Transcriptional coupling (Mfd) and DNA damage scanning (DisA) coordinate excision repair events for efficient Bacillus subtilis spore outgrowth

Abstract The absence of base excision repair (BER) proteins involved in processing ROS‐promoted genetic insults activates a DNA damage scanning (DisA)‐dependent checkpoint event in outgrowing Bacillus subtilis spores. Here, we report that genetic disabling of transcription‐coupled repair (TCR) or nucleotide excision repair (NER) pathways severely affected outgrowth of ΔdisA spores, and much more so than the effects of these mutations on log phase growth. This defect delayed the first division of spore′s nucleoid suggesting that unrepaired lesions affected transcription and/or replication during outgrowth. Accordingly, return to life of spores deficient in DisA/Mfd or DisA/UvrA was severely affected by a ROS‐inducer or a replication blocking agent, hydrogen peroxide and 4‐nitroquinoline‐oxide, respectively. Mutation frequencies to rifampin resistance (Rifr) revealed that DisA allowed faithful NER‐dependent DNA repair but activated error‐prone repair in TCR‐deficient outgrowing spores. Sequencing analysis of rpoB from spontaneous Rifr colonies revealed that mutations resulting from base deamination predominated in outgrowing wild‐type spores. Interestingly, a wide range of base substitutions promoted by oxidized DNA bases were detected in ΔdisA and Δmfd outgrown spores. Overall, our results suggest that Mfd and DisA coordinate excision repair events in spore outgrowth to eliminate DNA lesions that interfere with replication and transcription during this developmental period.

release of dipicolinic acid and monovalent and divalent cations from the spore core, hydrolysis of the spore cortex peptidoglycan, and uptake of water into the spore core to levels comparable to those in growing cells. The full hydration of the spore core completes germination and allows resumption of normal enzyme activity in the spore core including the replication and transcription machineries necessary for initiation of spore outgrowth (Setlow, 2003). During outgrowth, the α/β type SASPs are degraded, freeing spore DNA for transcription, and eventually for replication (Setlow, 1988(Setlow, , 2007. Results from a transcriptomic study revealed that genes encoding repair proteins are expressed during early (5-25 min) and late (40-50 min) outgrowth stages (Keijser et al., 2007). Furthermore, the free amino acids produced from proteolysis of α/β-SASPs support much of the metabolism early in spore outgrowth (Setlow, 2007).
The ability of spores to germinate and propagate depends on their genomic integrity (Setlow & Setlow, 1996). DNA repair cannot take place in metabolically dormant spores, therefore, DNA lesions generated by chemical or physical factors accumulated during the variable periods of spore dormancy, need to be repaired during spores' return to vegetative growth (Nicholson, Munakata, Horneck, Melosh, & Setlow, 2000;Setlow, 2007). It has been proposed that DNA lesions accumulated in dormant spores must be eliminated soon after completion of germination by DNA repair proteins produced and stored in the spore during its development (Pedraza-Reyes, Ramírez-Ramírez, Vidales-Rodríguez, & Robleto, 2012).
As noted above, the return to vegetative growth of B. subtilis spores represents a stage with increased oxidative stress due to the full hydration of the spore core during germination and activation of metabolism during outgrowth (Campos et al., 2014;Ibarra et al., 2008). The oxidative damage inflicted on DNA by reactive oxygen species (ROS) can be counteracted by KatX and the apurinic, apyrimidinic (AP) endonucleases Nfo, and ExoA (Bagyan, Casillas-Martinez, & Setlow, 1998;Ibarra et al., 2008). In the absence of Nfo and ExoA, DisA delays chromosomal replication and spore outgrowth until the genome is free of damage. Of note, DisA is not packaged into the forespore compartment during sporulation, but is synthesized during outgrowth (Campos et al., 2014).
During outgrowth, full reconstitution of many metabolic pathways, as well as nutrient uptake and cell replication requires macromolecular synthesis, which can be launched upon production of ATP (Keijser et al., 2007). In outgrowing spores, protein synthesis is dependent on de novo transcription; therefore, expression of a number of genes is required during the first minutes of outgrowth (Setlow & Primus, 1975;Setlow, 1975). In growing bacteria, genetic lesions occurring in transcriptionally active genes are preferentially repaired, in particular those occurring in the template strand, in a process termed transcriptioncoupled repair (TCR) (Hanawalt & Spivak, 2008;Selby, Witkin, & Sancar, 1991). In this process, the mutation frequency decline protein (Mfd) responds to RNA polymerase stalled by bulky or noncoding lesions and recruits the nucleotide excision repair system (NER) to the lesions through Mfd's interaction with the UvrA protein (Hanawalt & Spivak, 2008;Selby & Sancar, 1993). In B. subtilis the role of Mfd in vegetative growth and stationary phase has been well characterized (Ayora, Rojo, Ogasawara, Nakai, & Alonso, 1996;Pybus et al., 2010;Ross et al., 2006;Zalieckas, Wray, Ferson, & Fisher, 1998). Recent experimental evidence has further revealed that mfd is expressed in the forespore compartment of the sporulating cell and that TCR is required to contend with the noxious effects of bulky DNA lesions during spore morphogenesis (Ramírez-Guadiana et al., 2013). It was also proposed that Mfd stored in spores could play a role in processing DNA damage during spore outgrowth (Ramírez-Guadiana et al., 2013).
The efficient return of spores to vegetative growth requires a chromosome free of damage for appropriate transcription and replication (Keijser et al., 2007;Setlow & Setlow, 1996). Therefore, spore outgrowth offers the opportunity to study how Mfd and DisA modulate excision repair events to ensure efficient spore outgrowth.
Experimental evidence presented in this study shows that Mfd, DisA, and the NER protein UvrA coordinate excision repair events to deal with genetic lesions that interfere with transcriptional and replication events in outgrowing B. subtilis spores.

| Strain construction and culture conditions
The B. subtilis strains used in this study were derived from strain PS832, a prototrophic derivative of strain 168 and are listed in Table   S1. The strains were constructed using standard molecular biology techniques (Sambrook & Russell, 2001). A gene construct to disrupt disA was generated as follows. A 307-bp DNA fragment extending from nucleotides (nt) 275-582 from the disA open reading frame (ORF) was released from plasmid pPERM732 (Campos et al., 2014) by digestion with the enzymes EcoRI and BamHI and cloned into the pMutin4cat vector . The resulting plasmid pPERM1372 was used to transform competent cells of strain PS832 giving strain ∆disA (PERM1504). Competent cells of B. subtilis PERM1504 were transformed with chromosomal DNA of strains ∆mfd yqjH (PERM939) and ∆mfd yqjW (PERM940) (Ramírez-Guadiana et al., 2013) to generate strains ∆disA mfd yqjH (PERM1510) and ∆disA mfd yqjW (PERM1511), respectively. The appropriate recombination events into the homologous loci were confirmed by PCR using specific oligonucleotide primers (data not shown).
Spores of all strains were prepared at 37°C on Difco sporulation medium (DSM) (Schaeffer, Millet, & Aubert, 1965) agar plates without antibiotics, harvested and purified by water washing and stored as described previously (Nicholson & Setlow, 1990). All dormant spore preparations used in this work were free (≥98%) of growing cells, germinated spores, and cells debris, as determined by phase-contrast microscopy. Spores were generally used at an optical density at 600 nm (OD 600 ) of 0.5 corresponding to 0.75 × 10 8 viable spores/ml.

| Spore germination and outgrowth
Spore germination and outgrowth were performed in 2 × Schaeffer′s glucose (2 × SG) liquid medium (Schaeffer et al., 1965) supplemented with 10 mmol L −1 L-alanine. Spores in water were first heat shocked for 30 min at 70°C, cooled on ice, and inoculated into germination medium at 37°C to obtain an initial OD 600 of ~0.5. Where indicated, 0.5 mmol L −1 hydrogen peroxide (H 2 O 2 ) (Sigma-Aldrich, St. Louis MO) or 2 μmol L −1 4-NQO (4-Nitroquinoline-1-Oxide) (Sigma-Aldrich, St. Louis MO), equivalent to a 30% lethal dose of each drug, were added to cultures after most spore germination had taken place; that is, ~15 min after the mixing of spores with germinants. The OD 600 of cultures were monitored with an Ultrospec 2000 spectrophotometer (Pharmacia, Manassas Park, VA), and the values were plotted as a fraction of the initial OD 600 (OD 600 at time t/initial OD 600 ) versus time.
The rates of germination of disA mfd, disA uvrA, and wild-type spores were determined in 25 mmol L −1 Tris-HCl (pH 7.4) plus L-alanine of spore cultures. To this end, the fall in the relative OD 600 values was monitored over a period of 30 min and the linear portion employed to calculate the slope. The rate of germination of wild-type spores was refereed as 100%.

| Determination of chromosomal DNA content
To quantify genomic DNA from spores germinated and outgrown, chromosomal DNA was isolated as follows. Aliquots (3 ml; 1.5 × 10 8 viable spores/ml) of WT, ∆disA mfd and ∆disA uvrA dormant spores that had germinated for 30, 60, 90, and 120 min in 2 × SG were collected by centrifugation (13,500g for 2 min). The pellet of cells was washed two times with 1 ml of lysis buffer (50 mmol L −1 EDTA, 100 mmol L −1 NaCl [pH 7.5]), suspended in 0.3 ml of the same buffer and subsequently processed to isolate the RNA-free chromosomal DNA from the fraction that was directly susceptible to lysozyme degradation as previously described (Campos et al., 2014). The fraction of lysozyme-resistant cells was pelleted by centrifugation. This pellet, which consisted of lysozyme-resistant spores likely containing intact spore coats, was subjected to spore coat removal (Nicholson & Setlow, 1990), washed five times with STE buffer (150 mmol L −1 NaCl, 10 mmol L −1 Tris-HCl [pH 8], 10 mmol L −1 EDTA), and processed for chromosomal DNA isolation (Campos et al., 2014). After RNAse treatment, the chromosomal DNA isolated from both fractions was quantified by UV spectrophotometry (Sambrook & Russell, 2001). The DNA values from both fractions were combined to obtain the total DNA content.

| Microscopy analysis
Cell morphology and chromosome structure during spore germination/outgrowth were analyzed by fluorescence microscopy. Cell samples (1 ml) collected at different times during spore germination/ outgrowth, were centrifuged (11,500g, 2 min) and mixed with 0.

| Analysis of spontaneous and induced mutation frequencies
Determination of spontaneous and H 2 O 2 or 4-NQO induced mutation frequencies to rifampicin resistance (Rif r ) was performed as follows.
Spores were adjusted to a final OD 600 of 0.5 in 2× SG medium, supplemented with 10 mmol L −1 of L-alanine and treated (induced) or not (spontaneous) with the DNA-damaging agents, H 2 O 2 (0.5 mmol L −1 ) or 4-NQO (2 μmol L −1 ), which were added to cultures 15 min after the initiation of germination. 10 ml of cell samples collected 180 min after initiation of germination, were washed with 10 ml of phosphatebuffered saline (PBS; 0.7% Na 2 HPO 4 , 0.3% KH 2 PO 4 , 0.4% NaCl [pH 7.5]) and resuspended in 1 ml of the same buffer. Aliquots of cells were plated on six LB medium plates containing 10 μg ml −1 of rifampicin, and Rif r colonies were counted after 2 days of incubation at 37°C. The number of cells used to calculate the frequency of mutation to Rif r was determined by plating aliquots of appropriate dilutions on LB medium plates without rifampicin and incubating the plates for 24 h at 37°C.

| Identification of spontaneous rpoB mutations conferring rifampin resistance
Rif r colonies spontaneously generated from outgrown spores of the strains of interest were randomly chosen, resuspended in 100 μl of nuclease free-water and subject to cell lysis by heating the cell suspension at 95°C for 5 min (Nicholson & Maughan, 2002). The cell lysates were employed to PCR amplify a 716-bp fragment from rpoB with Vent DNA polymerase (New England Bio-Labs, Ipswich, MA) and the oligonucleotide primer set, RpoBFW 5′-CGTCCTGTTATTGCGTCC-3 (forward) and RpoBRV 5′-GGCTTCTACGCGTTCAACG-3′ (reverse).
The amplified 716-bp rpoB product contained the three hot-spot clusters (nt +1353 to +2069 relative to the ORF of the rpoB gene) where mutations confer Rif r in many bacteria including B. subtilis (Campbell et al., 2001). The amplified rpoB products were subjected to DNA sequencing to identify specific mutations conferring rifampicin resistance. The sequencing was performed in both directions of the rpoB PCR product of 20 clones from wild-type, ∆disA strain, ∆mfd strain, and ∆disA mfd strains. Sequencing was carried out by Functional  (Table S1). Wild-type, ∆disA, ∆uvrA, ∆disA/uvrA, ∆mfd, and ∆disA/mfd strains carrying the empty vector pDR111 at the amyE locus were also generated (Table S1) to quantify the basal fluorescence emitted by cells with these genetic backgrounds. Spores of the different strains carrying the recA-gfpmut3a fusion or the pDR111 empty vector were obtained and purified as described above. Heat shocked spores were inoculated into 2 × SG medium supplemented with 10 mmol L −1 L-alanine to an OD 600 nm = 0.5 and the cultures were shaken at 37°C/250 rpm. After 15 min, each culture was splitted in two equal subcultures; one subculture was made 250 ng/ml in Mitomycin-C (M-C) and the other was left untreated. After 60 min of incubation at 37°C with shaking, samples of 3 ml were collected from both subcultures, pelleted by centrifugation (10,000g for 2 min),  (Table S1), and with or without M-C, were subtracted from the values obtained with the strains harboring the recA-gfpmut3a fusion. The basal values of fluorescence were never superior to 10% (for the noninduced) or 1.5% (for the induced) condition in reference to the strains carrying recA-gfpmut3a.

| Statistical analyses
Differences in mutagenesis between untreated and treated with the DNA damage agents H 2 O 2 or 4-NQO as well as differences in fluorescence between untreated and treated with M-C strains were calculated by Mann-Whitney U test, and analyses were done using Minitab 17 software. p < .05 were considered significant.

| The lack of Mfd or UvrA delays outgrowth of spores lacking DisA
Oxidative DNA damage is a challenge faced by spores during the return to vegetative growth, as ROS-promoted lesions, including oxidized bases, AP sites, and single-strand breaks can be impediments to the transcription and replication machinery during spore outgrowth (Campos et al., 2014;Ibarra et al., 2008). During sporulation, significant amounts of the DNA repair proteins Mfd and UvrA are expressed and packaged in the developing forespore (Ramírez-Guadiana et al., 2013); in contrast, disA is not packaged into the forespore compartment but is synthesized very early in spore outgrowth (Campos et al., 2014). Of note, spores lacking Mfd, UvrA, or DisA alone exhibited germination/outgrowth curves that were indistinguishable from that of wild-type spores suggesting these proteins have either no or redundant functions in this developmental stage (Figure 1). In support of a redundant function for these proteins, loss of Mfd or UvrA in a strain also lacking DisA generated spores that were delayed significantly in their return to vegetative growth in comparison with spores of the wild-type strain. Some of this latter delay was due to a slightly slower germination of the mfd disA and uvrA disA spores compared to that of wild-type spores ( Figure 1). Thus, when germination of mfd disA, uvrA disA, and wild-type spores was monitored by the fall in the OD 600 in 25 mmol L −1 Tris-HCl (pH 7.4) plus L-alanine of spore cultures (Campos et al., 2014), the rates of germinations of the double mutants was ~90% to that of the wild-type spores. This small difference was seen with at least two different preparations of these spores. Experimental evidence has revealed that RecA, UvrA, and Mfd-dependent DNA repair is a relevant process for efficient spore morphogenesis (Ramírez- Guadiana et al., 2013Guadiana et al., , 2016. Therefore, some genetic defect resulting from the loss of these repair function seems to generate dormant spores slightly affected in spore germination but bearing much more significant deficiencies in spore outgrowth. Importantly, the disA uvrA and disA mfd strains exhibited essentially similar doubling times as the wild-type strain in vegetative growth-that is, 35 ± 2.5, 36 ± 1.2, and 34 ± 2, respectively. Therefore, the strong slow outgrowth exhibited by spores of these mutant strains cannot be attributed to vegetative growth defects.

| The delay in outgrowth of ∆disA/uvrA and ∆disA/mfd spores is accompanied by retardation of chromosome replication
The outgrowth defect exhibited by DisA/Mfd-and DisA/UvrAdeficient spores was further examined by epifluorescence microscopy. To this end, spores of these strains as well as spores of the wild-type strain and those bearing single mutations in disA, mfd, or uvrA were induced to germinate in a medium that supported outgrowth and vegetative cell growth, and samples were collected at different stages during germination/outgrowth, and DNA and membrane were stained with DAPI and FM4-64 dyes, respectively  (Table 1). In summary, the microscopic evidence together with results presented in Figure 1, strongly suggest that DisA together with Mfd (TCR) or with UvrA (NER) play a crucial role in repairing spontaneous DNA lesions that interfere with replication and thus delay spore outgrowth. To better support this notion, 1 × 10 8 dormant spores of the WT or ∆disA uvrA and ∆disA mfd strains were induced to germinate and the DNA content from the same amount of cells was determined at 30, 60, 90, and 120 min. The results showed that the DNA content 90 and 120 min after mixing spores with germinants, was significantly lower in outgrowing spores of disA/uvrA and disA/mfd strains than in outgrowing wild-type spores (Figure 3). Wild-type (90) ∆disA (90) ∆uvrA (85) ∆mfd (85) ∆disA uvrA (60) ∆disA mfd (18) Spores of wild-type, ∆disA, ∆uvrA, ∆mfd, ∆disA uvrA, and ∆disA mfd strains were germinated and outgrown in 2 × SG medium at 37°C. 90 min after the germination onset, at least 200 outgrown spores of each strain that were stained with DAPI were analyzed by fluorescence microscopy in at least six different fields to determine the number of cells showing replication and segregation of its chromosome. in spores of the same strains undergoing germination/outgrowth (Figs. 6 and S1).
As described above, metabolic conditions prevailing in outgrowing spores promote the synthesis of ROS, which attack DNA and induce the formation of different types of mutagenic lesions (Wang, Kreutzer, & Essigmann, 1998). In support of this notion, the ROS promoting agent, H 2 O 2 also increased the mutation frequency of outgrowing wild-type spores and of outgrowing ∆disA, ∆mfd, ∆uvrA, ∆disA/mfd, and ∆disA/uvrA spores (Figure 6b).
We next investigated whether DNA lesions that interfere with DNA replication and potentially affect transcription also contribute to mutagenesis during spore outgrowth. Our results supported this contention, since addition of 4-NQO during spore germination in-

| Base substitutions derived from DNA oxidation and deamination promote mutagenesis during spore outgrowth
To further investigate the types of spontaneous mutations occurring in germinated spores after experiencing outgrowth and cell division, a number of Rif r colonies of the wild-type and various mutant strains were randomly chosen. A 716-bp fragment from the rpoB ORF encompassing the three hot-spot clusters giving rise to rifampicin resistance in various bacteria (Campbell et al., 2001) was PCR-amplified from each colony; DNA sequencing confirmed that the Rif r phenotype in all but four of the colonies analyzed was due to base substitutions that occurred in the amplified region of the corresponding rpoB gene (Table 2). Interestingly, the major proportion of amino acid changes associated with the Rif r phenotype occurred in the cluster I of the rpoB open reading frame (Figure 7) (Campbell et al., 2001). Furthermore, the Rif r mutations identified had a large proportion of A→G and C→T transitions in the outgrowing spores of both the wild-type and mutant strains ( Table 2); these base substitutions are typical mutations produced by deamination of adenine to hypoxanthine or cytosine to uracil (Friedberg, Walker, Siede, & Wood, 2006). The disruption of disA promoted the appearance of A↔T transversions, G→A transitions as well as C→G and T→C substitutions, which have been reported to be elicited by oxidative stress (Wang et al., 1998). Interestingly, the A→T, T→G and C→A transversions that were absent in rpoB of outgrown wild-type spores were present in Rif r colonies from ΔdisA and Δmfd spores (Table 2). However, the base substitutions G→T and A→C that were detected in ∆mfd spores were absent in the wild-type spores. Remarkably, in Rif r colonies derived from outgrowing DisA/ Mfd-deficient spores G→A and G↔T substitutions predominated as well as the C→A transversion detected in outgrown spores of ΔdisA and Δmfd strains (Table 2). Notably, from 20 Rif r colonies analyzed in the ∆disA mfd strain, only 16 exhibited a base substitution in the sequenced rpoB region (Table 2). Thus, base substitutions in a different rpoB region may have generated the additional 4 colonies with a Rif r phenotype; indeed, mutations occurring in the N-terminal region of RpoB (from amino acids 132-136) have been reported to produce Rif r -resistant bacteria (Campbell et al., 2001).
Together, our results suggest that in germinating/outgrowing spores of B. subtilis: (1) spontaneous events of base deamination and oxidation contribute to transcriptional and replicative interference; and (2) Mfd and DisA operate on these types of lesions coordinating faithful and error-prone events of DNA repair.

| The SOS response is spontaneously activated during spore germination/outgrowth
We investigated if the lack of Mfd and UvrA in the ΔdisA background induces the SOS response during spore germination/outgrowth employing a recA-gfpmut3a fusion that was recombined into the amyE locus of the WT and different mutant spores. As shown in Figure

| DISCUSSION
In this work, an interaction of DisA with Mfd and UvrA (NER) in processing genetic lesions that are potential blocks for transcription and replication during spore outgrowth was uncovered. Thus, wild-type and disA-deficient spores treated or not with H 2 O 2 exhibited germination/outgrowth curves and kinetics of chromosomal replication that were essentially similar (Figures 1,2), suggesting that repair mechanism(s) operating in this developmental stage could suppress the checkpoint function of DisA. Spontaneous DNA lesions reported to stall RNA polymerase and potentially generate transcriptionreplication conflicts in different bacteria (Saxowsky & Doetsch, 2006;Wang et al., 1998) were also detected in outgrowing wild-type spores (  (Campos et al., 2014). In connection with these concepts, after the loss of dormancy, spores enter into an active stage of transcription preceding the first round of chromosomal replication (Keijser et al., 2007). Therefore, Mfd could be necessary to couple repair of lesions arresting the progression of the RNA polymerase in the transcribed strand of genes necessary for an efficient spore's return to vegetative growth. However, Mfd could be also involved in resolving structural conflicts resulting from encounters of the replication machinery with RNA polymerase stalled at DNA lesions (Merrikh, Zhang, Grossman, & Wang, 2012;Million-Weaver et al., 2015). In support of these notions, the outgrowth of spores lacking Mfd or UvrA (NER) was severely affected by 4-NQO, a DNA-damaging agent that interferes with DNA replication ( Figure 5). Therefore, the TCR and the NER pathways are not only crucial in sporulation (Ramírez-Guadiana et al., 2013) but as demonstrated here, also during spore outgrowth.
Previous studies have reported on the contribution of UvrA in processing AP sites and single-strand breaks during spore germination/ outgrowth (Campos et al., 2014;Ibarra et al., 2008). Results from this work showed that UvrA (NER) could also back up the function of DisA, as the absence of both proteins affected spore outgrowth in the presence or absence of H 2 O 2 . In parallel with this defect, outgrowing ∆disA uvrA spores were delayed in their first chromosomal replication ( Figure 2) and exhibited increased spontaneous Rif r mutagenesis ( Figure 6). Thus, during spore outgrowth, DisA and UvrA act coordinately to faithfully remove spontaneous genetic lesions that are potential blocks for replication. Of note, the role of UvrA in this developmental stage could be attributed in part to its contribution to TCR.
Furthermore, the C→A transversion, which is commonly generated by unrepaired 8-OxoG (Wang et al., 1998) was also detected in outgrown spores lacking DisA or Mfd. Importantly, G→T and A→C transversions resulting from incorporation of 8-oxoG into DNA as well as G→A mutations promoted by 5-OxoC (Wang et al., 1998), were identified in outgrown spores lacking Mfd or DisA, respectively. In conjunction, these results support the notion that DisA and Mfd work together to eliminate ROS-promoted nonbulky DNA lesions during spore outgrowth, suggesting a role of Mfd and DisA in coordinating proteins involved in repairing these lesions. Notably, 10 out of the Rif r colonies generated from 20 outgrown WT spores that were allowed to progress to the growth stage consisted of the same C→T transition, changing codon H 482 (CAC) to Y 482 (TAC) (Table 2, Figure 7). Importantly, this mutation was found to confer rifampicin resistance in B. subtilis spores (Nicholson & Maughan, 2002). Furthermore, 3 of the 20 Rif r colonies contained an A→G substitution, which changed codon H 482 (CAC) to R 482 (CGC) and this mutation was found not only in spores but also in vegetative cells of B. subtilis (Nicholson & Maughan, 2002). Therefore, physiological conditions encountered by outgrowing B. subtilis spores not only potentiate mutagenesis but also generate a differential mutational spectra with respect to that exhibited by vegetative cells. Finally, the A↔T, G↔C, T→C and A→C mutations were absent in outgrown ∆disA/mfd spores, strongly suggesting that additional repair function (