituents such as the calcium chelate of dipicolinic acid and small, acid-soluble spore proteins (α/β-type SASP), as protectants of spore DNA (46, 50). Binding of α/β-type SASP to spore DNA, coupled with spore core dehydration, appears to change the helical conformation of spore DNA from the B form to an A-like form (34, 48), which in turn alters its UV photochemistry to favor the production of 5-thyminyl-5,6-dihydrothymine, the unique spore-specific spore photoproduct (SP) (8, 32, 35, 50). For the removal of the SP, spores possess an SP-specific repair enzyme called SP lyase, encoded by the splB gene, that monomerizes the SP dimer back to two thymine residues in an adenosyl-radical-dependent reaction (4, 28, 42).

While the UV photochemistry of spore DNA and the repair of UV damage to DNA during germination are well described (12, 32, 33, 47, 50), there has been relatively little work on the nature of DNA damage in spores caused by ionizing radiation or extreme dryness and on the occurrence of a specific DNA repair system(s) for repair of this damage. It is assumed that DNA double-strand breaks (DSB), which are the most critical damage caused by ionizing radiation (57) and desiccation (9, 10, 39) in vegetative cells, are also induced in bacterial spores. Spores of B. subtilis contain a single chromosome arranged in a toroidal shape (16, 41); therefore, the homologous recombination pathway, which requires at least two homologous chromosomes, cannot operate on DSB during spore germination (55). An alternative repair pathway for DSB induced in spore DNA, nonhomologous-end joining (NHEJ) (3, 56), is considered here. This pathway as it occurs in eukaryotic cells requires a DNA end-binding component called Ku (Ku70 and Ku80) (58). The first step in the NHEJ DNA repair mechanism involves the binding of the Ku complex to the two DSB ends. The next proteins to be recruited to the complex are those that lead to resection of the ends of a DSB. The final set of steps in NHEJ leads to direct joining of the two ends by a specific DNA ligase that restores the integrity of the DNA. The ligation of DNA strands is an energy-dependent process, and ATP is required for the catalysis of the formation of a phosphodiester at the site of a single-strand break (SSB) in a duplex DNA (58). Recently, Weller et al. (57) identified in B. subtilis a Ku homolog (encoded by the ykoV gene), which retains the biochemical characteristics of the eukaryotic Ku heterodimer. The bacterial Ku specifically recruits a DNA ligase (encoded by ykoU) to DNA ends and thereby stimulates DNA ligation. Loss of these proteins leads to hypersensitivity of stationary-phase B. subtilis cells to ionizing radiation (57). From the observation that the Ku system is conserved in spore-forming bacterial species (e.g., Bacillus and Streptomyces spp.), Weller et al. (57) speculated that NHEJ via the prokaryotic Ku system might function during subsequent spore germination to repair DSB induced in dormant bacterial spores. Recently, Wang et al. (55) showed by microarray analyses that the ykoU and ykoV genes of the ykoWVU operon were under the control of sigma-G RNA polymerase during forespore development. They showed that ykoV and ykoU ykoV mutant spores were significantly more sensitive than wild-type spores to dry heat, a treatment known to cause DNA strand breaks, but no difference in hydrogen peroxide resistance between mutant and wild-type spores was observed (55). These observations led to the assumption that SP lyase, also expressed by sigma-G RNA polymerase during sporulation (32, 37), and NHEJ are both spore-specific DNA repair pathways, components of which are synthesized in the forespore during sporulation, packed in the dormant spore, and activated during germination.

We have been conducting ongoing studies on the role of DNA repair in the resistance of bacterial spores to the space environment, which is characterized by extreme vacuum, high solar UV flux, and ionizing radiation (reviewed in references 15, 18, 20, 32, and 33), all of which are implicated in the formation of DSB and SSB in DNA. Therefore, in this communication we report results from experiments examining the role of NHEJ in spore resistance to UV, ionizing radiation, and extreme dryness induced by high vacuum. To investigate the capability for DNA repair via the NHEJ pathway during spore germination, spores of B. subtilis 168 wild type (53) and of repair-deficient mutants with mutations in the homologous recombination gene recA (6), the SP lyase gene splB (11, 14), the NHEJ ligase-like gene ykoU, and the NHEJ Ku-like gene ykoV and the double ykoU ykoV mutant (57) were used (Table 1). All the strains are isogenic with the 168 wild-type strain. Spores of each strain were obtained by cultivation under vigorous aeration in liquid Schaeffer sporulation medium (43), purified, and stored as described previously (2, 25, 36). The mutations did not significantly affect sporulation efficiency. Spore preparations were free (>98%) of growing cells, germinating spores, and cell debris, as seen in the phase-contrast microscope. Spore samples consisted of air-dried spore monolayers (2 × 10⁷ spores) immobilized on 7-mm-diameter quartz discs (Heraeus Quarzglas GmbH & Co. Kg, Hanau, Germany), as described previously (40). For studying the effects of UV and ionizing-radiation-induced DNA damage on spore survival, samples were exposed to monochromatic UV-C radiation from a low-pressure mercury lamp (NN 8/15; Heraeus, Berlin, Germany) with a major emission line at 254 nm, to defined spectral ranges of UV-A plus UV-B (UV-A+B) (290 to 400 nm) or UV-A (320 to 400 nm) radiation obtained with a 1,000-W xenon arc lamp (Polytech, Waldbronn, Karlsruhe, Germany) and optical filter combinations (Schott AG, Mainz, Germany) (24), or to ionizing radiation provided in the form of X rays (150 keV/19 mA) generated by an X-ray tube (Mueller type MG 150, MCN 165; Phillips, Hamburg, Germany). Dosimetry and dose calculations were performed as described previously (23). For studying the effect of vacuum-induced extreme desiccation, samples were exposed to ultrahigh vacuum produced by an ion-getter pumping system (400 liter/s; Varian SpA, Torino, Italy) reaching a final pressure of 10⁻⁷ Pa (18, 29).

TABLE 1.

B. subtilis strains used in this study

Strain	Genotype and/or phenotypeb	Source (reference)
168a	trpC2	DSM 402, DSMZ (53)
WN463	trpC2 recA::ermC MLSr	R. E. Yasbin (6)
TKJ6324	trpC2 splB1	N. Munakata (14)
BFS1845	trpC2 ykoU::pMUTIN4 Cmr	A. J. Doherty (57)
BFS1846	trpC2 ykoV::pMUTIN4 Cmr	A. J. Doherty (57)
BFS1846	trpC2 ykoU ykoV::pMUTIN4 Cmr	A. J. Doherty (57)

^aObtained from the German Collection of Microorganisms and Cell Cultures (DSMZ) GmbH, Braunschweig, Germany.

^bMLS^r, resistant to lincomycin (25 μg/ml) and erythromycin (1 μg/ml); Cm^r, resistant to chloramphenicol (5 μg/ml). splB1 carries two point mutations causing changes in the amino acids G168R and G242D in SP lyase (14).

To recover the spores from the quartz discs after treatments, the spore monolayer was covered by a 10% aqueous polyvinyl alcohol solution; after air drying, the spore-polyvinyl alcohol layer was stripped off as described previously (19). Spores were resuspended in 1 ml sterile distilled water, resulting in >95% recovery of the spores. This procedure has no geno- or cytotoxic effect on the viability of the spores (21). Spore survival was determined from serial dilutions in distilled water as CFU after growth overnight on nutrient broth agar (Difco, Detroit, MI) at 37°C. The surviving fraction was determined from the ratio of N/N₀, with N being the number of CFU of the treated sample and N₀ the CFU of the nontreated controls. Plotting the logarithm of N/N₀ as a function of each treatment, survival curves were obtained. Each experiment was repeated at least three times, and the data shown are expressed as averages ± standard deviations. The results were compared statistically using Student's t test (21, 24, 25). Values were analyzed in multigroup pairwise combinations, and differences with P values of ≤0.05 were considered statistically significant.

Inactivation kinetics of wild-type and mutant B. subtilis spores were determined in response to mono- and polychromatic UV radiation spanning wavelengths from 254 to 400 nm (Fig. 1). Nearly exponential UV survival curves were obtained for spores of all strains of B. subtilis tested (Fig. 1), and the best-fit curves were used to calculate F₁₀ values (i.e., fluences reducing survival to 10%) for statistical comparison (Table 2). With UV-C, the recA, ykoU, ykoV, and ykoU ykoV strains produced spores with slightly (1.5- to 2-fold) but significantly lower F₁₀ values than wild-type spores (Table 2). In contrast, spores carrying the splB mutation were threefold more UV-C sensitive than wild-type spores (Table 2). The slight but significant UV sensitivity of ykoU and ykoV mutant spores likely reflects the fact that the majority of the 254-nm UV photoproducts in spore DNA are SP, whereas strand breaks such as SSB and DSB are only minor photoproducts (52). Relative to wild-type spores, in response to all UV wavelengths (UV-C, UV-A+B, and UV-A), splB spores showed the highest rate of inactivation (ranging from three- to ninefold), followed by the ykoU ykoV double mutant, ykoV and ykoU single mutants, and the recA-deficient strain (Fig. 1; Tables 2 and 3). Interestingly, both DNA strand break-repairing mechanisms, homologous recombination mediated by recA and ykoUV-mediated NHEJ, became progressively more important to spore resistance in response to longer UV wavelengths (Tables 2 and 3). These observations are in good agreement with the previous demonstration that exposure of spores to longer artificial or solar UV-B and UV-A wavelengths produced a higher proportion of SSB and DSB in spore DNA (52). The results suggest that NHEJ provides a major DNA strand break repair strategy for spores irradiated with UV-B and UV-A wavelengths of >290 nm, such as encountered on Earth's surface (32).

FIG. 1.

Survival curves of B. subtilis strains in response to 254-nm UV-C (A), 290- to 400-nm UV-A+B (B) and 320- to 400-nm UV-A (C) radiation. Strains: 168 (wild type) (•), recA mutant (○), ykoU mutant (▴), ykoV mutant (▵), ykoU ykoV mutant (▪), and splB mutant (▿). Data are averages and standard deviations (n = 3).

TABLE 2.

Survival characteristics of treated B. subtilis spores^a

Strain genotype	D10 (Gy) of X raysb	F10c			% S after vacuum desiccationd
		UV-C (J/m2)	UV-A+B (kJ/m2)	UV-A (kJ/m2)
Wild type	838.1 ± 98.0	273.1 ± 52.2	11.4 ± 1.4	364.5 ± 12.3	33.4 ± 10.1
recA	378.9 ± 57.7*	189.2 ± 29.5*	8.3 ± 2.1*	136.3 ± 23.0*	(43.1 ± 17.4) × 10−2*
splB	694.4 ± 86.9	91.7 ± 26.1*	2.2 ± 0.6*	43.9 ± 8.3*	19.8 ± 7.6
ykoU	280.3 ± 46.1*	183.2 ± 30.3*	6.9 ± 1.2*	110.2 ± 19.7*	(7.6 ± 2.3) × 10−2*
ykoV	188.4 ± 40.1*	164.7 ± 25.0*	5.7 ± 0.9*	86.7 ± 15.5*	(0.4 ± 0.1) × 10−2*
ykoU ykoV	146.4 ± 35.4*	138.8 ± 17.6*	4.1 ± 0.6*	63.2 ± 9.5*	(0.2 ± 0.1) × 10−2*

^aData are averages and standard deviations (n = 3). Asterisks indicate survival values that were significantly different (P ≤ 0.05) from the survival value for wild-type spores of Bacillus subtilis 168.

^bDose reducing the survival of the spore population to 10%.

^cFluence of UV-C (254 nm), UV-A+B (290 to 400 nm), and UV-A (320 to 400 nm) irradiation reducing the survival of the spore population to 10%.

^d% S, survival after 30 days of exposure to high vacuum (10⁻⁷ Pa).

TABLE 3.

Repair capacity of spores of B. subtilis 168 and mutants after irradiation with mono- and polychromatic UV, X-ray irradiation, and high-vacuum-induced extreme desiccation^a,b

Strain genotype	F10REP/F10WT			D10REP/D10WT (X rays)	SREP/SWT (desiccation)
	UV-C	UV-A+B	UV-A
Wild type	1	1	1	1	1
recA	0.7 ± 0.1*	0.7 ± 0.2*	0.4 ± 0.1*	0.5 ± 0.1*	(1.3 ± 0.3) × 10−2*
splB	0.3 ± 0.1*	0.2 ± 0.1*	0.2 ± 0.1*	0.8 ± 0.2	0.7 ± 0.2
ykoU	0.7 ± 0.2*	0.6 ± 0.1*	0.3 ± 0.1*	0.3 ± 0.1*	(22.9 ± 6.3) × 10−4*
ykoV	0.6 ± 0.1*	0.5 ± 0.1*	0.2 ± 0.1*	0.2 ± 0.1*	(1.2 ± 0.3) × 10−4*
ykoU ykoV	0.5 ± 0.1*	0.4 ± 0.1*	0.2 ± 0.1*	0.2 ± 0.1*	(0.6 ± 0.2) × 10−4*

^aREP, repair-deficient spores, WT, wild type.

^bS, spore survival after 30 days of vacuum-induced desiccation. Asterisks indicate values that are significantly different (P ≤ 0.05) from the wild-type value, as determined by multigroup pairwise combinations (Student's t test). Data are averages and standard deviations (n = 3).

Ionizing radiation induces less specific damage to DNA than does UV radiation. There are two alternative routes of radiation damage to biological components such as proteins, RNA, and DNA: direct energy absorption (direct radiation effect) and interactions with radicals, e.g., those produced by radiolysis of cellular water molecules (indirect radiation effect) (22). Both processes mainly cause SSB or DSB in DNA (9-11, 22, 30). The decreased water content of the spore core, which may reduce the indirect effect, has been suggested to be responsible for increased spore resistance to ionizing radiation (32, 49). However, to date no DNA repair system has been identified that acts specifically on spore DNA damage caused by ionizing radiation. Therefore, wild-type and mutant spores were assayed for their resistance to ionizing radiation delivered in the form of X rays (Fig. 2). After exposure to X rays, strictly exponential survival curves were obtained for spores of all repair-deficient strains of B. subtilis tested, while the wild-type strain showed a slight shoulder in its survival curve (Fig. 2). More dramatic differences were observed in the spore survival kinetics of the various strains exposed to X rays (Fig. 2) than the same strains exposed to UV-C (Fig. 1A). Indeed, the X-ray data closely mimicked the response of the recA, ykoU, ykoV, and ykoU ykoV strains to UV-A (Fig. 1C; Tables 2 and 3), further supporting the notion that both UV-A and X rays produce substantial SSB and DSB in spore DNA. This contention is further strengthened by the observation that spores of the splB strain, which is defective in repair of the UV-induced photoproduct SP, were not significantly more sensitive to X rays than wild-type spores (Tables 2 and 3).

FIG. 2.

Survival curves of B. subtilis strains in response to X-ray irradiation. Strains: 168 (wild type) (•), recA mutant (○), ykoU mutant (▴), ykoV mutant (▵), ykoU ykoV mutant (▪), and splB mutant (▿). Data are averages and standard deviations (n = 3).

In the case of X-ray resistance, the spores of recA, ykoU, ykoV, and ykoU ykoV strains were remarkably more sensitive than wild-type spores, up to a factor of 7.9 in the case of ykoU ykoV spores (Tables 2 and 3). The observed differences in X-ray resistance of the mutant versus wild-type spores, determined as the repair capacity, were much more pronounced for X rays than for UV radiation (Table 3). These differences in sensitivity of the spores to X rays showed a tendency similar to that observed for vegetative cells of the B. subtilis mutants (57). However, Weller et al. (57) reported that in stationary-phase B. subtilis cells, ykoU mutants were more X ray sensitive than ykoV mutants; interestingly, in spores we observed the opposite (Fig. 2; Tables 2 and 3).

Spores are clearly much more resistant than their vegetative counterparts to extended desiccation both at atmospheric pressure and in vacuo (9-11, 29, 32, 39). When typical laboratory vacuum systems are used for simulated extreme desiccation, wild-type spores often exhibit no detectable killing (9, 10). A major reason for spore resistance to these processes is protection of spore DNA by SASP. SASP-deficient spores are much more sensitive to extended desiccation, and killing by this treatment is accompanied by mutagenesis and DNA damage e.g., single-strand breaks and DNA-protein cross-links (11, 49). Previous studies of spores exposed to extreme desiccation via ultrahigh vacuum indicated that wild-type spores were quite resistant while DNA repair-deficient spores showed significant sensitivity to extended dryness via vacuum (reviewed in reference 32). Although significant killing of wild-type spores was not observed, the survivors exhibited significant mutagenesis (29), indicating that DNA was targeted. In the present study, spores were exposed to ultrahigh vacuum (10⁻⁷ Pa) for up to 30 days and assayed for survival (Fig. 3). Strict exponential inactivation kinetics were observed for all strains (Fig. 3). After 30 days of vacuum-induced extreme desiccation, wild-type and splB spores both showed high levels of survival (33.4% and 19.8%, respectively), which were not statistically different (Tables 2 and 3). In sharp contrast, spores carrying the recA, ykoU, ykoV, or ykoU ykoV mutations were dramatically more sensitive to high vacuum (Fig. 3), ranging from 77-fold (recA spores) to >16,000-fold (ykoU ykoV spores) as measured after 30 days (Tables 2 and 3). The extreme sensitivity of ykoU ykoV-deficient spores to high vacuum-induced desiccation is entirely consistent with the observation that exposure to high vacuum induces SSB and DSB formation in DNA (9, 10, 20, 32). Thus, NHEJ is an extremely important determinant of spore survival of the ultrahigh vacuum prevailing in space, a critical prerequisite for interplanetary transport of spores by natural processes or human spaceflight activities (15, 18, 20, 32).

FIG. 3.

Survival curves of B. subtilis strains in response to ultrahigh vacuum (10⁻⁷ Pa). Strains: 168 (wild type) (•), recA mutant (○), ykoU mutant (▴), ykoV mutant (▵), ykoU ykoV mutant (▪), and splB mutant (▿). Data are averages and standard deviations (n = 3).

Recent data indicated that ykoU and ykoV are functional in the ionizing radiation stress response of B. subtilis, Mycobacterium tuberculosis, and Mycobacterium smegmatis (1, 7, 38, 57, 58). Current models of bacterial NHEJ propose that multiple subunits of YkoV bind to DSB ends and recruit YkoU ligase to join the ends in an ATP-dependent manner (3, 17, 38, 58). It is worthwhile to note that dormant spores do not contain detectable amounts of endogenous ATP; rather, high-energy phosphate is stored in the spore as a large depot of 3-phosphoglyceric acid (3-PGA). In the first minutes of spore germination, 3-PGA is rapidly converted to ATP by the lower branch of glycolysis (45, 51). Thus, the activity of the NHEJ pathway is dependent upon reactivation of ATP generation during spore germination.

The above results raise a question about the potential role of ykoW in the ykoWVU operon in NHEJ. According to the SubtiList database (http://genolist.pasteur.fr/SubtiList) (26), the putative ykoW product is similar to sensory box proteins and/or diguanylate cyclase/phosphodiesterases potentially involved in DNA repair. In this regard, it is interesting that Mun et al. (27) reported that nonhomologous ends were removed by a DNA phosphodiesterase to eliminate radiolytic products (e.g., thymine glycol adducts) at broken DNA ends in Deinococcus radiodurans. In addition, YkoV ligases themselves from a number of bacteria are reported to encode both DNA polymerase and DNA end-processing activities (reviewed in reference 3). These observations point to the intriguing possibility that the ykoW product may also function to prepare DSB ends for resection, a possibility that is currently being addressed.

In conclusion, when faced with prolonged exposure to harsh environments, dormant bacterial spores must repair accumulated DSB to ensure their survival and genome integrity. Our results lend support to the hypothesis that NHEJ is a key strategy used during spore germination to repair DSB caused by terrestrial UV-B and UV-A radiation, as well as UV, ionizing radiation, and ultrahigh vacuum-induced extreme desiccation encountered during exposure to space.

(These results will be included in the Ph.D. thesis of Ralf Moeller and in the research reports of the HIMAC project [17B463].)